Welcome back to the tasty morsels of critical care podcast.

We’re going to cover a bit of an environmental/tox topic today and look at carbon monoxide poisoning from Oh’s manual chapter 83 on burns. I have previously covered this on the old tasty morsels of EM series back when i was doing my EM fellowship exams.

As you no doubt remember from school chemistry classes, carbon monoxide is a colourless, odourless, tasteless gas produced when combustion occurs with insufficient oxygen.

We’re likely to see this in a couple of contexts.

1) the house fire victim, pulled from the fire unconscious and sick

2) the sub acute or chronic poisoning in a patient presenting with headaches and flu symptoms that seem to get better when they leave the problem environment. The classic EM example is the whole family who present with flu symptoms and no fever and even the dog is sick. We’re much less likely to see this cohort in the critical care side of things.

How does it make people sick? Haemoglobin is a fickle little protein, while evolved to carry oxygen to needy tissue beds it actually has a distinct preference not for our beloved oxygen but for carbon monoxide. Introduce some carbon monoxide at the alveolus and the haemoglobin molecule will bind to CO with an affinity 240 times that than for oxygen. I take that number of 240 somewhat at face value but I presume someone got a PhD from working that out. In visual form my preferred means of explanation for this would be the distracted boyfriend meme where the haemoglobin boyfriend looks longingly over his shoulder at the carobon monoxide while his oxygen girlfriend looks on in horror. Hopefully you get the idea.

So instead of having lots of circulating oxyhaemoglobin we’re instead left with lots of not especially useful carboxyhaemoglobin. Let’s imagine 50% of our Hb is now carboxyHb and 50% is OxyHb we’re left with a sort of severe fucntional anaemia where half of our Hb is out of action. One might be inclined to think that this is the major cause of morbidity and mortality in CO poisoning but in fact this is only a small portion of the problem. CoHb actually has a direct cytotoxic effect on things cytochrome oxidase and myoglobin function. As such it interrupts the whole process of oxidative metabolism and life as we know it.

We can measure the level of CO fairly easily, any blood gas machine worth its salt should be able to give you a break down of the types of Hb present in the sample. This is co-oximetry and typically it’ll show you oxy, deoxy, carboxy and met haemoglobins. All these different forms of Hb absorb different wavelengths of light. The lowly pulse oximeter does not have the subtlety to distinguish the different wavelengths as it only functions at wavelengths of 940 and 660nm. Indeed the pulse ox often demonstrates a non diagnostic number somewhere in the 80s rather than a true reflection of the CarboxyHb or OxyHb present.

Severe CO poisoning resulting in obtundation is going to have high level of COHb on our cooximeter. >10% is quoted but it’s more often over 30%. Patients are going to be pretty sick often from multiple pathologies but COHb on its own is enough to produce severe neurological injury, shock and even cardiac injury is also quite prevalent. Expect a high lactate given the disruption of oxidative metabolism. Resuscitate and investigate as you would any sick patient.

Treatment is nice and simple in that we just give loads of oxygen. Oxygen reduces the half life of CO in the blood quite dramatically, commonly quoted numbers are

• the haf-life of COHb in an FiO2 of 0.21 is 300 minutes
• the half-life of COHb in an FiO2 of 1.0 is 60-90 minutes 

There is a substantial rationale and literature on the use of hyperbaric oxygen as a means of accelerated clearance of COHb. But the RCTs that have been done don’t seem (to me at least) to give a clear benefit. The Lindell Weaver NEJM RCT in 2002 did suggest a neuro benefit but only 8% of the patients in this trial were intubated. A follow up trial in 2011 by ICU steroid guru Djilalli Annane did not find a benefit . So if anyone should get this it might be the non intubated isolated COHb poisoining. This is not really our cohort. Our cohort is likely to be tubed, shocked, with multiple injuruies and not someone you want to transport cross county to put in a single person hyperbaric chamber for hours at a time.

Reading

Oh Manual Chapter 83

Weaver, L. K. et al. Hyperbaric oxygen for acute carbon monoxide poisoning. The New England journal of medicine 347, 1057–1067 (2002). Annane, D. et al. Hyperbaric oxygen therapy for acute domestic carbon monoxide poisoning: two randomized controlled trials. Intensive Care Medicine 37, 486–492 (2011).

 

 

 

Welcome back to the tasty morsels of critical care podcast.

We’ve been talking about pulmonary hypertension, last time we had a pretty broad overview with a focus on group 1 or pulmonary arterial hypertension. This time we’re going to go through some management strategies that might keep you between the hedges on a night on call or a fellowship exam viva.

We briefly mentioned the PH specific drugs that someone might be on. The evidence base for these is almost exclusively in group 1 PH. But what should we do with these meds in someone with group 1 PH who has just arrived back from theater after a laparotomy and a hartmans and they’re on a bit of noradrenaline? The simple answer is continue them. The more complicated answer is you should usually continue them. For example there will be the very rare patient whose pulmonary vascular resistance is kept low in the community with a PICC line and an epoprostenol pump. They are critically dependent on this drug with a very short half life and it should be continued at all costs. Think about it like an adrenaline infusion running at 10mcg/min, not something you can tolerate a break in.

A recurring message from the review papers on critically ill patients with PH is to focus on treating PVR not PA pressures. This is a somewhat philosophical approach that reminds us that the PA pressures themselves don’t prognosticate especially well but a failure of flow from right to left will result in cardiogenic shock and death.

We have a lot of vasoactives to choose from in helping with this, most of which have varying impacts on the PVR. Vasopressin has some animal data suggesting it causes less rise in PVR than our beloved noradrenaline but take that with an appropriately loosely defined portion of salt given that animal data is not ICU patients. Milrinone seems like a great idea as an inotrope that is easy on the PVR but the often dramatic drop in SVR is often a disaster. Dobutamine has the benefit of at least having substantial clinical experience in PH patients even if the tachycardia and even worse the a fib is less than desirable.

The ventilator is a bit of a poisoned chalice. Not only do you have to tolerate a significant risk of peri-intubation cardiac arrest even once you get them on the vent you have to deal with the adverse effects of positive pressure on the RV. The only upside of the vent is that it might make them easier to oxygenate but only if the cause of the hypoxia was a big shunt physiology like a pnuemonia. Oxygen is a great tool for reducing PVR so if we can leverage that then that’s great. However, a lot of hypoxia in end stage PH is reduced mixed venous oxygenation due to low cardiac output and the vent does nothing good for this.

Once on the vent we want a goldilocks’s zone of lung unit recruitment. Too little PEEP we have atelectasis and shunt and hypoxia and vasoconstriction. Too much PEEP and we have overdistension which itself can raise PVR by squeezing the pulmonary vasculature. Finding that sweet spot for the PEEP is a whole post or 10 on its own.

While on the vent it’s a good opportunity to deliver some inhaled therapies. The original gangster here is of course nitric oxide which is one of our target molecules in PH. In a crisis and a failing RV, this might get you out of a tricky spot. But given its expense and not being widely available its worth considering other inhaled options, particularly intermittent nebs of iloprost or a continuously nebulised eporprostenol solution both of which i have seen implemented to good effect.

In terms of monitoring should we be reaching for a PAC? Well, take a step back to start with. We probably need the CVP more. The RV is the first downstream organ that suffers under the burden of worsening PH and if the RV is failing then the CVP will be rising. Like any monitoring tool, a PAC in itself is going to do nothing but provide you with scary looking numbers, particularly the PA pressures which, remember, you should largely ignore. But picking up a severely raised wedge for example might push you to be much more aggressive with your diuresis and left heart management. A continuous cardiac output monitor will allow you to titrate your vasoactives with a great deal more confidence and accuracy

The other monitor I would reach for would be echo. I am a self confessed echo phile so take that into consideration but one of my targets of treatment is going to be how the heart looks. Is the IVS becoming less flattened, is the RV less distended, is the TAPSE improving etc… Echo early, echo often in my book.

Atrial fibrillation is something of a right of passage in the ICU. Have you really been critically ill if you haven’t even had an episode of fast AF? When it comes to PH it’s often poorly tolerated and the approach to rhythm and rate control probably needs to be a bit more aggressive than usual. Our usual choice of vitamin A, amiodarone is a good start but you may need other agents like dig or even DCC to get control. A consistent message from the reviews is to avoid beta blockers. The negative inotropic effect on an RV that is already functioning at peak capacity is not going to be good.

Our first reaction when faced with hypotension is often to load with fluid, this makes sense when we think of the frank starling mechanism, we want to be sure our LV is appropriately pre loaded. But in PH the issue is a failure to deliver volume or flow from the right heart to the left. We can dump a litre into the venous side of the circulation but the PVR just stops it getting efficiently through to the LV. If your patient is hypotensive then the RV is already failing in its basic function of delivering volume and flow to the LV while keeping the CVP low. More fluid is almost never going to fix this.

Indeed diuresing the hypotensive patient may well be the way to go. If you can decongest the right side and reduce the bowing of the septum you’ll get both the RV and the LV working more efficiently

This is only a taster of things you might want to try in a critically ill patient with severe PH. It is important to emphasis that they are not evidence based overall. Most of it is interpretation of clinical physiology at the bedside and applying the available manipulations. Which is of course what makes it so much fun.2

Reading

My own rambling review of pulmonary hypertension on JFICMI website.

2022 ESC Guidance

McLaughlin, V. V., Shah, S. J., Souza, R. & Humbert, M. Management of Pulmonary Arterial Hypertension. J. Am. Coll. Cardiol. 65, 1976–1997 (2015). Jentzer, J. C. & Mathier, M. A. Pulmonary Hypertension in the Intensive Care Unit. J. Intensiv. Care Med. 31, 369–385 (2015). Johnson, S. et al. Pulmonary Hypertension: A Contemporary Review. Am. J. Respir. Crit. Care Med. 208, 528–548 (2023). Barnett, C. F., O’Brien, C. & Marco, T. D. Critical care management of the patient with pulmonary hypertension. Eur. Hear. J. Acute Cardiovasc. Care 11, 77–83 (2022).

Welcome back to the tasty morsels of critical care podcast.

Today we tackle a somewhat nebulous syndrome. Something we throw around with a few hand wavy explanations but often light on detail. Hopefully in a few minutes you’ll at least have a few morsels more of information to stave off all the trainees who are undoubtedly much smarter than you on the ward round. But perhaps I’m getting too autobiographical already.

This does not appear with any great frequency in Oh’s manual but there is a nice scientific statement from the AHA that is referenced below. Though when you call it a statement you imagine some nervous spokesman in front of a camera trying to explain why is boss has done something naughty. Instead this is a 39 page epic review of the topic.

To start with there are apparently 5 types of cardiorenal syndrome. I’ll let that sink in. You all thought there was one didn’t you?

Type 1 is the acute deterioration in kidney function seen in cardiogenic shock from ACS. Type 2 is the slow and chronic deterioration of kidney function in the chronically failing heart.

They get sneaky with type 3 calling it renocardiac syndrome. You see what they did there they just reversed cardiorenal syndrome and called it renocardiac syndrome. In this scenario the kidney has acutely been injured and the consequences such as fluid overload cause heart failure. Type 4 is again renocardiac with the kidneys causing the heart failure but on a chronic basis. With me so far?

Type 5 is the big bucket where they put all the left over disease that cause both kidney and heart failure eg things like amyloid, or sepsis or cirrhosis.

Certainly when i use the term in daily practice i was only ever thinking of types 1 and 2 and that’s what we’re going to focus on in this  tasty morsel.

Why does this happen. I’ll paraphrase the opening part of the pathophys section from the scientific statement. Conventionally we focus on poor forward flow from the heart causing poor renal perfusion, poor GFR and activation of the RAAS. But in the style of a telemarketing TV advert “wait there’s more”. Poor forward flow is by no means the only pathology and in fact high pressures on the venous side likely contribute to the phenomenon of cardiorenal syndrome. for example we know that a MAP of 65mmHg is a generic target for perfusion pressure for most organ beds. However the actual calculation of perfusion pressure is probably better represented as MAP-CVP. Therefore in those with CVPs chronically sitting in the 10-15 range, you are going to struggle to effectively perfuse their kidneys. You’ll even here this called congestive renal failure on occasion.

Along the same lines it’s worth thinking about the impact of intrabdominal pressure on renal perfusion, those with tense ascites from heart failure are also going to struggle. There are of course much more complex neurohumoral, inflammatory type cytokiney thingies going on but as you can tell they are well over my head so I’ve skipped them for now.

You might think that diagnosis of cardiorenal syndrome might be straightforward. We just check a creatinine and if it’s high it’s a problem. But there are a fairly bewildering array of tests available for assessing renal function beyond the very blunt stick of creatinine. Things that rejoice in names like NGAL or cystatin C or looking at albuminuria; all may have a role in teasing out CRS from other issues. Valuable as it is for filling the 39 pages of the scientific statement i can’t see any great relevance to the jobbing intensivist. Of note in the paper, and perhaps obscured by the detail of the available biomarkers is the note that fluctuations in creatinine are often poor representations of actual kidney injury. I took home from this discussion that as long as they are still diuresing effectively we shouldn’t be in a rush to hold the diuretics purely because the creatinine bumped.

Of note as part of the diagnostic work up the statement does give a shout out to the much maligned and greatly missed PAC. This might allow us to effectively assess congestion while avoiding the terrors of hypoperfusion from volume removal.

Moving swiftly on to management strategies I think it’s clear that diuretics have a clear role in congested heart failure patients. However there does seem to be a reluctance to give diuretics once the creatinine bumps up anywhere above the normal range. There is a pervasive (and unfounded) belief that loop diuretics are directly nephrotoxic and as such should not be given. But if we’ve been paying attention so far we’ll realise that congestion itself may be causing the kidney injury and decongestion may well fix things. Now of course we need to be a doctor about this and have a think about other causes of AKI beyond simple congestion but for the sake of the podcast we will assume that we have the correct diagnosis.

Let’s say we have done the right thing and given a decent dose of loop diuretic despite the bump in the creatinine, we often encounter something called the “braking phenomenon”. This refers to the idea that we get less and less response to each successive dose of diuretic, and this can develop over hours. The pathophys of this is beyond the scope of this podcast but involves the nephron doing what it does best in a crisis and tries to hold on to more sodium. You can get around this by making a flanking attack on the nephron by bringing in something like a thiazide in addition. Indeed the concept of the Nephron Bomb covered in tasty morsel 68 (first made popular to me by Joel Topf known as kidneyboy on twitter) is a clinically compelling and somewhat entertaining way to approach pharmacology of diuresis.

Of note there comes a certain point where no matter the diuretic stratgey the volume of wee wee produced is insufficient. And this indeed portends a poor prognosis. Ultrafiltration with whatever mode of RRT you choose seems a compelling option but has performed poorly in most trials to date. Either because it simply doesn’t work or possibly because those sick enough to qualify for an ultrafiltration trial have already found themselves in a category of patients likely to do poorly no matter what.

This segues relatively nicely into a section of the document on palliative care. It is important to realise that a referral to ICU for refractory cardiorenal syndrome may simply be a sign that the patient is reaching end of life. Adding an extra machine to a patient at end of life is not good form and it is incumbent upon us to do the work to figure out if we have some degree of reversibility (eg from acute congestion) or if this is just progression of an underlying irreversible disease process

Reading:

– Rangaswami, J. et al. Cardiorenal Syndrome: Classification, Pathophysiology, Diagnosis, and Treatment Strategies: A Scientific Statement From the American Heart Association. Circulation 139, e840–e878 (2019).
– Mullens, W., Verbrugge, F. H., Nijst, P. & Tang, W. H. W. Renal sodium avidity in heart failure: from pathophysiology to treatment strategies. Eur Heart J 38, 1872–1882 (2017).
– Mullens, W. et al. Evaluation of kidney function throughout the heart failure trajectory – a position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail 22, 584–603 (2020).

 

Welcome back to the tasty morsels of critical care podcast.

Oh Chapter 37 is dedicated to NIV in the ICU and is probably worth some time given that this is a common respiratory support both in the ICU and throughout the hospital.

Many of the benefits of NIV are similar to those seen with ventilation with the blue plastic tube through the vocal cords.For example you still get:

• positive airway pressure which recruits alveoli and improves oxygenation
• improved alveolar ventilation which improves minute volume and lowers CO2
• reduction in work of breathing as the machine is doing some of the work
• stabilisation of the chest wall eg in rib fractures
• reduction in transmural LV pressures acting as a sort of poor man’s IABP (more on that later)

The big advantage of course is that you get all the positives but avoid the blue plastic tube through the cords and all the hassle and complications that come with that.

But it’s not all unicorns and rose petals, the mask itself has a tendency to macerate the face over time and patients who are already feeling breathless and suffocating often don’t take kindly to having a plastic mask shoved over their face. Even if they do tolerate the mask it is frequently difficult to make a decent seal and maintain that lovely positive mean airway pressure that you’re looking for.

And while i did wax lyrical about the potential positives of positive pressure ventilation at the beginning of the post, it seems only fair to point out the negatives of positive pressure ventilation. It is clear that positive pressure ventilation is non physiological and is known to cause its own form of lung injury when applied through a plastic tube through the cords. The alveoli only see the pressure and care not which device it’s being delivered through, so there’s no good reason why NIV wouldn’t cause similar problems.

This of course brings up the unanswered and quite entertaining controversy over P-SILI or patient self induced lung injury that hit its zenith during the worst days of the COVID-19 pandemic. There were back and forth letters in the journals between some of the heaviest hitters in the ventilation world bouncing back and forth whether they actually believed self induced lung injury was a thing. Now this is not the post to explore it, but perhaps suffice to say that someone sitting with a resp rate of 30 for a week on 80% O2 and a PEEP of 10 on NIV may well be undergoing some of the same lung stress that any typical ventilated ARDS patient may be undergoing. NIV is not necessarily a free pass.

When it comes to modes, the names are, as ever, confusing and baffling. Overall they split into some kind of CPAP mode where airway pressure is constant throughout the respiratory cycle and a mode with pressure support set above the PEEP where the pressure increases above the baseline CPAP when the patient inspires. To make matters worse there’s no clear consensus in how the numbers are described. For example, our portable, single limb circuit, ward based NIV machines use the terminology EPAP and IPAP to describe the pressures with both numbers starting from zero. for example 10/5 would be a CPAP of 5 with an additional 5cmH2O pressure support whenever the patient expires. On an ICU vent this would be described as 5/5.

When would you reach for NIV over say one of the aforementioned blue plastic tubes through the cords? Well there are a number of now well established indications where it is entirely appropriate to try and temporise with NIV rather than just putting the tube in. I’ll give a brief summary of a few of them below:

Pulmonary oedema.

• the heart is poor, the lungs are wet and heavy and the sats are low. The patient is crying out for some CPAP. How might it help, let me count the ways.
  1. by increasing intrathoracic pressure you are decreasing the gradient of pressure between the low pressure at end disastole in the LV and the high pressure at end diastole in the aorta. As a result the LV has to do less work to pop open the AV and get blood moving forward to the aorta. This mechanism is somewhat akin (though probably not nearly as effective)  to the afterload reduction with an IABP. hence the description of CPAP as the poor man’s balloon pump
  2. improving oxygenation by recruiting alveoli
  3. reducing work of breathing by giving a little boost as the chest wall tries to expand those wet and heavy lungs
  4. applying a little +ve hydrostatic pressure to the alveoli to get the fluid back into the vasculature where the furosemide can do its glorious work of diuresis.
• there is a clearly proven benefit for reducing intubations and improving oxygenation though the signal for mortality improvement is not as clear.

Exacerbations of COPD

• in this scenario the lungs are scarred and the airways constricted and obstructed. A minor sniffles can be enough to push them over the edge of respiratory failure and the CO2 is rising and the pH is falling and they do not have the respiratory reserve to up their work of breathing
• NIV, and in particular a mode with increased support during inspiration can improve the minute volume and clear the CO2 and wake them from their CO2 narcosis.
• This is a very well supported intervention with 14 or so RCTs showing benefit and an NNT to avoid intubation and death of 4 and 10 respectively .

Asthma

• now we have rapidly strayed into the evidence lite zone. It seems somewhat counterintuitive that a disease where the main issue is difficulty breathing out would be helped by adding positive pressure down the airway but it may be that the extrinsic PEEP is helping them overcome the intrinsic PEEP, or it could be that its reducing the work of breathing or it could be any number of  potential arguable benefits. Still to be proven but commonly tried.

ARDS

• For a long time this was firmly in the controversial box and many would have argued that ARDS needs low volume low pressure ventilation, all of which we can not control in NIV. Meaningful trials were lacking. Then COVID came along and we all went mad with the old CPAP handing out CPAP masks with the coffee and dexamethasone on the morning ward round. Allowing that the bilateral infiltrates of COVID meet the definition of ARDS by any standard then it seems that NIV had found a role in the world of ARDS. That role and how far you might push CPAP before biting the bullet and tubing them remains to be defined. Yes I can keep this person going for 2 weeks on 90% FiO2 and a PEEP of 10 but is that really the best thing to do? There’s no sarcasm here, i really don’t know the answer to that…

Post extubation support

• so you’ve taken the tube out and they’re struggling, should you stick an NIV mask on or just put the tube back in? Again, data not exactly clear on that but it seems that if they’re failing extubation due to either pulmonary oedema or bronchospasm/COPD then it’s probably worth a go. hardly surprising seeing as they’re the two most solid indications we have already. If its’ not either of those then probably best just to put the tube back in.

 

Reading:

Oh’s Manual of intensive Care chapter 37

 

Welcome back to the tasty morsels of critical care podcast.

Today we’re covering the ambitious topic of CRRT in the ICU. Something that occupies a central part of the daily job, but also occupies Oh Chapter 48, Irwin and Rippe chapter 201 and a few other review papers thrown in for good measure. We’re only going to get so far as the modes today so let’s not get too carried away.

The obvious initial distinction in RRT modes is between IHD and CRRT with IHD being intermittent as the name suggests and CRRT being continuous. These are obvious temporal discriminators to do with how long the machine is attached to the patient but under the hood there are more fundamental differences between how the two modes work.

In broad terms we can compare dialysis (the movement of small molecules across a membrane along an osmotic gradient) with ultrafiltration (the squeezing of plasma through a big sieve that retains the big bits of the plasma and lets the other bits leak out). The best analogy I’ve seen for this comes from one of my colleagues in his yearly introduction talk to RRT. Dialysis is the tea bag as ultrafiltration is to the espresso machine.

Alas such simply categorisations fall by the way side when we encounter the actual workings of one of the big green machines in the unit as it often presents several modes to us. We can run a continuous heamofiltration, a continuous haemodialysis or a combo mode of continuous haemodialfiltration. These rejoice in the acronyms CVVH, CVVHD and CVVHDF respectively.

Lets start with CVVH, continous venoveno heamofiltration. In this set up blood is drained from the venous side and entered a circuit initially under negative pressure then post pump becomes positive pressure. This positive pressure is used to force blood through a haemofilter containing many hollow fibre microtubules making up around a metre squared squeezed into that tiny plastic cylinder. Hydrostatic pressure drives the water into the filter compartment from the blood compartment with “solute drag” bringing along small and middle molecules with it. The principle here is convection with both the transmembrane pressure and the semipermeable barrier characteristics both contributing to how much filtrate is generated. The filtrate produced looks just like urine and collects in a big bag at the bottom of the machine. The yellow stuff contains things we want out of the body like potassium and urea. It is very easy to remove water from the body in this manner and in general if left to filter without replacement fluid then your patient will become very negative very quickly, hence the large 5L bags of replacement crystalloid fluid that run simultaneously as the yellow stuff is being produced.  At its simplest the yellow ultrafiltrate has all the same concentrations in it as the plasma minus the large molecules like albumin.

In pure CVVHD (continuous venovenous haemodialysis) the patient’s blood is on one side of a membrane with a dialysate fluid running in the opposite direction on the other side of the membrane. In this scenario solutes (such as Na and K and urea) leave the blood compartment to the dialysate compartment down a concentration gradient. In this scenario the water follows the solute which is in distinction to haemofiltration. When running CVVHD the dialysate flow rates are usually very modest at maybe 30ml/min in distinction to IHD dialysate flow rates of 500-800ml/min

CVVHDF is a combination of the two with a little bit from column A and little bit from column B so to speak, with plasma being squeezed through the haemofilter and a modest counter current dialysate flow happening at the same time. The yellow stuff produced in this mode is a combination of ultrafiltrate and the spent dialysate that has passed through the filter.

One would think that we’ve already covered enough acronyms for one day but unfortunately there are several other important ones still to cover, thankfully they use most of the physiological principles already covered.

Let’s start with SCUF, slow continuous ultrafiltration. Simply put this is CVVHaemofiltration without fluid replacement. Blood enters a haemofilter and through the interaction of hydrostatic pressure and membrane characteristics ultrafiltrate is produced and the remaining blood is returned to the body minus some electrolytes and water. Because the electrolyte concentration in the ultrafiltrate is the same concentration as that in the blood there’s no major drops in Na or K to worry about as long as you don’t remove too much fluid. Solute clearance overall is very poor as effluent rates are more like 100-200/hr (~2ml/kg/hr) vs the usual 2000ml/hr (25ml/kg/hr) produced in CRRT. But this is not an issue as this is a mode that you use in someone who has reasonably working kidneys but has about 10 or 15 kg of water to remove. Removing 100-200ml of ultrafiltrate an hour will rapidly dry out your patient as you might imagine.

The final ICU specific mode of RRT is probably SLED. Slow, Low efficiency Dialysis. It must be said that they’re really not selling it with a name like that and perhaps the inventors need to up their branding game a little. Unlike all the other modes so far this needs a standard IHD machine and the plumbing that goes with it. The idea here is to run the blood and dialysate flows at a much lower rate over a longer period of time. Given that it’s a diffusion mode it is much more effective at solute removal than CRRT. The idea is that your ICU patient can have a busy day with trips to CT, line changes, rehab activities and then get plugged in for the night shift to SLED to give the blood a nice wash ready for another day of clinical progress the next day. I have zero experience with this but people who do have it feel it’s the best thing since loaves came pre-cut up and ready for the toaster.

CRRT is clearly ubiquitous in every day ICU practice and one might think this is due to an overwhelming collection of evidence suggesting its superiority over IHD but perhaps unsurprisingly such an evidence base supporting a mortality benefit does not exist and the stability and availability of CRRT has led to its current market leading position. There does seem to be a suggestion of better renal recovery with CRRT over IHD which might be related to less kidney hypoperfusing episodes of hyoptension when CRRT is used.

Reading:

Oh Chapter 40

Iriwin and Rippe 201

Deranged Physiology

 

Welcome back to the tasty morsels of critical care podcast.

Nestled towards the end of Oh Chapter 51 we have a section dedicated to SAH. Given that a lot of ICU bed days are given over to managing SAH, I felt it might have warranted its own chapter. Indeed, looking at its prevalence in fellowship examinations it does seem that a fair deal of attention should be given to SAH. It stands apart from the usual intracranial bleeding where the typical  treatment and discussions are all focussed on supportive care and the the nuance only comes in when you get to BP management. Whereas in SAH you have a whole bunch of interesting and well proven interventions that can improve outcome for the lucky patients who haven’t already prognosticated themselves by presenting with a GCS of 3.

As a starter for 10: in which meningeal space in the brain do you find an SAH? The clue, thankfully  is in the name. The space between the brain adhering pia matter and the filmy arachnoid matter is where you’ll find an SAH. This is the space that CSF flows in from it’s genesis in the choroid plexi of the ventricles on its journey to reabsorption in the arachnoid granulations. Also in this space lies the cerebral vasculature that has a tendency to become aneurysmal and rupture arterial blood into this space.

Blood in the sub arachnoid space is easily seen on a simple dry CT scan, particularly in the first few hours. It has now become a test so good that people would suggest that if you have a negative CT in the first 6 hrs then you probably can skip the de rigeur LP that has been all the rage for the past century. Though i’ll admit that that question is delving much more into the realm of EM than hard core crit care.

In a critical care exam type stem you might be faced with someone in their 60s with a history of poorly controlled hypertension, who smokes, takes cocaine and has polycstic kidneys. All of these are identified as risk factors for SAH, though such a combination, i imagine only exists on exams. In the stem they’re likely to have a reduced GCS in the 13-14 range with a BP of somewhere north of 170mmHg. You’ll be given a CT scan showing some diffuse SAH but you’re waiting on an angio etc… Imagine a question like: what are your immediate priorities in management.

Given that 85% of SAH is aneurysmal, and they need definitive treatment likely not available in your hospital then getting that angio done is certainly a priority. But probably more acutely will be the basics of assessments of ABCs with particular attention to getting that BP under control. The biggest risk to life in the first few hours is going to be a rebleed which happens in maybe 20% of patients. Getting the BP down to somewhere south of 160 is likely a good idea with the ubiquitous labetalol probably being the most accessible and available option. Avoid the GTN and the foil wrapped madness of nitroprusside as both can cause a little cerebral vasodilation that you want to avoid. Bonus points for a decent analgesic (which will help the pain) and an antiemetic as vomiting does indeed tend to make the BP spike a little.

The stem continues and the plot thickens. While waiting for the CT angio the patient becomes obtunded gets intubated (where great attention was paid to the heamodynamics). Now the CT shows more blood, some hydrocephalus and a big posterior communicating aneurysm. What now genius?

Hydrocephalus is a relatively common event in SAH and the theory is that blood in the CSF space blocks up the arachnoid granulations preventing reabsorption and with ongoing production and failure to reabsorb you get hydrocephalus. There may be other reasons including a clot in one of the intricate drainage canals in the brain but either way you get more CSF than you want with a concomitant rise in ICP that quickly becomes life threatening if not drained with an EVD. Now if you’re a neurosurgeon and someone gives you the story of GCS 13-14 with SAH and an aneurysm your interest is definitely piqued but this is likely to be a transfer within 24 hrs to get some coiling done but is unlikely to require any surgery per se by the neurosurgeon at 3am. However if you give them that same story but now add some hydro and a falling GCS you have the type of thing that will buy your patient an emergency blue light transfer over to the OT in the neurosurgical centre.

So what’s going to happen with the aneurysm? Assuming we keep the BP under control and correct any coagulopathy then it needs secured. If possible this should be done by a neurointerventionalist with a coiling procedure, and not by craniotomy and clipping of aneurysm. This is now well defined and supported by randomised controlled level evidence. Given the complexity of these procedures the timescales provided for which it has to be secured is generally in the 24-48hr range and this allows them to be done as day time procedures most of the time.

The scene fades and the time jumps and the question stem now is in the ICU on day 5. On your daily sedation break it is noted the patient is not moving the left side. What pray tell is this new calamity? Has the poor soul now had a big embolic stroke in addition to his SAH? While this is definitely cerebral ischaemia it is not stroke and instead rejoices in the name of “Delayed cerebral ischaemia”  or DCI to its friends or “vasospasm” to the people it knew in high school but haven’t bothered to stay in touch.  DCI is a clearly recognized phenomenon in SAH patients and is typically found to co exist with the radiological phenomenon of vasospasm where the artery spasms and has reduced flow. Vasospasm itself is very common in up to 70% of SAH patients but only about 30% of vasospasm is DCI. DCI requires a focal deficit or drop in GCS in addition to imaging findings of ischaemia.

The typical response in days of yore would have been “triple H therapy” consisting of hypervolaemia, hypertension and high haemoglobin. Over the passage of years the only remaining tenet of the grlorious trio is induced hypertension which does seem to have some kind of effect on improving those ischaemic symptoms. Don’t be surprised to find your neurosurgeons requesting MAP targets like 90 or 100mmHg in these scenarios. The single best treatment for DCI we have is actually a prophylactic treatment in the form of the calcium channel blocker nimodipine. 60mg given 4 hrly is the standard recommended from day of rupture given for 3 weeks or so. It’s not entirely clear how this works but it’s fairly clear that it does. lastly it’s worth noting that DCI is not a day 1 concern for your SAH patient with incidence peaking in the 4-10 day range.

Reading:

As a broad overview Oh Chapter 51 covers all the bases.

Welcome back to the tasty morsels of critical care podcast.

Today we are going to talk about triggering on the ventilator. Now given the ubiquity of the word “triggering” in contemporary discourse I must confess that i do find it quite “triggering” to walk up to a vent and see the pressure support set at 11 or some other horror show like a PEEP of 7… I mean, who would do such a thing. But let me clear we are talking about a very different type of triggering.

If i was on a ventilator and somewhat engaged in the process of respiration at least at a brainstem level, I would feel a much more content if the ventilator cycled to inspiration whenever I requested it to. Indeed I would also find myself greatly contented if said ventilator did not randomly produce new inspirations any time it detected the slightest change in airway pressure. All of this is dependant on ventilator triggering.

Let’s start with the basics, the ventilator can be triggered to cycle to inspiration in a number of ways:

1. time (in the case of mandatory ventilation, in fairness this is not really a trigger as the patient has no input)
2. pressure trigger. The patient must produce enough negative pressure in the circuit to trigger the vent
3. flow. The patient must produce a certain amount of inspiratory flow in the circuit to trigger the vent

My experience has been overwhelmingly with the ubiquitous servo ventilators found in many ICUs in Ireland. On the servo-i when you scroll through the menus you’ll see a dial for trigger. This dial is defaulted to flow trigger with a dimensionless number from 1-10 based on a proprietary software from Maquet. The more clockwise you turn the knob the lower the flow in the circuit the patient has to generate and therefore the easier it is to trigger inspiration. Swing it all the way right for the poor GBS patient who struggles to trigger. As the dial is turned left (or anticlockwise) then the trigger will magically switch to a pressure trigger with actual numbers in cm H20. These define the negative pressure in the circuit that has to be generated before the vent will trigger a breath. Thus flow triggers are easier for the patient and pressure triggers harder.

But when would you ever want to make the trigger harder for the patient? Typically it’s not actually that you want to make it harder for the patient, it’s more that you want to avoid autotriggering. A good example of auto triggering is commonly seen in the patient who has become dead by neurological criteria. The story at handover will typically be a  devastating brain injury with some haemodynamic instability and loss of pupilary and cough reflexes but the trainee notes that brain death cannot have occurred because they are still triggering the vent. In this scenario it is quite common for the ventilatory to be auto triggering due to the minor fluctuations of flow within the circuit caused by the substantial cardiac oscillations of the hyperdynamic circulation of the person undergoing  brain death. Simply switching from a flow trigger to a pressure trigger typically eliminates these auto triggers. Alternate sources of auto triggering can be the big air leaks of a bronchopleural fistula or a water logged circuit with a meniscus of rained out water oscillating back and forth in the tubing.

Failure of triggering is very common. In this scenario there has been a neurological trigger that may have even initiated some diaphragmatic contraction but it was missed by the ventilator. An oesophageal balloon is probably the gold standard here and you can use it to see if a negative deflection on the balloon is matched by a breath.

In the absence of a balloon (and aren’t we all?) we have to use some surrogates. It’s hard to detect but in some patients you can see a -ve deflection in the pressure waverform that is not matched by a breath. This may be the patient trying to trigger but failing. The flow waveform is similar but this time we’re looking for a +ve deflection of the expiratory slope. There are some nice pictures in the multiple references at the end of the post.

While it may seem inconceivable to many there is always the option of actually examining the patient. A hand on the sternocleidomastoid or tummy might make patient generated effort easier to recognise.

Intrinsic PEEP or gas trapping is one of the commonest causes of a failed trigger. Let’s say a COPD patient is emerging from propofol and fentanyl induced haze of 3 or 4 days on the vent for pneumonia. They are transitioning to a spontaneous mode as their respiratory drive increases. Unfortunately their obstructive lung disease is still an issue and the expiratory flow has not returned to zero before they try and take their next breath. Air is still exiting their body at a certain flow and pressure so they need to generate enough flow and pressure to reverse this gas in the circuit in order for gas flow to move from expiratory to inspiratory limb to allow the vent to recognize a trigger. You can often see this as artefact in the flow waveform.

There is an interesting technology called NAVA or Neurally Adjusted Ventilatory Assist . This involves a fancy NG tube that is placed in the distal oesophagus and picks up electrical signals from the diaphragm. This is then connected to the vent and allows the vent to know with a high degree of precision when the diaphragm is contracting and match the beginning of the breath to this. So even if the diaphragm is weak and ineffective the NAVA can pick up on the neural signal to breathe. Like most such things it’s been tricky to bring to widespread practice and trials showing signficant benefit are sparse.

Moving on from failed triggers there are 2 more concepts to discuss. 1) double triggering 2) reverse triggering. These can look quite similar at times and are often mistaken from each other but are quite distinct. Double triggering can be seen when neural inspiration is longer than mechanical inspiration; in other words the patient wants to take a really long drawn out breath in but the vent for any number of reasons has cycled to expiration before the patient was finished. This will be particularly common in partially controlled mode of vent where you’ve tried to set a small and “safe” tidal volume but the patients brainstem is having none of it.

The second one, reverse triggering is much more recently described and can be really quite subtle. It is usually seen in deeply sedated patients undergoing a control mode of ventilation. In this scenario the vent triggers the breath itself based on the set program. During the mandatory breath the diaphragm is activated and so as soon as the mandatory breath is over the vent senses the diagphrgm induced flow change and cycles into inspiration again.  If you have something like NAVA or an oesophageal balloon you can see the diaphragmatic activation on the trace. Without one of these it can look almost like a hiccup. In addition look for a mandatory breath followed by a triggered breath during the expiratory phase.

As always this is by no means a comprehensive review of triggering but hopefully a little intro to some potential very examinable topics.

Reading:
– Georgopoulos, D. & Roussos, C. Control of breathing in mechanically ventilated patients. Eur Respir J 9, 2151–2160 (1996).
– Good lecture on triggering from Brochard at the Toronto course
– Dres, M., Rittayamai, N. & Brochard, L. Monitoring patient–ventilator asynchrony. Curr Opin Crit Care 22, 246–253 (2016).
– Artigas, R. M. et al. Reverse Triggering Dyssynchrony 24 h after Initiation of Mechanical Ventilation. Anesthesiology 134, 760–769 (2021).
– Oto, B., Annesi, J. & Foley, R. J. Patient–ventilator dyssynchrony in the intensive care unit: A practical approach to diagnosis and management. Anaesth Intens Care 49, 86–97 (2021). [images above]

Welcome back to the tasty morsels of critical care podcast.

Today we are going to do our best to charm the yellow snake of the intensive care unit and cover the pulmonary artery floatation catheter. Like a lot, indeed practically all of these topics, I do not in any way consider myself to have great expertise in the topic but I have had to upskill as much as I possibly can in lieu of the typical mis spent youth doing cardiac anaesthesia that most of my colleagues have had.

As such the source list for this post is quite varied in terms of its references.

The focus here will be on the basis, the nuts and bolts of how to put in and what type of numbers you might obtain from a PAC.

The insertion carries a lot of similar complications to any typical central vascular access procedure. But the big ones come from the fact that you’re trying to place the catheter through the heart rather than in close proximity to it. Perforation is of course a real possibility but perhaps more likely are nasty arrhythmias precipitated by the catheter irritating the myocardium. Expect to see this more in the cold, shocked post bypass patient or in someone who’s already having a lot of arrhythmias.

The PAC is also famous for the knots it can manage to tie itself into that can make extraction more than a little challenging.

There are lots of good materials online on insertion so I’ll only mention a few basics in passing. The tiny little balloon at the tip catches the flow of the venous return and pulls the catheter along with the flow.

In the absence of flurosocopy it can be tricky to know quite where the tip of the catheter is at any given time so we use the changes in waveforms to tell us what chamber or vessel the tip is at any given time. The pattern we expect to see should be CVP waveform, RV waveform then PA waveform and finally a wedged waveform. If all plays ball the you should those patterns at roughly 20cm, 30cm, 40cm and 50 cm respectively. The challenge is usually transitioning from the RV to the PA and the key change in waveform to look for is the “step up” in the diastolic pressure from the RV waveform which has a diastolic in the low single digits to a PA diastolic which is in the low double digits.

Once the procedure bit is done we typically take a CXR looking for the tip. Typically the natural curve of the catheter leads it to ending up in the right PA most commonly though this is by no means guaranteed. It can be tricky to tell from a simple CXR but ideally we want the tip in a West zone 3 part of the lung, typically in the inferior portions. West zones may be a distant memory from medical school but for our purposes the estimate of the left atrial pressure produced by our pulmonary capillary wedge pressure is only valid when the alveolar pressure is less than the pulmonary venous pressure, a situation that exists only in West zone 3. If you’re in zone 3 you should be able to see a and v waves (analagous to the a and v waves of the CVP waveform)

In some of the linked papers at the end there are some excellent images of troubleshooting various waveforms. One of the more useful ones was dealing with the failing RV (the very scenario where a PAC is likely to be needed) In this scenario, the RV diastolic pressures can approach the PA diastolic pressures with a loss of the “step up” as you move into the PA. The key difference to note in this scenario is that when the PAC is in the RV the diastolic run off (the period before the next ejection) is upsloping and the disatolic run off is downsloping when the PAC is in the PA.

There are lots of measurements we can take from the PAC. Directly measured PA pressures are of course useful but the typical catheters used these days also have a thermodilution filament built in so that we can measure continuous cardiac output (on the principle that the RV cardiac output is equivalent to the LV cardiac output). The contemporary catheters use semi random pulses of heat (up to 44 degrees) in order to calculate a thermodilution cardiac output. In general it needs at least a 15% difference in CO to be detectable and it averages things over 5-10 minutes rather than from beat to beat. There is often a “stat CO” measure that averages it over a more like 60 seconds.

In another success of marketing over function there is typically a continuous oxygenation sensor at the tip of the catheter. This gives a continuous reading of the true mixed venous oxygenation but is probably worth calibrating with an actual co-oximetry reading from a blood gas taken from the tip of the catheter.

With a PAC in place we have the potential for measuring the pulmonary capillary wedge pressure which given a long number of assumptions can allow us to infer things like a left atrial pressure or left ventricular end diastolic pressure, key variables for assessing the filling status of the left heart. The principle involves the tip being in a west zone 3 branch vessel, the balloon is then blown up creating a theoretical continuous column of blood between the tip of the catheter and the left atrium. Once wedged the displayed number will typically be a mean, however the PAOP should be obtained at end expiration and in end diastole which often means reviewing a screenshot with your monitor and using a cursor to identify the pressure, timed at the onset of the QRS. There of course are lots of subtleties and caveats to the number obtained and even more about how to respond to it.

Finally if you want to be really hard core there is a way of compensating for the effect of high levels of PEEP (>10) on the PAOP. The transmission index (TI) gives you a “corrected” PAOP taking this into account. The TI is calculated by looking at the PAOP  in inspiration and expiration. The difference between these two numbers is then divided by the driving pressure on the ventilator, this is your TI. The corrected PAOP is then the measured PAOP minus the total PEEP multiplied by the TI. This type of maths does not translate well to audio format and indeed there are actually several of these calculations available just to make it even more confusing.

There is a substantial literature behind the utility, or lack thereof of the PAC that has led to a massive decline in its use preceding the mid noughties when i started practicing. However they remain a key tool in the intensivists arsenal and if you deal with sick hearts on a regular basis it’s vital you have a decent grasp on charming the yellow snake.

Reading:

Irwin & Rippe Chapter 19 (an excellent source of a textbook if you want detail on any topic not particularly well served by Oh)

Deranged physiology has as expected an even higher level of excruciating details for those interests, presented of course in an excellent fashion.

LITFL

– Bootsma, I. T., Boerma, E. C., Scheeren, T. W. L. & Lange, F. de. The contemporary pulmonary artery catheter. Part 2: measurements, limitations, and clinical applications. J Clin Monitor Comp 1–15 (2021) doi:10.1007/s10877-021-00673-5. – Bootsma, I. T., Boerma, E. C., Lange, F. de & Scheeren, T. W. L. The contemporary pulmonary artery catheter. Part 1: placement and waveform analysis. J Clin Monitor Comp 1–11 (2021) doi:10.1007/s10877-021-00662-8. – Teboul JL, Pinsky MR, Mercat A, Anguel N, Bernardin G, Achard JM, Boulain T, Richard C. Estimating cardiac filling pressure in mechanically ventilated patients with hyperinflation. Crit Care Med. 2000 Nov;28(11):3631-6. doi: 10.1097/00003246-200011000-00014. PMID: 11098965

Welcome back to the tasty morsels of critical care podcast.

The subject of solid tumours in the ICU gets a whole chapter in Oh’s hallowed pages, number 46. I suppose the term solid is in place to distinguish it from the “liquid” tumours of the bone marrow and domain of the haematologists, something we covered in tasty morsel number 58

Historically the idea of admitting people to the ICU with malignancy was somewhat unusual and the idea of admitting someone with some degree of metastatic disease even more heretical. But now on a daily occurrence I find people admitted post op from debulking surgery, or hepatic resections for metastases or the slightly “out there” concept of HIPEC procedures (heated intraperitoneal chemotherapy). Many of these patients are on a potentially curative trajectory not thought possible 20 years ago.

That being said many patients with disseminated malignancy, even those on active treatment are typically approaching end of life when the old multi organ failure and ICU referral appears.

An intensive care unit admission is bad for cancer as there is a sort of immunosupression that comes about with being critically ill, leading to depletion in things like natural killer cells who were rightfully killing off cancer cells till the pneumonia, steroids and blood transfusions came along. For example the beta stimulation from adrenergic agents has a direct effect on reducing cytotoxic t cells and natural killer cells.

Let’s look at some complications of chemotherapeutic agents that might land a patient in the ICU. One would hope that the oncologist will pick this up but it’s still important for us to have an awareness of the potential issues. I’m going to skip over the most common presentation to us , that of neutropaenic sepsis from a beaten down bone marrow, and instead focus on the niche ones.

Bleomycin is a relatively commonly used agent for multiple malignancies, perhaps most famous for its potential to cause lung injury, namely a form of pulmonary fibrosis. The pneumonitis associated has been reported in up to 40% of patients according to Oh, but a quick review of the relevant UTD article suggests it’s more like 10%. Will generally be seen in the first 6 months since treatment but can occasionally be late. So the cancer patient with new diffuse alveolitis has a very broad differential including multiple infections but perhaps wise to keep bleomycin in your head. Don’t be surprised if someone gives lots of steroids as that might be beneficial.

Ifosfamide (a sort of cyclophosphamide in disguise) is another agent commonly used in multiple cancer types and is famously associated with an encephalopathy (and also a nephrotoxicity that i mention only in passing). This neurotoxicity can happen in around 20% of patients and is thought to be related to a break down product rejoicing in the name chloroacetaldehyde. Both Oh’s manual, and the oncologists in the two cases i saw gave methylene blue as there was a suggestion it might help. Interestingly the European society for oncology guidelines recommend specifically against using it.

The anthracyclines are a broad group of drugs including such tongue twisters as daunorubicin, doxorubicin, and epirubicin. These are the kind of drugs that can bring a patient to the ICU 5 years after their cancer treatment with severe heart failure and dilated cardiomyopathy. This type of toxicity is one reason for the pre treatment echo requested on these patients and the interest in things like global longitudinal strain to predict early signs of complications of treatment.

Finally I’ll mention some cautions with regards to platinum based chemo agents, easily identified by the “platin” at their end, the most common one being cisplatin. I’ve noticed our chemo protocols have specific note not to use gentamicin in neutropaenic sepsis associated with platinum based chemo agents. Given that gentamicin is one of my top 10 drugs, up there with propofol and steroids, and noradrenaline, it made me intrigued as to why. After a brief and really quite shallow rabbit hole it seems that the prohibition is all related to ototoxicity. Aminoglycosides, we all know are associated with ototoxicity and it seems cisplatin does the same by a very similar mechanism. There doesn’t seem to be any kind of synergy or pharmacological trickery here, it’s simply that giving two drugs with risk to the inner ear is simply a bad idea.

The massive monoclonal antibody shaped hole in this post is the immunotherapeutic agents now so common in much of modern oncology. They have their own list of very specific complications that are deserving of their own post but for now I would point you towards the excellent IBCC post and podcast that cover the topic beautifully.

Reading:

Oh’s Manual Chapter 46

Schacht J, Talaska AE, Rybak LP. Cisplatin and aminoglycoside antibiotics: hearing loss and its prevention. Anat Rec (Hoboken). 2012 Nov;295(11):1837-50. doi: 10.1002/ar.22578. Epub 2012 Oct 8. PMID: 23045231; PMCID: PMC3596108.

Welcome back to the tasty morsels of critical care podcast.

This time we look at Oh Chapter 52, focused on cerebral protection. There is, I must admit some repetition and cross over here, particularly with tasty morsels 20 and 39 respectively which cover more with regards to TBI. But in all honesty a little repetition is often very helpful for such subjects. We talk a lot about cerebral protective measures in the ICU and hopefully this will give you a little of the basic physiological background.

We’ll start with a few basic factoids. The brain apparently receives 15% of the cardiac output, though I imagine by the end of a typical ICU on call shift, that proportion will have dropped quite significantly. The squishy pale blob of folds and ridges in our skulls uses a surprising amount of glucose and oxygen and is very dependent on a continuous and uninterrupted supply of glucose, or ketones to allow ATP generation. It is not an organ able to tolerate an oxygen debt and has no real capacity for anaerobic metabolism. Basically it’s a bit of a diva and seems to take the position that as its the only organ capable of producing consciousness and self awareness that therefore it’s a bit special.

It’s circulation has some redundancy built in with 4 separate vessels (2 verts and 2 internal carotids) all filtering into the big round about that is the circle of Willis. However once you take the branches off this circle the redundancy and collaterals are somewhat lacking.

As a reminder cerebral perfusion pressure is equal to MAP – ICP. So for most normal humans from day to day this works out at a perfusion pressure of ~60mmHg. Autoregulation of blood supply to brain is very well controlled being able to control perfusion and flow very precisely anywhere in the ranges of MAPs from 60-160mmHg. It does this by vasoconstricting when arterial pressure is high and vasodilating when pressure is low. As expected chronicity will have some effect on the brain’s response here.

There is an intricate mechanism called “flow-metabolism coupling” that allows the brain to match supply with demand. The mechanism for this is described in Oh as unclear with lots of intriguing theories but it remains hard to grasp how metabolic products being washed away on the venous side cause a neatly matched dilation of the arterial side of things.

Outside of that broad pressure range of 60-160mmHg, flow becomes more linear with higher pressures leading to higher flow and lower pressures to less flow. In the injured brain (either traumatically or medically) this autoregulation becomes much more challenged and we find ourselves stepping into augment MAPs with things like vasopressors in order to ensure a reasonable CPP.

There are also systemic (ie non local factors) that can significantly affect cerebral blood flow. Most notably CO2 and temperature. CBF (cerebral blood flow) increases 3-4% for each mmHg increase in PaCO2. For those of us tied to kPa it’s perhaps easier to express it as a doubling of PaCO2 will double the cerebral blood flow and having the CO2 will have an equivalent effect. The overall metabolic rate of the cerebral tissues is lowered 8% for every degree celcius that the body temperature is lowered. Being able to reduce flow to the brain (either by hyperventilating or cooling) is a nifty trick when you’re in trouble with the ICP. However hyperventilating is reducing flow to a brain that needs it while temperature control is reducing the actual need for the flow and therefore clearly seems the more elegant of the two. As mentioned in prior posts it’s fairly definitive at this stage that therapeutcially cooling the injured brain to sub normal temps does not improve outcomes but it would still seem prudent to take the TBI with a temp of 39 down to a more normal range. Equally, hyperventilating to a PaCO2 of 3kPa might buy you an ICP drop on the lift to theatre but is not something you want to be doing as a definitive strategy.

Osmolality is another manipulable physiological variable we can tinker with. Oh states that a 3mosm drop in osmolality can lead to a 7% increase in overall cerebral volume as fluid shifts into the brain. An overnight sodium drop in a TBI patient from 135 to 130 all of a sudden seems a little more significant now. The most basic implications of this are to avoid hypotonic fluids in the brain injured patient and the more advanced might be manipulating the serum Na north of 150 in the hope of shriveling the brain up like a prune so it fits within the cranium instead of herniating out the foramen magnum.

As a little bit of a bonus, I’ve been working through Thomas Woodcock’s highly recommended “Fluid Physiology” textbook. The focus on this for me has been the revised starling mechanism, best described elsewhere online but there is a nice chapter on the CNS. He points out that recent studies suggest there is a form of lymphatic system in the brain (previously it was thought to be non existent) that probably helps with CSF re circulation beyond the traditional understanding of re absorption through the arachnoid granulations. He describes 3 forms of cerebral oedema evident.

1) vasogenic due to disruption of the blood brain barrier

2) cytotoxic due to plasma hypoosmolality, mostly thought to be related to failure of various ion pumps.

3) interstitial due to CSF circulation issues.

He also makes a nice point about mannitol. We claim this works by osmotic reabsorption. Something he says does not normally occur across an intact blood brain barrier. But if the endothelium and glycocalyx is damaged only then mannitol will becomes effective by the mechanism of osmotic reabsorption. Of no clinical import but interesting none the less.

Reading

Oh’s manual Chapter 52

Fluid Physiology by Thomas Woodcock.

Welcome back to the tasty morsels of critical care podcast.

In yet another departure from the stone tablets of Oh’s manual, today we’ll talk a little about one of favourite gram +ve cocci: staphylococcus aureus. Diagnosis and management of infections with this bug are a common occurence in the ICU and it behoves us to have a working knowledge of some of the complexities of its investigation and treatment that often fall to our micro or ID colleagues. So accept this for what it is – an intensivists summary of someone else’s expertise.

About 30% of us are colonized with staph aureus. Much of this is simple MSSA but there are increasing cohorts of MRSA in the mix there too.

When causing an actual infection we see staph aureus implicated in lots of different types of infection such as

• skin and soft tissue infections from cellulitis to foliculitis to abscesses to impetigo
• a toxin mediated GI illness from eating food contaminated with staph aureus
• bone and joint infections
• a sepsis with toxic shock syndrome
• pneumonias especially in the hospitalised population
• cerebral infections in the context of trauma, neurosurgery and drains
• blood stream infections from IE to device infection

Many of the list above will have nothing to do with critical care but staph really comes into its own when it gets into the blood. Getting a phone call from micro that your septic patient in the ICU has staph aureus in the blood is a fairly common phenomenon and should trigger a specific sort of response. A staph bacteraemia carries with it a mortality of roughly 20% so it’s something to take very seriously.

This is a disease which has significant metastatic complications so once you realise there’s staph in the blood you should be hunting out the other sites it could have spread to. It’s often difficult to tease out if the infected site is a primary site of infeciton or has seeded form somewhere else but a good list of places to look at would include

• heart valves
• cardiac devices
• the spine
• bones and joints
• embolic abscesses in the lungs or even a primary pneumonia
• septic emboli in the brain
• skin and soft tissue infections

In ICU level staph bacteraemia I find it’s the heart, the spine and the lungs are the main sites of infection. Needless to say that if there’s a collection of pus somewhere in staph bacteraemia strong consideration should be given to draining it acknowledging the complexities of decompressing a spine etc…

The brain primarily becomes an issue in someone with IE and a valve that is falling to pieces. In these patients surgery is typically considered but there is some understandable reluctance to give 40000 units of heparin to anticoagulate someone for a bypass run when there may or not be a lump of staph aureus in the cerebrum. As a result they often end up getting MRI to ensure the brain is clean.

Interestingly the PET-CT has become a key tool in the diagnosis and management of staph bacteraemia. At least it seems to have become key in centres with timely access to PET-CT. I can say that I have never managed to get an ICU patient through a PET-CT and they inevitably get it later on when all the excitement of the ICU stay has died down.

Once we know there’s a staph aureus in the blood what should we be reaching for treatment wise? Typically the patient will already be on some antimicrobials before you even get the result. So usually it’s a question of rationalising the bug juice. Vancomycin is typically our go to until we get an ID as the MRSA status is often unknown at this point. If it turns out to be MSSA then the swap is typically to a high does flucloxacillin, say 2g  every 4-6 hrs. This swap is important as an anti stpahylococcal penicillin like fluclox is a better drug for killing staph aureus than vancomycin, so if it’s an option it should be used.

There are some nuances once you get into the weeds (and beyond the scope of an ICU exam) about addition of aminoglycosides (not routinely recommended) or addition of rifampicin (generally only in the context of prosthetic joints or hardware).

The next job is to document clearance from the blood. once on the bug juice repeat the cultures. If you’re killing the bug it should be gone within 48 hrs. failure to do so suggests an issue with source control so go back and have a think about what you need to image further – eg the heart valves, the spine etc..

Duration is split into uncomplicated and complicated. an example of uncomplicated might be someone with a PICC line staph bacteraemia that clears their cultures very quickly, gets the lines out and as a good quality negative TTE. They have uncomplicated staph aureus bacteraemia and typically 2 weeks is sufficient. Complicated might be someone with a spinal osteomyelitis not needing surgery and maybe it took a week before the cultures cleared. They likely need 6 weeks.

Reading:

largely a summary of the UTD article on staph bacteraemia.

Welcome back to the tasty morsels of critical care podcast.

Today we’re not so much looking at a chapter of Oh’s manual but at the physiologic concept of respiratory compliance. I approach this with a degree of trepidation as the probability of screwing this up is infinitely higher than simply translating Oh’s manual into podcast form.

Compliance is relatively simply defined as change in volume per change in pressure. Put another way, for every 1 cmH20 pressure i apply with the ventilator I get 100mls of volume. The compliance in this scenario would be 100, ie 100 divided by 1 = 100. 100ml/cmH20 also happens to be normal compliance of the human lung.  You’ll sometimes see compliance written as delta volume divided by delta pressure.

There are a few different types of compliance described for the respiratory system.

1. static compliance = compliance in the absence of flow. This consists of the compliance of the lung tissue and the chest wall and is the number we look at generally
2. dynamic compliance = compliance in the presence of flow. This consists of chest wall and lung tissue compliance AND airway resistance (and will always be lower than static compliance)
3. specific compliance = compliance normalised for lung volume (kind of like an indexed value so adults and kids can be compared)

Compliance will vary depending on distension of the lung with ideal compliance usually just above the FRC. When overdistended and about to pop, you can imagine that increases in pressure will only produce small changes in volume. The same is true when the lung is at very low, atelectactic volumes where the lung tissue is squished solid and large changes in pressure are needed to produce a change in volume. This is nicely represented in the graph from deranged physiology in the show notes that shows a nice sigmoid curve of lung volume plotted against airway pressure. The steep part of the curve represents the ideal compliance as you get the most “bang for your buck” in that small increases in pressure will result in substantial increases in volume.

We are very interested in lung compliance in the intensive care unit. We talk a lot about stiff lungs and spend a lot of our time and energies trying to optimise ventilation of those with poor compliance. So how do we measure or assess compliance?

This becomes a sort of reflex over time where you simply walk in the room and look at the vent and the driving pressure and the tidal volume produced and a synapse somewhere ignites and tells you  that 25cmH20 pressure to produce 250 mls of tidal volume is not good. If you do the basic calculation of delta volume divided by delta pressure, of 250/25 you get 10 which is indeed a very low compliance and of great concern. This is most of what you need to know for day to day practice.

However for examinations of brownie points you might wish to know more and understand many of the circumstances where that kind of heuristic might be wrong.

The gold standard is apparently something called the super syringe method which involves inflating the lung in 100ml increments with a 2-3 sec pause at each inflation. This measures static compliance and i mention it mainly cause it has a cool name.

In real life we measure compliance by fiddling with the inspiratory and expiratory hold buttons and looking at what the ventilator spits out. This is technically the compliance of the respiratory system rather than true static compliance but I remain somewhat in the dark as to the subtleties of the difference.

What you do with the number is a whole different question. Stiff lungs do worse. That’s hardly a surprise. Given that compliance is typically best just above the FRC we can titrate PEEP to idealised compliance. This is best explained on a critical care now post by Matt Siuba, linked in the show notes. The basics of this involve a passive patient in a volume control mode and the PEEP is dialled up and down with a fixed volume to see at which PEEP you get the best driving pressure (ie the lowest amount of pressure to produce the set volume).  This should place you on the steep part of that curve and just above the FRC.

There are actual a number of methods trying to attain the same thing and I don’t mean to imply that this is proven best but I have put a few links in the show notes for those looking more and will hopefully do a whole post on setting PEEP and recruitment at some point.

Reading

Deranged Physiology

Critical Care Now

Sahetya, S. K., Hager, D. N., Stephens, R. S., Needham, D. M. & Brower, R. G. PEEP Titration to Minimize Driving Pressure in Subjects With ARDS: A Prospective Physiological Study. Respiratory care 65, 583–589 (2020).

 

 

Welcome back to the tasty morsels of critical care podcast.

This time we’re looking at pulmonary hypertension. Mainly cause I recently had to give a talk on it so it’s fresh in my rapidly diminishing brain cells and thought I should get it all written down before I forget it. We’re going to try it as a 2 parter. Part 1 will cover a broad overview of pulmonary hypertension and part 2 will focus on management strategies for a PH patient in the ICU.

Saying a patient has PH does not really tell you very much. All we mean is that pressures in pulmonary circulation are higher than they should be. Saying someone has PH and not quantifying it is a little like saying someone has cancer but not saying which organ or how advanced it is. We need to go a bit further than just say they have PH and quantify the cause or rather which group of PH they’re in. We also need some way of quantifying the severity of it.

The definition of PH since the 2022 ESC guidelines is a mean PAP of 20mmHg on a right heart catheter. Echo can be used to screen for “probability” of PH but the right heart cath is needed to make the diagnosis. Once you’ve defined that the pressure is high the real doctory work begins as you have to figure out the likely cause. The language the guidelines use is “group”. You should be able to put your patient into 1 of 5 groups.

To give an example you are handed over someone who has known PH. You dig a little deeper and see they have an mPAP of 27 on a recent right heart cath. Their echo shows a poorly functioning LV and severe MR. The PH here is going to be group 2, PH secondary to left heart disease. This is by far the commonest.

Or another example, you are told someone has PH. You dig a little deeper and see an echo report that says the left heart works well but the right side is dilated. You dig a little deeper and see the clinic letters describing severe end stage emphysema. This is likely to be group 3 PH, PH secondary to lung disease.

In both those examples the PH is a problem but it is a downstream effect of other disease. And unless you can fix the heart or lung disease then the patient is in trouble, indeed if the patient dies in the coming weeks to months it’s likely going to be the left heart disease or the lung disease that kills them.

Let’s spend a few minutes talking about group 1 PH, sometimes called PAH. This is rare but often very severe and progressive and comes with some unique medications so it’s worth discussing. These people should have normal lung parenchyma and normal left hearts. There are a variety of specific causes in group 1 but a lot of it is described as “idiopathic”. It is a progressive pulmonary vasculopathy where the tiny arterioles suffer intimal proliferation and eventual fibrosis due to a variety of vasoactive molecules. This transforms the pulmonary circulation from a very compliant, low resistant circuit into a narrow and stiff group of pipes. The right heart is evolved and very comfortable with assisting large volumes of blood through a low resistance circuit. In hroup 1 PH, the change in pulmonary vascular resistance is more than the right heart can cope with and the right heart over time starts to fail in its primary purpose of maintaining a low CVP while delivering preload to the LV.

Over the past decades a number of classes of drugs have been developed that target the vasoactive molecules that cause the vascular changes. These can be split into 3 classes

1) endothelin receptor angtagonists which do exactly what the name says: reducing endothelin. Drugs like macitentan fall in that category

2) PDE5 inhibitors. These inhibit the enzyme you expect from the name but the key outcome is that there is an increase in nitric oxide something that causes pulmonary vasodialtion. Sildenafil or tadalafil are two common drugs in this group

3) prostacyclins. These vasodilate and reduce proliferation in the vascular bed and typically IV epoprostenol is the drug of choice here.

These drugs have proven disease modifying benefit but only in group 1 PH. We have not been able to prove any benefit for those with PH from left heart disease or lung disease. The more severe their disease the more drugs they might be on. Some patients are even on IV epoprostenol in the community to keep their PVR compatible with life.

These 3 classes of drugs have had a significant impact on both length and quality of life in PH. But the prognosis in group 1 PH is still one of progressive irreversible disease in the longer run. There are lots of features that are well validated on an outpatient basis to determine prognosis however that is rarely the question we’re faced with. For example we know ICU admission is a poor prognostic sign in severe PH but this is generally the very point we get involved at. As usual i suspect decisions about prognosis in the ICU setting are typically decisions about limitation or withdrawal of life sustaining therapy and they all depend on reversibility. If someone with severe PH has a pneunonia then we can probably turn that around then that’s something to consider. However if the right heart and liver are failing due to worsening congestion from progressive PH then that’s a different question.

That’s enough for today and to give you an overview of PH, next time we’ll focus on some management strategies.

Reading

My own rambling review of pulmonary hypertension on JFICMI website.

2022 ESC Guidance

Welcome back to the tasty morsels of critical care podcast.

Today we’re going to talk about some of the basics of some of our favorite drugs intensive care – the diuretics. As always this is planned to be a brief overview of the essentials rather than the deep dive.

https://memorang.com/flashcards/157643/Renal+Pharmacology+-+Diuretics+site+of+actionClick for source

As a starter pretty much all diuresis is conducted by convincing the kidney to lose more Na. Lose the Na and the water will follow.

First on the list of course is furosemide. This is one of the commonest drugs we use in intensive care and we really should just be mixing it in with the NG feed or the propofol given how commonly we use it.

Furosemide is one of several loop diuretics. By loop we mean its site of action is at the loop of henle, site of the much loved countercurrent mulitplication system. In particular furosemide acts by blocking the NaK2Cl pump in the thick ascending loop of henle. This sounds all very technical and impressive but how does blocking said channel cause an increase in the wee wee in the bag? Ultimately you end up with a lot more sodium arriving at the collecting duct. The presence of extra sodium in the collecting duct decreases the osmotic gradient between the medulla and the tubule and as a result less water is reabsorbed and more comes out in the urine.

Furosemide is normally highly protein bound. As a result it can’t get into the nephron through the glomerulus, which in the healthy state won’t let large things like albumin through. Therefore to get to its site of action in the loop of Henle it gets secreted into the proximal tubule and washed along with the ultrafiltrate towards the loop. This feature of secretion in the proximal tubules is one of the things we see with the furosemide stress test, typically used to predict need for RRT in AKI. A lack of response to a healthy (ie  at least 1mg/kg) dose of furosemide tells us the proximal tubules are in big trouble and there will be a likely need for RRT.

In terms of side effects the ones that are perhaps clinically most apparent are the electrolyte losses (primarily potassium and magnesium) and hypernatraemia (as the water loss is in excess of the Na loss). There is a corresponding “contraction alkalosis” that is nicely explained at the deranged physiology post or in audio form over at the curious clinicians. Longer term the one worth knowing about is the ototoxicity commonly seen with high doses of furosemide especially in conjunction with our other favourite ototoxic drugs – the aminoglycosides. Though to be honest, only our most chronically critically ill patients stay long enough with us for us to pick up the ototoxicity.

Next on the list are our thiazides. Typically for our local practice that means metolazone. Thiazides work just a little further down the windy nephron river from the loop of Henle at the distal convoluted tubule. Once there it inhibits the NaCl transporter system again meaning increased delivery of Na to the distal tubule where most of the water reabsorption occurs. Thiazides aren’t especially powerful as a diuretic strategy but they are additive from a Na wasting (and hence water losing) perspective as you’re targeting a different part of the nephron. I find the metolazone often gets added when you still want to diurese but you’re a bit worried about the rising Na. The idea is that you get the Na content of the urine to the sweet spot where you lose equal amounts of salt and water and the serum concentration stays the same. Like most things in ICU this is likely physiological wishful thinking rather than good science and it keeps us amused while the disease process resolves on its own.

Continuing our journey through the nephron we have the aldosterone receptor antagonists. A class largely occupied by spironolactone. Spiro (to its friends) works by blocking the very important ENaC (or epithelial sodium channel), especially in the collecting duct. When these are blocked Na no longer is reabsorbed in the collecting duct and hey presto water follows the Na out of the nephron into the ureters. The main side effect of the increased concentration of Na in the collecting duct will be a reluctance to secrete K into the duct thus preventing K wastage and ultimately increasing the serum K. It is not an especially effective diuretic in terms of producing volumes of urine but more importantly it does have a significant long term mortality benefit in patients with heart failure unlike crowd favourite furosemide. It is of course difficult to extrapolate findings from massive cardiology heart failure trials to the ventilated patient with a dodgy ticker with multi organ failure in the ICU but there you go.

The final drug we’ll mention today is acetazolamide. It has its site of action way back in the early nephron at the proximal convoluted tubule. It is a carbonic anhydrase inhibtor, unsurprisingly inhibiting the action of carbonic anhydrase. From the name “carbonic anydrase” we can hopefully deduce that it inhibits the process of removing water from carbonic acid. Ultimately this impairs HCO3 reabsorption at the proximal tubule creating a scenario somewhat similar to renal tubular acidosis. The drug clearly causes a diuresis and does indeed increase the Na wasted in the urine though the precise mechanism is not entirely clear. My anecdotal experience when taking the stuff climbing Kilimanjaro nearly 20 years ago suggests indeed it does make you want to pee a lot more. There are a a few small trials looking at its use in ICU none of which are hugely compelling for benefit but I find myself reaching for it when the fursoemide driven alkalosis is causing issues or you’re playing a game of “diuresis jedi” and want to complete all the steps of the “nephron bomb”

Reading:

Deranged Physiology

• Overview of diuretics

Curious Clinicians

Mullens, W., Verbrugge, F. H., Nijst, P. & Tang, W. H. W. Renal sodium avidity in heart failure: from pathophysiology to treatment strategies. Eur Heart J 38, 1872–1882 (2017).

Bell, R. & Mandalia, R. Diuretics and the kidney. Bja Educ 22, 216–223 (2022).

Welcome back to the tasty morsels of critical care podcast.

Today we’re looking at asthma. In reality I find this is much more commonly discussed than seen in real life. No doubt this is due in part, to an improvement in asthma care chronically which is of course a good thing. I think it gets discussed and comes up on exam papers so much partly because it is such a nice illustration of physiology and ventilation.

I am guilty of over teaching this myself having delivered not one, but 2 talks on the subject, and even a prior tasty morsel of EM on the subject.

Oh’s manual devotes a whole chapter, number 35 to the subject.

We definitely see much less of this than we used to. I suspect that’s largely due to better access and provision of primary care but there remains a cohort of fairly brittle folk out there who will occasional crop up in resus or the ICU.

To begin with, let’s cover some aetiology and pathophysiology, asthma has well described allergic and atopic associations as we all learn in medical school but also some important environmental triggers such as the infamous “thunderstorm asthma” that occurred in Australia some years back with 1000s of patients affected.

There are several major consequences of severe asthma that Oh describes:

1. increased work of breathing – dynamic hyperinflation a big part of this
2. V/Q mismatch and shunting
3. CV instability from intrathoracic pressure

Status asthmaticus is a term commonly found in textbooks but I don’t think it has anywhere near the utility of it’s Latin equivalent status epilepticus. Oh applies the term to those not responding to nebulised bronchodilators which could be fairly broad.

For management of asthma like this, the mainstay of treatment is inhaled beta agonists with a chaser of ipratropium and some steroid. There is plenty of evidence suggesting simple inhaled beta agonist with an MDI can be as effective as neublisation but for ICU level asthma (which this post is aimed at) you will be reaching for an oxygen driven nebuliser aiming to get particle sizes somewhere in the 1-3um range. however it is well known that <10% of the drug gets delivered to target and it is likely that in the most severe asthmatics where very little gas is moving that the drug delivery is even worse. Hence the existence of the IV therapies.

All of these are controversial on some level and I am not here to advocate for one or the other but more to provide a pithy line or two on each that one could reasonably throw into an SAQ or a viva answer and look somewhat smart.

IV salbutamol is commonly used in the UK and Ireland but like pretty much all of these therapies could not be said to have a robust evidence base. Concerns have been expressed that it adds a significant metabolic load to the work of breathing with the inevitable rise in lactate and fall in BE leading to an increased minute volume and an increase burden of respiration.

IV magnesium is likely more benign and given out extremely commonly in these cases but once again the evidence base is hardly stellar.

IV adrenaline is a common go to and has some physiolgoic rationale beyond flogging the already overstimulated beta agonists. It’s alpha agonist effect may have beneficial effects on secretion burden and plugging.

Aminophylline continues to be used though anecdotally I’ve not seen it to be that helpful

Heliox often crops up in the text books as it allows for the goldilocks phenomenon of laminar flow. However given the limtations on FiO2 at ~30% (generally 30:70 seems to be the mix) it’s not a great option in a population where hypoxia is a real concern.

No post would be complete without mention of crowd favourite ketamine and its potential bronchodilating properties. It is likely overstated but as an induction drug would seem reasonable. Inhalational anaesthetics are somewhat similar but importantly likely to be inaccessible when you need them.

Let’s say you’ve failed all these therapies and a tube has gone in. How would one ventilate such a patient. The answer should be “with great difficulty”. If it turns out they’re completely easy to ventilate with normal pressures and no gas trapping then you have just intubated someone with vocal cord dysfunction, sometimes known as paradoxical vocal cord motion. Worth another post perhaps but can be a common mimic of life threatening asthma.

More likely you’ll find yourself faced with a ventilator complaining loudly that all the pressures are too high. If lucky you’ll sort things before the inevitable CV collapse from intrathoracic pressures or tension pneumo. The emergent response if you’re faced with high pressures and hypotension is disconnect the patient from the vent. If high intrathoracic pressures are the problem then decompressing the thorax through the ETT may be  sufficient to temporarily fix the problem. If they have a PTX they’ll need a decompression of the pleural space of some kind.

Disconnecting the vent is of course not a long term solution so how should we appropriately ventilate these patients. The simple answer is probably “very slowly”. The best thing I’ve seen on this is Dave Tuxen’s paper from the late 1980s and the more recent podcast he recorded with the Intensive Podcast. A fairly simple summary is keep the minute volume low ie 6-7L/min at most. If you try harder you’ll cause harm. You can get this with a resp rate of around 10-12/minute with a Vt of ~500-600mls. Without even touching the I:E ratio you’ll end up with an expiratory time around 4 seconds. Tuxen argues that there’s little to be gained by prolonging expiration beyond this or lowering your minute volume below this. The 1980s paper provides some data to back this up. You will undoubtedly get high peak pressures on the vent that reflect airway resistance rather than the pressure being felt at the alveoli. Plateau pressures here are important here to ensure safe pressures and an aspirational goal of <25 cmH20 seems reasonably. Volume vs pressure control is a great debate it seems but my own preference would probably be volume control for what it is worth.

Finally my preference would be to have these people deeply sedated and even paralysed but there is an association between paralysis and a necrotic myositis that can be an issue in weaning and rehab and this is distinct from the usual ICU acquired weakness and myopathy.

References:

IBCC

Deranged Physiology

My talk at EuSEM 2021

David Tuxen on intensive podcast

Tuxen, D. V. & Lane, S. The Effects of Ventilatory Pattern on Hyperinflation, Airway Pressures, and Circulation in Mechanical Ventilation of Patients with Severe Air-Flow Obstruction. Am Rev Respir Dis 136, 872–879 (1987).

Welcome back to the tasty morsels of critical care podcast.

Oh chapter 26 devotes a whole chapter to this and for those of us in cardiac units the arrival of several post cardiac surgery patients a day in your unit is a routine part of the day. At least it was pre-pandemic anyhow. As such it presents a fairly predictable work load and patient cohort for your ICU. Given the bewildering number of conditions that can present to a mixed ICU on a given day, knowing that you have a couple of hearts moving along the production line provides a degree of predictability to the workflow.

The scheduled, usually elective nature of cardiac surgery lends itself to large scale outcome prediction and indeed, cardiac surgery has found its outcomes examined very closely over the past few decades. Outcomes are often examined in great detail and are more likely than most patient cohorts to be reflected onto the hospital and even individual surgeons.

Today we’ll focus mainly on the routine cardiac surgery patient and some of the common significant issues you’ll see.

Before we look at any specific cardiac surgeries, it is worth addressing the hand over. There is frankly a lot of information to assimilate and sift through. You will often be receiving handover from the anaesthesia team but also trying to tease out information from the surgeons almost simultaneously. Key points of information to glean are:

• pre surgical state, including what the heart looked like on TOE pre bypass – eg. is the ventricle good or bad at baseline
• what was done and how the plumbing stands as of now
• bypass time, cross clamp time
• how did separation from bypass go – are they on 2 pressors and 2 inotropes and some nitric? or have they arrived on some propofol and a tincture of noradrenaline and a big sticky label saying “extubate me”
• heart rhythm and presence and need for pacing (indeed if not being used is it at least on backup?)
• what did the heart look like on TOE following the surgery – is the valvular lesion fixed? are they, dare i use the terrible phrase, “underfilled”
• products given and current state of reversal and need for protamine
• where the drains are and what they’re doing (are they just mediastinal or are they pleural too? Was that 400mls of red stuff there when you left theatre or has it just appeared)

Contra to most ICU patients, cardiac surgery patients often benefit from a bit of the salty water stuff. Likely driven by rewarming induced vasodilation and hypothermia induced diuresis they can be hypovolaemic. It doesn’t take them long to transition to the more conventional ICU patient where fluid does nothing but increase the oedema but in the first 6-12 hours fluid resuscitation often has a role.

Episodes of hypotension are common and the major concerning causes are going to be surgical bleeding or tamponade. The drains that you checked at handover are both diagnostic and therapeutic. A big gush of blood from the drain and hypotension usually points to the problem while at the same time relieving any potential tamponade. Significant bleeding might be >200ml/hr in the first hour or two and >100ml/hr after that.

Unfortunately blood can clot focally around the heart causing a focal tamponade. This can be a bit trickier to diagnose as one might imagine, and indeed some form of imaging is often needed to make the diagnosis. Oh is very down on the utility of TTE in cardiac surgery patients and while of course you aren’t going to get all the windows, a good sonographer can usually answer many of the key questions. That being said if the patient is crashing you often need a TOE to look at the tricky spots like behind the LA where clots have tendency to form obstructing LA inflow. Both surgical bleeding and tamponade are surgical issues that once diagnosed should prompt a return to theatre. Any major issues with shock in the first 12-24 hrs should always have “return to theatre” somewhere on the list of possibilities. Finally for tamponade don’t forget the CVP. While much maligned in general critical care, when it comes to cardiac surgical patients the CVP provides some nice screening information for issues like tamponade and RV function

Atrial fibrillation is common everywhere in the ICU and it is unsurprisingly common in the cardiac surgery population. Its incidence comes in somewhere around 20%. Much of it occurs beyond the first 24 hrs, often when they have left my particular unit. Causes are numerous and much of the basics will happen automatically with correction of potassium and magnesium. It remains unclear what the best way is to manage cardiac surgery related a fib, but the usual suspects of beta blockers and amiodarone are the commonest interventions with vitamin A leading the way, particularly when they’re still a bit sick and shocked and ventilated, and beta blockers playing more of a role once the tube is out and the pressors are gone. Cardioversion is probably not the way to go for these patients.

Interestingly I just read an RCT that suggesting leaving the pericardium open a small bit at the end of surgery reduced the rates of post op a fib from 30% to 17% with no significant consequences. It was single centre and may well disappear into the midst of unreproducible research but would be a nice move if it turns out.

Reading:

Oh Chapter 26

 

Welcome back to the tasty morsels of critical care podcast.

Today we’re talking about dead space. While it may sound like something from The Expanse, we’re actually talking about the physiological concept of dead space here. This is pretty core physiology that crops up in clinical practice all the time so I think it’s worth thinking about.

As usual this represents a sort of idiot’s guide to the topic with just enough information to scrape by in an exam and in clinical practice but likely with large gaps, simplifications and occasional frank errors in description.

Definition is “the fraction of tidal volume which does not participate in gas exchange.” But let’s be clear the word participation here refers more to an inability rather than a surly choice by the dead space fraction not to participate as it didn’t get picked for football till last. The dead space fraction never has the option of participating in gas exchange as it never reaches any functional gas exchange surface.

At its most basic (and that’s the only form I’m interested in) it can be split into:

• Apparatus dead space – the amount of gas in the circuit and associated dongles like ETT, an NIV face mask or an HME.
• Physiological dead space – this is split further into:
  • anatomic- gas in the conducting airways and
  • alveolar – gas in non perfused alveoli

Phsyiological dead space usually takes up ~20-30% of the Vt. As mentioned above it splits into two components, anatomic and alveolar. As you can imagine the anatomic is pretty fixed but the alveolar dead space can vary markedly depending on V/Q matching.

Anatomic dead space is ~2ml/kg (about 150mls) but this includes the oropharynx that will be bypassed with the placement of an ETT or even better a tracheostomy with both of these interventions reducing anatomic dead space. I think the most important clinical take away about anatomical dead space is that it is fairly fixed. Assuming a 2ml/kg anatomic dead space, if you’re ventilating someone at 8ml/kg PBW and want to reduce to 6ml/kg PBW the fraction of anatomic dead space in each breath goes from 20% to 33%. In other words, while you’ve only reduced the Vt by 20% you’ve reduced the portion of gas participating in gas exchange by a third. There is of course good empiric evidence that a lower Vt is better but in terms of clearing CO2 dropping the Vt disproportionately reduces the fraction of gas available at the alveolus and may cause big issues with your CO2. Indeed at some point a rate reduction rather than Vt reduction may be the more favorable factor to reduce overall mechanical power delivered to the lung. That all seees very persuasive and logical but is countered by the simple fact that it doesn’t seem to be true when tested.

It seems that at very low Vt gas exchange continues to be more effective than one might expect likely due to 2 mechanisms beyond simple mass gas movement.

• laminar flow occurs allowing a central column of gas to move in and out
• there is expiratory gas mixing – basically diffusive gas mixing that ensures the right molecules are in the right place at the right time

Moving onto alveolar dead space, there are a number of things that might increase it:

• reduced cardiac output: eventually lung units stop receiving perfusion
• parenchymal disease: air space cavities no longer with effective perfusion due to thickening of interstitium
• high airway pressures: ensures aerated alveoli but may limit blood inflow to the lung unit
• pulmonary vascular occlusion: eg PE (one of probably several mechanisms of hypoxia)
• Posture – this will affect the west zone of particular lung units

The main consequence of increased dead space will be primarily seen in your CO2 with either hypercapnia or a requirement for a huge minute volume. As noted in the alveolar gas equation it will affect oxygenation much less but eventually it will impair oxygenation.

Reading

Deranged Physiology

The unfortunately short lived and much missed basic science clinic podcasts from Steve Morgan and Sophie Connolly.

Welcome back to the tasty morsels of critical care podcast.

Of the many things I poorly understand, I suspect that haematology holds a special place. Knowing the intricacies of the haematological malignancies was not exactly core knowledge for emergency medicine and to be fair an exhaustive knowledge is hardly key to ICM either. However in ICM there is a need to have a broad understanding of what some of the haematological acronyms might mean given that a fair number of these patients end up in the ICU. Most of this post will be navigating the basics of the diseases rather than super specific ICU management.

Oh dedicates a whole chapter, number 101 to the haematoloigical malignancies implying that it is certainly worth our attention. As a broad definition haematological malignancies involve the bone marrow or the lymphoid tissue, they occupy a different niche in the oncology world with the haematologists running the show rather than the general oncologists. They are also distinct in histology and outcomes from the solid organ malignancies.

We’ll start with the leukaemias and these can be split neatly into myeloid and lymphoid leukaemia. The cells gone bad in AML are the myeloid precursor cells, the cells gone bad in ALL are the lymphoid precursor cells. This type of statement is however only useful if you have any concept of how a myeloid precursor cell is different from any other type of cell in the bone marrow.

The attached image on the show notes, comes from the leading source of all medical knowledge, wikipedia. It’s a nice overview of the different types of cells stemming (see what i did there…) from myeloid and lymphoid precursors. It clearly works really well as an educational aid in audio form…

Myeloid cells differentiate into, well, most of blood cells that appear on your FBC, things like red cells, platelets, neutrophils, basophils. On the other hand, lymphoid cells have a much smaller and narrower family tree differentiating into different types of lymphocytes and plasma cells.

For AML there are a variety of causes from various genetically triggered issues, to transformation from a myelodysplastic syndrome or related to prior chemo or radiotherapy. It also includes the very ICU relevant disease of acute promyelocytic leukaemia. As a result, expect to see more AML in the older adult population. ALL is much more common in younger people with a much heavier CNS component, hence the prevalence of intra thecal treatment.

Each of the acute leukaemias has it’s own chronic version. With CLL being a form low grade lymphoma. CML begins as a chronic, somewhat indolent process that accelerates towards a blast crisis towards the end of the disease and for most people is more of a comorbidity than a malignancy.

Lymphomas, understandably come from lymphoid cells, these could be b cells or t cells for example. Classically lymphomas get lumped into two big categories of hodgkins and non-hodgkins with the former generally having the better outcomes.

Finally on the list of common haematological malignancies is multiple myeloma. This is a cancer of plasma cells which are the grown up and left home versions of B lymphocytes. In general plasma cells have developed to produce large amounts of proteinaceous antibodies and are triggered as part of an immune response. In myeloma they are inappropriately making large amounts of their specific protein or globulin reflected in the high total protein count and hyponatraemia associated with this disease.

So that’s a very broad, sub medical student overview of the different malignancies but why would they end up in your ICU?

Sepsis is probably number 1 on the list. People have no immune system due to marrow infiltration or their marrow being wiped out with treatment, and they tend to struggle with the old bugs. They get all the usual bugs that we see, but also we need to worry about candida and aspergillus, plus all their vascular access devices and a wide range of other opportunistic infections. Treat early and broadly and involve your micro or ID folk to be sure you’re getting the right cover for the known and unknowns.

There are a number of rather intense treatments used for haematological malignancy that often precipitate an ICU admission. We’ve already mentioned infection which is a common sequelae of chemotherapy but we’ve covered a lot of the issues in tasty morsel number 34 on chemo agents.

Overall chemotherapy treatment can be viewed in 3 stages:

1. induction: achieving remission by making the level of leukaemic cells undetectable
2. consolidation: eliminates residual undetectable disease – this is when you might tentatively start using the word “cure”
3. maintenance: used sometimes in ALL and AML to maintain remission (stage 1)

Predicting outcomes in these patients is tricky and historically there has been a degree of skepticsm in admitting them to intensive care as things end badly with a high degree of frequency. Oh quotes a large case series that currently puts mortality at 60% for this cohort of ICU patients. This is of course high but when you compare it to things like OOHCA it’s certainly not awful.

Finally haematopoetic stem cell transplant is an increasingly used potentially curative treatment for all kinds of hematological malignancies. This has a whole variety of specific indications and complications and rightly deserves its own note in due course.

Reading

Oh Chpater 101

 

 

Welcome back to the tasty morsels of critical care podcast.

Today we’re looking at a small section of Oh Chapter 58 covering myasthenia gravis. I don’t think I’ve ever looked after a true myasthenic crisis in the ICU. Likely because they’re well managed by their neurologists on an OPD basis or well managed from an anaesthetic perspetive when they need an operation done. I have made the diagnosis twice de novo in the ED (or at least admitted them with that as the leadning diagosis) so it is out there.

It does make excellent exam level material as there is some interesting physiology and compare and contrast type tables to be made comapring with other neuromuscular diseases.

To give a flavour of what might you see in the ED (and these patients will rarely need ICU) it’s typically some kind of cranial nerve issue, typically ptosis and diplopia complaints, sometimes with some speech and swallowing issues. The cardinal features of the neuro dysfunction is flucutating weakness. Typically this is described as fatigueability. For example the ptosis isn’t too bad in the morning but by afternoon it’s much worse.

Involvement of bulbar muscles (BTW, bulbar being an archaic name for the medulla and the cranial nerves that stem off it) should be recongised as a bit of a concern given that swallowing and airway protection fall under the remit of cranial nerves 9-12.

The edrophonium test that you may have heard about in medical school can be safely forgotten about as it is no longer recommended. On the other hand the ice test can be used as a cool demonstration of the physiology. In its essence, the ptosis that improves after having some ice on the eyelids would suggest myasthenia.

The pathophysiology of the illness is likely one of the more testable aspects here. Myasthenia is an auto immune disease where antibodies are made against acetylcholine receptors in the post synaptic neuromuscular junction. As a result Ach cannot bind to these receptors and therefore cannot complete the transmission of neurologic impulse to the muscle. The ice test works because NM transmission is apparently more efficient at lower temperatures.

Once the diagnosis is made you can look super smart by thinking about their thymus as thymomas are found in ~15% of people with antibody +ve myasthenia. Many more are found to have some kind of abnormality in the normally negleted and unloved thymus.

As an outpatient these people will typically be established on pyridostigmine, a nifty medication that potentiates the remaining ACh at the NM junction. They are usually immunosupopressed on some kind of steroid or maybe some azathioprine. Some may have had their unsuspecting thymus removed in the interim.

In the ICU we’re likely to see someone in myasthenic crisis. This is commonly seen when tapering immunosuppressives or when faced with some sort of actue stress like spending time under anaesthesia while a surgeon ectomises some part of your body. There are also a large list of drugs that we can misprescribe that can mess people up.

The fundamental feature of a myasthenic crisis will be respiratory insufficiency, this is defined as need for NIV or intubation. Remember it is unusual for myasthenia to affect respiratory muscles so if it is you’re looking at big trouble. Expect this to be a quiet, undramatic sort of respiratory failure. FVC and cough will quietly disappear without any of the usual increased work of breathing we usually use to quantify respiratory failure. Hence they look fine until they’re really not.

There are a variety of vital capacity cut offs described as reasons to intubate. But as discussed in the GBS post these are somewhat arbitary. For exams a VC of 15ml/kg is certainly a good red flag to keep in mind. Conveniently the same number can be spouted for a question on GBS.

Without getting into the weeds on NMBA in myasthenia, for a non depolarising agent like roc you can expect the effect to last much longer so people have suggested a smaller dose. Given that if they’re being intuabted in ICU the prolonged effect is much less of a concern as long as you’re runnning the sedation which of course you are.

Once they’re in ICU and ick enough to be tubed we need chart ourselves and the patient a way out of this mess. The key treatments for both real life and for examinations are going to be plasma exchange or IVIG. Plasma exchange removes all the nasty antibodies. IVIG on the other hand does something akin to witchacraft and probably binds the antibodies. There is no clear data favouring one over the other with the only RCT being neutral but the trend it seems is towards PLEX.

As is typical for these immune binding/removal type therapies they need an immunosuppressive “chaser” to stop production of more autoantibodies. Typcially this will be steroids and typically you’ll not be making the decision anyhow.

The more interesting question for us is time to resolution. A common quoted median is 2 weeks of invasive ventilation. This is much shorter than might be expected for a GBS case where it is not uncommon to intubate and tracheostomise in the same day given the expected course. For myasthenia it might be reasonable to give them some time to assess response to treatment before commiting to the tracheostomy.

In terms of meds to be cautious with, the list is long but the commonly implicated bad actors include NMBA, aminoglycosides, the fluroquinonlones and crowd favourite magnesium is more likely to cause NM weakness than usual.

Finally it’s worth knowing that there is such a thing as a “cholinergic crisis” desrcibed in myasthenia that is due to excessive cholinergic effects from too much pyridostigmine. It is vanishingly rare at this stage. It’s interest is that it forms another cause of respiratory failure in the myasthenia patient that you might mistake for a myasthenic crisis but if you’re a betting man or woman (and in medicine we all are) then if your myasthenia patient has respiratory failure it’s goint to be the myasthenia almost every time

References

UTD

Oh’s manual 58

 

Welcome back to the tasty morsels of critical care podcast.

Today we look at everyone’s favorite mould – aspergillus. We see a number of fungal infections in the ICU, most commonly it’ll be the yeasts – forms of candida. Yeasts are single celled organisms. The moulds, of which aspergillus is a member are multicellular organisms. To continue into a brief foray of wikipedia inspired irrelevance the name aspergillus comes from the liturgical implement known as the aspergillum, more commonly known as the thing your priest man shakes to sprinkle the holy water. Apparently it was named aspergillus by the Italian priest who discovered it under his microscope and named it for the resemblance.

The biggest issue comes for us in the ICU in differentiating colonisation from active infection. Proper invasive pulmonary aspergillosis is characterised histologically by invasion across tissue planes, particularly into vessels. As you can imagine getting a lung biopsy to prove such on ICU patients can be a tad challenging so we’re stuck with the usual conundrum of trying to work it out based on probabilities and surrogate tests.

the LITFL entry has 4 types of aspergillosis described that we should be aware of:

1) allergic bronchopulm aspergillosis. This is generally an OPD condition that is rarely the cause of why the patient is in the ICU

2) the aspergilloma – the dirty great fungus ball hanging out in one of the lobes of the lung causing all kinds of bother. Surprisingly this can also often be an out patient problem

3) chronic necrotising pneumonia (described as semi-invasive in the post)

4) invasive pulmonary aspergillosis – the type we’re likely to see and most worried about.

The IBCC covers aspergillosis very well, and if this podcast does no more than refer you to the IBCC then my work is done. Josh makes an excellent point of pointing out that there is probably a different clinical pattern in the neutropaenic vs the non neutropaenic patient.

For any kind of fellowship exam you would be expected to reproduce a somewhat cogent list of risk factors for such an illness. Of note we are all exposed to aspergillus and exposure to aspergillus is a simple fact of being alive. But generally it doesn’t cause us a trouble unless something else is going on. A reasonable (but by no means complete) list of risk factors might include

• most famously stem cell and lung transplants or anyone with prolonged neutropaenia, typically >10 days (remember that haematological malignancy are the main groups to reach this level of prolonged neutropaenia)
• can also occur in much less immune suppresses people who are just really sick in ICU, or even COPD with recurrent steroids (typically at least 20mg/day of pred)
• cirrhotics
• patients post severe influenza seem to be a risk (looks like severe superinfection 3-5 days into course)
• most other solid organ transplants seem to be fairly low risk.
• COVID-19 as a risk factor is almost certainly a risk factor. However many of the studies have assumed that growth on a sputum implies invasive disease which is probably a bit of a stretch. That being said if you find that the whole airway is covered in plaques on bronch then you’ve probably established a diagnosis.

So a common clinical context might be a haematological cancer patient in the ICU as part of a neutropaenic sepsis process. They might be in a week or two with profound neutropaenia and develop recurrent fevers and a respiratory deterioation. A sequence of micro growth and adjunctive tests established a diagnosis of aspergillosis.

More recently we’ve seen it in the unfortunate chronically co morbid patient who gets a bad dose of the auld COVID and 2 weeks into their vent course they deteriorate and the bronch shows white plaques all the way down.

In terms of testing there are a number of potentially useful modalities. When it comes to radiology, CT is your friend and as always giving your radiologist a specific query might be helpful. Look for cavities, or a monod sign (air around a fungus ball). The two other signs described are the halo sign and finger in glove sign. Good examples available on radiopaedia.

In terms of blood tests we typically reach for the beta d glucan or galatcomannan. Both are cell wall compents of fungi but the galactomannan is probably more specific for aspergillus. Both have sensitivities quoted around 75% so they are by no means perfect. There are a variety of assays and types available most of which i defer to a microbiologist rather than attempt to understand. There were historical issues of false +ve with beta lactams like pip-tazo but apparently these are historical issues and not relevant to contemporary assays.

Bronchoscopy is typically performed to get a decent sample. A galactomannan on the BAL sample is something we tend to lean a lot heavier on than the serum test. In my reading i saw reports of eosinophilia and elevated serum IgE being associated with aspergillosis but have not seen that in the wild.

Ultimately making the diagnosis in the ICU is very difficult as there are lots of confounders – mainly being lots of colonisation and it can be difficult to distinguish from invasive disease. Open lung biopsy as a gold standard comes with a sensitivity of 60% so let’s just bin that idea. In general we look for it and if the patient fulfills the “sick as shit” category (which they almost always do) then we treat.

When it comes to treatment there are a few recommendations you can choose from. In general voriconazole leads the way. There have been some recent shortages of the IV form so it’s nice to know the PO version works pretty well. It does not need some monitoring of levels. Occasionally there’ll be dual antifungal cover but as always i would take advice from an expert in that type of situation.

If there’s a huge aspergilloma it’s important to note that surgical resection is a very realistic option but as you might suspect this applies to the slightly more stable population.

UPDATE:

Since recording this one of my colleagues who does both ID and ICU pointed out this paper that looks particularly at aspergillus in the ECMO population.

Rodriguez-Goncer, I. et al. Invasive pulmonary aspergillosis is associated with adverse clinical outcomes in critically ill patients receiving veno-venous extracorporeal membrane oxygenation. Eur J Clin Microbiol 37, 1251–1257 (2018).

References:

• IBCC
• ATS 2019 Guidance: Hage, C. A. et al. Microbiological Laboratory Testing in the Diagnosis of Fungal Infections in Pulmonary and Critical Care Practice. An Official American Thoracic Society Clinical Practice Guideline. Am J Resp Crit Care 200, 535–550 (2019).
• LITFL
• Deranged Physiology
• Fekkar, A. et al. Occurrence of Invasive Pulmonary Fungal Infections in Patients with Severe COVID-19 Admitted to the ICU. Am J Resp Crit Care 203, 307–317 (2021).
• Verweij, P. E. et al. Taskforce report on the diagnosis and clinical management of COVID-19 associated pulmonary aspergillosis. Intens Care Med 1–16 (2021) doi:10.1007/s00134-021-06449-4.

 

Welcome back to the tasty morsels of critical care podcast.

This time round we’ll look at an oldie but a goodie: salicylate poisoning. I have not seen one of these in quite some time but it is a classic tox question for exams in both EM and ICM.

Oh Chapter 90 has the ambitious task of covering all poisonings so unsurprisingly it’s a little brief but this post is supplemented by a few other excellent resources linked to at the end.

Salicylates are primarily found in our part of the world in aspirin. Locally the commonest use for aspirin these days is primary or secondary prevention of vascular disease, ie the baby aspirin tablets that come in 75mg. You would need to take a large number of these to get into trouble. The analgesic doses of nearer to 600mg are less commonly used, especially when compared to the ubiquitous paracetamol, but 15-20 of these big aspirins could get you into big trouble. It does exist in other forms, most notably in “oil of wintergreen” which, in kids can be a potentially fatal ingestion at low volume.

Like most ingestions, the context or the patient  will often be the give away to the diagnosis. But if they haven’t told you directly, you might start by asking questions about tinnitus, dizziness and vomiting. On exam you might find fever, tachypnoea and even impaired consciousness as things get more advanced.

These clinical signs can be explained by looking at the pathophysiology. Aspirin, being salicylic acid is by nature an acid, one would think that this is the reason you get the metabolic acidosis. In overdose it does indeed form part of the anion gap of unmeasured anions along with lactate. But in reality the salicylate apparently contributes only a small amount of the gap here and other unmeasured anions like lactate and ketones form most of the gap.  The metabolic acidosis induces an appropriate kussmaul like response observed in the tachypnoea. Minute ventilation is increased to lower CO2 as a “compensation” for the metabolic acidosis. More interestingly aspirin has a direct effect on the brainstem causing outflow to the respiratory centres to increase resulting an additional increase in minute volume beyond that appropriate to the metabolic acidosis. As a result you get the classic blood gas of someone with a mixed metabolic acidosis and respiratory alkalosis with a CO2 lower than that expected with something like Winter’s formula or perhaps a normal pH (remember respiratory compensation for acidosis should not correct so much as to normalise the pH). In general the pH in these patients will be normal or high and indeed if an acidaemia develops you’re really in trouble.

The tinnitus and dizziness is thought to be a direct effect on vestibulocochlear centres inducing the symptoms. The fever is likely related to the uncoupling of oxidative phsophorylation and possibly on hypothalamic set points.

Aspirin has multiple potential mechanisms of pathology that could potentially lead to death.

1. probably the biggest is uncoupling of oxidative phosphorylation, something so key to life that pretty much all aerobic life depends on it. High lactates are probably the best measure for this
2. penetration of the CNS with salicylates, leading to the usual tox spiral of seizures, coma, death
3. promotion of fatty acid metabolism creating severe ketosis and contributing to hypoglycaemia

The non ionised form of aspirin causes all the nastiness and the dissociation between ionised and non ionised is highly pH dependant. An aspirin level of 400 at pH 7.4 might be tolerable but the same level at a pH of 7.2 is likely to be rapidly lethal. These 2 components, the aspirin and the blood pH form the subtleties of management at this stage.

Levels are easily obtainable from every lab I’ve ever worked at. This gives you a total salicylate. It does not tell you the unionised salicylate which determines the badness as mentioned above. Hence, overall levels on their own are poorly predictive of mortality. Levels >500mg/L are definitely where you should be worrying and thinking about dialysis (note that outside UK and Ireland levels may well be in mg/dl rather than mg/L so know what you’re looking at!) The absorption of salicylates is also variable, unlike paracetamol where we have a fairly robust curve for dertermining treatment threshold. Salicylates can be tricky and repeat levels every few hours are usually recommended to see if there is ongoing absorption from the gut and indeed may be a reason to use repeat doses of charcoal.

Management follows the usual tox paradigm (RESUS-RSI-DEAD is a good mnemonic here) of ABCs first then think about risk assessment, antidotes, enhanced elimination and supportive care.

Alkalinisation has been the mainstay of treatment for years. This probably works by two ways 1) reducing penetration into the CNS 2) enhancing elimination through the kidneys. There are a variety of ways to achieve this but typically this will be through short infusions of the typical 50-100ml 8.4% bicarb amps stocked on resus trolleys or more prolonged infusions of the “isotonic bicarb” made by mixing 150mls of 8.4% in 850mls of DW5. Either way you’re looking for the urine pH to be >7.5

Toxbase in the UK describes the indications for HD as follows

• level >900
• level >700 with acidaemia
• coma with salicylates

(Of note the EXTRIP group have their own published guidance on this)

As with all tox cases, IHD would be better at clearance than CRRT but in reality the only modality available when you need it will be CRRT. Filtration is poor at clearance so this is the situation where you want to use CVVHDF with a larger exchange than usual.

Given the importance of maintaining normal or alkalotic pH in these patients, it can be problematic to pursue intubation as the mechanical ventilator will complain noisily and typically fail at maintaining a minute volume of 20 l/min. The inevitable drop in minute volume following your sedative and paralytic of choice will drop the pH dramatically unleashing salicylates to do their devilry. In fact describing it as problematic is understating what is more likely to be a homicidal move unless really needed. “really needed”, is not exactly that well defined here but should be done with extreme caution.

Finally as an interesting side note of little consequence, the INR is typically raised in those with salicylate poisoning. This is apparently linked to a warfarin like effect on part of the vitamin K cycle. This can be corrected with vitamin K if needed.

Reading:

LITFL

IBCC

Oh’s Manual Chapter 90

Tox Handbook 2nd edition

UK Toxbase Gudiance

Welcome back to the tasty morsels of critical care podcast.

This time round we’re going to have a look at some chest wall injuries you should know about. The main reference here is Oh’s manual chapter 79.

The vast majority of what we see here is going to be simple pneumothoraces and the elderly patient with some rib fractures and contusions or a developing pneumonia. That kind of thing is our bread and butter. This post will focus on some of the more esoteric injuries which of course occur with disproportionate frequency in fellowship examinations.

There are a fairly small number of immediately life threatening injuries we need to recognise and the list could include:

• tension PTX
• open/sucking PTX
• massive haemothorax
• pericardial tamponade

Massive haemothorax is typically defined as >1500mls immediately or more than 200ml/hr is certainly a concern that should prompt a surgeon to have a look inside. While not mentioned in Oh, the main concern with these is a sort of “damned if you do, damned if you don’t” scenario. When presented with a massive haemothorax and hypotension, it is not always immediately clear what the primary physiology causing the hypotension is. For example a large haemothorax with tension physiology will kink the SVC and obstruct the IVC leading to hypotension due to low preload to the heart. They may also be hypotensive form frank hypovolaemia because all the blood is in the pleural cavity instead of the blood vessels. The bit you can’t account for is how much this tension phenomenon is actually providing some kind of tamponade effect and keeping the remaining intravascular volume in the vasculature. The concern here is that when you decompress the haemothorax the patient is no less hypovolaemic than they were before. The blood is now in the chest drain rather than the pleural space. This hasn’t really fixed the hypovolaemia but has relieved the tension phenomenon obstructing the preload to the heart. Unfortunately it may have also unleashed the remaining circulating volume to enter the pleural cavity and swiftly out through the plastic conduit you’ve placed and into the chest drain.

All this is a very long and convoluted way to say that it’s complicated. I think we will always end up draining that massive haemothorax but it would be wise to have someone capable of dealing with major bleeding inside the chest, immediately on hand.

Speaking of thoracotomies, what follows is a list of interventions that might be potentially useful to do once the chest is open.

• drain pericardial tamponade, this is 1st, 2nd and 3rd for me in terms of importance and utility.
• control intrathoracic bleeding – which is a nice coverall term for all the various bits blood could be squirting out of
• control of massive broncho venous embolism. In this scenario a pulmonary vein is lacerated and air is being entrained into the left side of the heart. This is bad form as one might imagine so it would be wise to clamp it
• control of massive bronchopleural fistula. Maybe a lung has been avulsed proximally and you can see the ET tube through the bronchus. All the Vt is disappearing into the pleural space and you should something to stop it
• temporary blocking of the aorta. Commonly done in an attempt to preserve the circulating volume to the heart and brain. You might better achieve this without opening the chest with a REBOA or a SAAP catheter but that’s a whole different kettle of fish
• internal cardiac massage.

In terms of aortic injuries these are often fatal pre-hospital but if you do find one they’ll typically be at the junction of the fixed and tethered aorta and the slightly more mobile arch. This junction occurs at the isthmus  just distal to the take off of the left subclavian. Your cardiac surgeons will likely decide but this may be an open repair or some kind of TEVAR type stenting if they survive long enough.

As mentioned briefly above, tracheobronchial injuries can be a real challenge. The classic region of injury in blunt trauma is at the take off point of the right main possibly because of the steep angle from the trachea. Expect to see a PTX and some mediastinal emphysema. In itself, that might not be a big issue but the real clincher to the diagnosis is the massive PTX and emphysema that occurs when they get intubated and transition to positive pressure ventilation. The flexible bronch is your friend here and allows you to confirm diagnosis as a quick look down the tube will let you see mediastinum and pleura through the bronchus.

Systemic air embolism is more typical in penetrating than blunt injuries and again the problems really begin when you move to positive pressure ventilation. In this instance we’re talking about pulm vein or maybe SVC injury. In negative pressure ventilation the pressure in the pleural space is lower than in the vasculature so blood will flow into the pleural space which is a problem in itself, however this pressure gradient will ensure air is not entering the circulation. Once intubated and in positive pressure ventilation then the gradient reverses and air can enter the vasculature with disastrous haemodynamic and even neurologic consequences if it enters the left sided circulation. A few rescue moves might be to selectively intubate the good lung, get them spontaneously breathing by reversing the rocuronium and get them on 100% O2 in the hope of switching out the non absorbable nitrogen for oxygen.  Thoracotomy is likely the next step but again a decision for the surgeons.

There are of course other chest injuries out there but I suspect that’s plenty of exam worthy minutiae for today.

Reading

Oh’s Manual Chapter 79

 

Welcome back to the tasty morsels of critical care podcast.

Usually the topics here follow the well trodden path of Oh’s manual, but we’re looking at something primarily because it is  an ideal question for a fellowship exam. In this type of scenario you will almost be guaranteed to find that Deranged Physiology has already covered it comprehensively and in great detail. So as expected the DP post forms the bulk of the material for this short audio snippet.

First thing to consider is how long do you expect your filter to run for in the first place. This will likely vary between hospitals and more signfiicantly between different types and brands of filter. A common time to empirically take down the filter is around 72 hrs or so. At that point you’re at end of life as far as the manufacturer is concerned. All the microfibres are clogged with all the gunk your kidneys usually deal with and the filter isn’t going to work. Think about it like your vacuum cleaner bag getting full and needing a change.

I’m not sure a clear definition of recurrent clotting exists but I would usually consider it to be recurrent clotting when the nurses tell me it is, given that they spend all the time running the machines they, as always typically know best.

Once we know there’s a problem then it’s time to start thinking of some potential causes.

1) is the anticoagulation being done poorly?

There can be quite a lot to this really. Sometimes it can be as simple as just not following the citrate or the heprain protocols as you should. That’s a fairly simple “just read the instructions” type problem. However, particularly when it comes to heparin the tests we use to judge the level of anticoagulation are fairly blunt instruments in explaining the intricacies of the finely tuned coagulation system. Heparin resistance might be a problem or any number of other myriad of causes. I have linked to the excleent IBCC post looking at anticoagulation that I return to frequently. You might even be in the enviable position of being enlightened and use anti-Xa levels to monitor your anticoagulation.

Given all the issues with heparin I think the most appropriate answer to the question of “how to antioagulate your filter ciruit with heprain” is to use citrate instead.

2) Is your filter clotting because the flow rates are too poor?

There are a lot of reasons for poor flow rates, the catheter itself being a major contributor. That being said most units have only one type of catheter available so you usually don’t have much flexibility here. You will likely have different lengths of the same gauge of catheter available and remember that pouseille’s law might get the better of you with the flow being inversely proportional to the length of the catheter. In other words the longer the catheter the worse the flow rates.

You probably can optimise position. I don’t mean to pretend that these are definitively settled issues but RIJ is often favourable due to direct access to the central circulation and minimal impairment with mobilisation. However, it is interesting to note that the shiny, ivory tower of the Alfred in Melbourne have the femoral vein as their location of choice with some local obseravtional data supporting extended filter life.

Once you’ve decided on site you should consider tip placement. A brief review of a CXR of someone with a permacath, a tunnelled long term device, usually shows that the tip is probably in the RA. Typically I was taught that we shouldn’t have the tip in the RA and instead aim for that RA/SVC junction. The issue with the RA/SVC junction is that often one of the ports lies against the wall of the vessel and can obstruct and suck down and cause issues with flow. Again, hardly definitive but there is another small RCT of positioning in the RA that would suggest this is a safe and effective practice.

Finally, at least in this entirely non comprehensive list, we often worry about “filling” or “volume status” in inverted commas as an issue for flow rates. If the SVC is collapsing even with simple respiration then pulling off 150-200ml/min might be a challenge. Often a judicious fluid bolus can help. My issue with this is it can be fairly difficult to know if this is the problem without just giving it and seeing what happens

3) Is your patient in some kind of prothrombotic state?

The answer will usually be no. At least not in any more of a prothrombotic state than the usual ICU patients. But could this be a HIT, could there be something weird and wonderful underlying or is it just COVID like everyone else.

4) is viscosity an issue?

By viscosity I don’t really think they have waldenstrom’s macroglobulinaemia but you more like you not be diluting enough pre-filter . This should be taken care of by your protocols but the pre filter dilution is likely helpful in reducing clotting.

Fixing the clotting issue is obviously directed at the causes outlined above. But given the difficulty in diagnosing the problem in many of these cases I find myself addressing multiple different causes and switching anticoagulation and changing lines for a few days until either the patient gets better or the need for CRRT goes away. Honestly, it’s remarkable that they let me do this job at all.

References:

DP article is a must

Intensive blog by Vui Kian Ho on mastering the vascath

Morgan D, Ho K, Murray C, Davies H, Louw J. A randomized trial of catheters of different lengths to achieve right atrium versus superior vena cava placement for continuous renal replacement therapy. Am J Kidney Dis. 2012 Aug;60(2):272-9. doi: 10.1053/j.ajkd.2012.01.021. Epub 2012 Apr 11. PMID: 22497790

 

Welcome back to the tasty morsels of critical care podcast.

Last time i was butchering my way through a diagnostic approach to hyponatraemia, particularly the forms likely to end up in the critical care end of the hospital. This time we’ll take a punt at how you might approach management. In an ideal world of course you would have all of the diagnostic tests back and you’ve been able to make a very solid diagnosis of the cause of hyponatraemia and you would institute a bespoke treatment course for the underlying disease and the resultant hyponatraemia. But as we all know in critical care we often work with less than ideal information and have to begin treatment while the diagnostic process is ongoing. Hopefully what follows will provide enough broad brush strokes to get you through a night on call or even worse a viva.

We’ll start with truly emergent situations. Older person presents to the ED after being unwell for several weeks. They have a seizure on arrival and a Na comes back at 105. This is a fairly solid indication to give hypertonic saline. In this scenario they are seizing because of the low Na and rapid increase of the Na is needed to stop the seizure. The European Hyponatraemia Guidelines would suggest 150mls of 3% saline over 20 mins aiming for a rise in the Na of 5mmol/L. This bit is usually pretty straightforward. The sodium rises, the patient stops seizing everyone relaxes but then the Na continues to rise, well above the 5mmol we wanted and a panic ensues.

The guidelines suggest a max rise of 10mmol in the first 24 hrs and 8 mmol/day after that. It is hard to overemphasise how easy it is to blow past that target unless you are paying attention. So how do you control the rise in the Na? If it’s rising too quick it’s often because the patient is losing lots of water through the kidneys which concentrates the plasma raising the Na in the blood. You can replace that water loss by giving a decent bolus of free water in the form of something like 5% dextrose. An alternative method involves using the wonderfully named DDAVP clamp. In this scenario you’re using the DDAVP to tell the kidneys to excrete less water therefore limiting the rise of the Na. I have not seen particularly strong data on one method vs the other for limiting the rise and indeed I have seen clinicians use either or indeed both to good effect.

The European guidelines do use the phrase “severe symptoms” as an indication for a bolus of hypertonic. Unfortunately it’s a little less clear what constitutes severe symptoms. A seizure seems fairly easy to define but “coma” is a little bit more vague.  The guidelines are clear that you have to be able to put the symptoms down to the hyponatraemia and not some other cause. But as we all know patients often have multiple reasons to be obtunded including sepsis or intoxication or multiple other causes. As such the decision to give hypertonic can be a little subjective and fudgeable.

For many patients the best thing you can do is very little. A former consultant I worked for had somewhat facetious plans to start a hyponatraemia clinic that involved locking the patient in a room and denying them access to water and letting the body sort it out over several days. There is an element of truth to that as for many of the hyponatraemics simple fluid restriction and time will correct things.

Lastly, our hypertonic of choice is typically 3% saline with an osmolality somewhere in the range of 1000 or so. Typically we’re a bit reticent to give such concentrated solutions through a peripheral IV but there are a few papers suggesting that this is fine at least on a limited basis. I will say that once the hypertonic is in and you’re reaching for a 2nd or a 3rd you should probably be thinking about a CVC as the access for administration and indeed regular sampling is really helpful.

Reading

European Hyponatraemia Guidelines

Oh Chapter 95

Khasiyev, F., Hakoun, A., Christopher, K., Braun, J. & Wang, F. Safety and Effect on Intracranial Pressure of 3% Hypertonic Saline Bolus Via Peripheral Intravenous Catheter for Neurological Emergencies. Neurocritical Care 1–6 (2024) doi:10.1007/s12028-024-01941-3.

Welcome back to the tasty morsels of critical care podcast.

There’s not a huge amount of notes on procedural stuff that I accumulated for the exams but I did collect some interesting bits on bronchoscopy, particularly because it was so novel to me as an EM trainee who really had no experience with bronchoscopy prior to starting to my critical care fellowship.

After 2 years of frequent, if not daily use of the bronch, I find it hard to see how I would manage in an ICU without it. Now, when I say bronch I want to be clear that I’m talking about the flexible, disposable pieces of gear. The respiratory people sometimes come and laugh at us when they arrive with their big stacks and multi-thousand euro pieces of kit.

Why might we pull out the bronch in the ICU?

Well I’m not sure there is a clear or authoritative list, but the following would be reasonable

• as an assistance in intubation
• as a guide in percutaneous tracheostomy
• as a therapeutic in mucous plugging
• sudden refractory hypoxia without obvious cause. I often find putting the bronch down quickly answers a lot of questions about tube position and mucous plugging and bleeding a whole lot quicker than waiting for the CXR
• BAL. We all love to know what we’re treating and especially in the hospital acquired or  immunocompromised, it can be lovely to grow a bug. Though I don’t want to suggest this is particularly data driven.
• Airway bleeding. Finding the bleeding, clearing it, treating it.

In general we will almost always be passing the bronch through some kind of device. I have on occasion post tracheostomy spent some time practicsing getting the bronch through the cords in a sort of a poor man’s fibre optic intubation, and damn if it’s not tricky… Credit should be due to the respiratory guys who do this all the time. All be it with much weller patients and nicer and shiner equipment.

What do we look for then once we’ve passed a bronch into the large airways? We should have a good look at the airways themselves. Is there a lot of suction trauma? Is there bronchomalcia, (esp in the slow trache weans), are there hyphae and mushrooms growing in the airways suggestive of aspergilllus? We often forget to simply look at the walls as we chase down into the lobes to get our sample.

Once it comes to the process of BAL itself then I realise I’m often a little shy and cautious in my installations of saline. When I went and read up on this it turns out typically recommended volumes for installation of saline are more like 100mls rather than the paltry 20-40mls I was using when I first started. A lot of what I (and I suspect many of you) have been doing is perhaps better classified as bronchial washings rather than true bronch alveolar lavage. Also important, once you stick the saline in, to give it a decent amount of time (30-60 secs) before aspirating.

In terms of where we are aiming for, typically aim for the segment that looks worst on imaging. If it’s all awful or nil focal then the RML and the lingula are often recommended.

In some of the reading there was discussion on “wedging” the bronch which is not the infantile practice of pulling someone’s underwear up really high when they’re not suspecting it. Bronchoscopic wedging involves getting the tip of the bronch into a segment far enough where there is partial airway collapse (though not complete) on suction. If it doesn’t collapse at all you’re too proximal, If it collapses completely then you’ll likely not get a good aspirate.

Once you’ve got your samples then the qunadry you face next is what to test it for. This will likely depend on the situation, but you have a variety of options available to you

• routine culture
• cytology, things like eosonophils for daptomycin induced pneumonitis, inclusion bodies for things like CMV/HSV
• galactomannan on a BAL sample apparently more useful for aspergillus than serum
• PJP PCR (very sensitive, not specific) and silver staining is a useful test
• CMV reactivation is another common concern in the critically ill

Bronchs are not without their potential complications and can be summarised as follows

• bleeding
• barotrauma – this may be related to the sudden and extreme pressure changes from being on +ve pressure one moment and attached to wall suction the next. The vent settings itself will also change things as if you really ask the vent to deliver 400mls while the airways are largely occluded with a plastic bronch then expect airway pressures to be very high

References

• the UTD article was useful on this but you will need access
• LITFL covers it well as expected
• there is a nice simulator but it needs flash and it seems flash has not graduated into the pandemic era so it’s hard to run
• finally there’s a nice and brief anatomy review that is worth reviewing on a regular basis so you actually have some idea where the hell you are in the lungs at any given time.

 

Welcome back to the tasty morsels of critical care podcast.

Sometimes the tasty morsels are exam sized snippets of my knowledge on a given topic. More frequently they are literally all I know on the subject. Today’s topic of parenteral nutrition is a good example of this. I have no doubt there is a lot more I should know about what goes into those plastic shrouded bags at the end of the bed but I am afraid I do not.

Firstly, indications. From our point the overall message should be that we use PN when we fail to establish enteral nutrition. The definition of when we have failed to establish EN is a little less well defined, but comes in somewhere around the 5-7 day mark. In real life clinical practice we often start much earlier , particularly when the surgeons feel that the gut, or perhaps more accurately, the anastamosis mightn’t handle it.

In terms of complications, the maintouted one has been infection, with a long standing belief that PN increases the risk of sepsis. This is probably not true. However the need for central access on an ongoing basis does certainly increase the risk of line related infections but it does seem unfair to blame the PN for that.

The bogey man of refeeding syndrome rears its head in particular in relation to PN as the insulin that has lain dormant for so long in the starved patient finally comes out to play and cause havoc as the body moves from catabloism to anabolism. Again, this may not be due to PN specifically and maybe more reflects the fact that the calories are actually delivered in PN as opposed to calories being aspirated up the NG every 4-6 hrs in enteral nutrition.

There is a NAGMA or hyperchloraemic metabolic acidosis described that is associated with PN and may be caused by the amino acid solutions it contains. One to add to the list of NAGMA causes for me.

Hypercapnoea, causing a resp acidosis and hyperventilation is described as “overfeeding” in PN and I think I have seen this once but more commonly this tends to crop up on the data interpretaton section of any fellowship exam. If too many calories (eg >4mg/kg/min of glucose in the septic patient) are delivered then the body will covert these to lipids. This is a CO2 generating process and may contribute to difficulty weaning or tachypnoea driven by a high CO2.

LFTs are frequently monitored and are frequently abnormal and are frequently ignored. That being said there are a variety of potential issues with PN and LFTs, in particular, hepatic steatosis, intrahepatic cholestasis, and biliary sludge are all an issue. Unfortunately most of these are already common in the critically ill and would be hard to tease out which is which.

In terms of what is actually in the bag, while you can get pre made bags, in reality the composition is determined by the super clever dieticians in the units I’ve worked in.

Calories can come from any of the components (eg carbs or lipids) but the energy split is typically 70:30 carb:lipid with the carbs being generally 50% dextrose. Lipids are a more concentrated way of providing energy and can avoid some of the hyperglycaemia, though there is apparently concern for an immunosuppression effect. As I’m sure you already know the lipids are typically provided from intralipid – the magic white stuff we use when we give too much local anaesthetic. The required nitrogen comes from l-amino acids solutions. It will also include some tasty vitamins (though thiamine, folate and vit k are particuarly prone to depletion)

There are a number of important nutrition trials that occurred early in my career before I was really paying attention. LITFL has a good summary but my TLDR is that there is insufficient evidence to say PN is better than EN for now we go enterally when we can and parenterally when we can’t.

Reading:

Oh’s Manual Chapter 96

LITFL

From Twitter, after publishing, Lisa, one of my excellent dietician colleagues, added a grerat summary of the guidelines that you can access here.

Welcome back to the tasty morsels of critical care podcast. This is number 50, so for all 7 of you out there, well done for making it this far especially when you can’t even get CPD points for it.

Today we’ll look at Oh’s Manual Chapter 80 written by the one and only Oli Flower of SMACC and CODA fame.

Like TBI we can split up spinal cord injury into primary and secondary injury. Primary injuries include direct mechanical injury like compression, haematoma, laceration, traction or even complete transection which is thankfully rare.

Secondary injuries include local ischaemia that begins at site but extends progressively in both directions (ie the cord level of injury can get worse). There is loss of autoregulation of blood supply and lots of inflammatory stuff. In addition there is often bleeding into the cord with oedema.

The assessment of SCI is driven by the ASIA score which is a systematic severity assessment tool that includes pictures and tells you what all the dermatomes and myotomes are so you don’t need to actually carry them all in your brain. It is a useful and at this stage well validated tool for motor prognosis that forms the cornerstone of assessing SCI. It spits out a grade A to E which is unhelpfully the opposite of what you want in your A Levels as a grade A is a complete injury with very low chance of recovery.  B is described as sensory incomplete, which again is confusingly named as it suggests that there is an incomplete sensory injury but in reality it means a severe motor injury with preservation of sensory function below the level of injury.  C and D are varying degrees of motor preservation below the injury and grade E is normal

A further key point to help us speak the language of the spinal surgeons is that of neurological level of injury. The Neurological level of injury = most caudal segment with normal sensory and antigravity (ie 3+ ) motor function. Remember that the neurological level does not usually equal the radiological level as the spinal cord is much shorter than the spinal column.

There are a variety of cord syndromes described that are certainly exam worthy and worth knowing about. The central cord syndrome consists of

• weakness and sensory loss in the arms>legs,
• the useful mnemonic is MUD (motor, upper, distal)
• think hyperextension in a grotty neck with pre existing arthritis
• the pathophys here is central ischaemia/haematoma

The anterior cord syndrome looks like

• loss of motor, pain, temp below injury.
• can also seen in aortic pathology

The brown-seqard syndrome is more notable for teaching anatomy than it is for clinical practice but for completeness look for

• ipsilateral loss of motor, proprioception and find touch
• contralateral pain and temperature loss

Diagnosis and imaging these days is all about CT and MRI. There is still a robust literature in well done plain films for exclusion of c-spine injury in the lower risk patients, but by now I think everyone has just moved to CT. The controversy in ICU at this stage is whether a CT is sufficient. Oh cites a 4% miss rate for CT, ie injuries not seen on CT that will show up on MRI but more importantly only 0.3% actually need an intervention. CT remains better for bones but MRI is brilliant for cord and ligaments.

In general, from what I have seen, if the patient has a neck but is unconscious, then they end up getting an MRI, generally several days after the original injury and normal CT c spine. This leads to a prolonged and somewhat hard to quantify decrement in the patient’s care as maintaining spinal precautions is challenging in the ICU patient. The EAST group makes a conditional recommendation for clearance of the C-spine following a normal CT c-spine but this doesn’t seem to have made it across the Atlantic to our orthopaedic teams even if it seems our local neurosurgeons have embraced the idea. Bottom line is that this remains a topic of controversy where there is a very small (but not zero) risk of removing the collar without the MRI.

One thing to remember when ordering imaging is vertebral artery injury. While not routine you should have a low threshold for adding an angio when there is a fracture of the transverse foramen, base of skull, facet joint, or preexisiting spinal disease. It is not good form to wake them up two days later to find they’ve stroked from a dissection.

Moving onto management. An incomplete injury with compression on imaging needs urgent decompression. The term “urgent” is somewhat ill defined in this statement but it doesn’t seem to be a 3am urgent in my experience.

Managing an SCI patient in the ICU is best broken up by systems. At least it is for exam purposes.

From a respiratory point of view if it is a high c spine injury, then expect loss of the diaphragm and apnoea, If the diaphragm is preserved then there will still be loss of cough. Somewhat paradoxically, they generally breath better flat rather than upright. When supine the abdominal contents push the diaphragm up giving it mechanical advantage for respiration. Expect to see reduced vital capacity and FRC with atelectasis, reduced chest wall compliance with spasticity of chest wall muscles. There is often wheeze due to sympathetic loss (ie the beta outflow) and 75% of cervical injuries will need intubation. Ventilation is often done with higher Vt (eg 10/kg) to prevent atelectasis and allow lower PEEP (again, giving the diaphragm the advantage). If they are extubatable then extubation to NIV is reasonable. If you have a complete injury above C5 almost all need trache so perhaps best to crack on and do it early. Expect a prolonged wean.

When it comes to haemodynamics, the loss of sympathetic outflow can cause profound vasoplegia, hypotension and bradycardia. Asystole is common with simple suction in first week and then resolves with time rather than a pacemaker typically. Keeping the MAP>85 might help and has some low level data in support and has a guideline recommendation for 7 days post injury. This can often lead to a request for a critical care bed for an otherwise well patient who needs some noradrenaline and an art line to meet the MAP targets. Neurogenic shock is of course a real thing in the immediate post injury phase but as always in the immediate post injury phase you should be thinking bleeding first second and third before considering that as a diagnosis.

Autonomic dysreflexia describes hypertension due to symapthetic stimulation in pts with injuries above T6. It only occurs after the initial period has settled.  There is sympathetic vasoconstriction below level of the lesion, and parasympathetic reaction above the lesion (brady, sweating and vasodilation). Bladder and bowel stimuli are the main culprits but remember ureteric stones can also do it. Treatment involves sitting them up and using agents like GTN and captopril. IV SNP andGTN can be used if unresponsive.

Finally, as I’ve gone on too long already, a note on outcomes. Hospital survival is 90%. The clinical exam (ie the ASIA score, even on day 1)  rather than things like MRI, remains the best predictor of outcome. There is a 7% conversion from ASIA A (complete) to B (some sensation) at one year with none progressing to C (ie an ASIA A will not get motor recovery). That is indeed a bit miserable but on the brighter side 50% of ASIA B convert to C or D within a year. Even improving a level of injury (ie from a C8 to a C7) will lead to a significant improvement in hand function and ultimately quality of life.

Reading

Oh’s Manual Chapter 80

Anatomy for emergency medicine series on the spinal cord

• Part 1
• Part 2
• Part 3

Eastern Association for Surgery for Trauma 2015 systematic review

Welcome back to the tasty morsels of critical care podcast.

Today we’ll look at everyone’s favourite yeast – Candida. Firstly, remember the distinction between yeasts and moulds. Yeasts, like Candida species are single celled critters whereas moulds like aspergillus are multicellular.

Candidiasis is common in ICU patients. It grows commonly in our BAL samples, it grows in our urine samples. It’s often the only +ve micro we turn up because the patients are mostly picked in meropenem and linezolid and the candida is all that is left or becomes selected out. It’s important to note that these BAL and urine samples of candida are common and generally reflect colonisation rather than infection and in general they shouldn’t be treated. That being said knowing what’s there in the event of the patient getting super sick might be somewhat helpful in guiding therapy.

the LITFL post describes candida as invasive if:

• grown in blood cultures
• found in a sterile cavity
• cultured from 2 non contiguous sites (so for example you could argue that candida in sputum and urine fills this criteria though in reality I find we rarely treat this type of growth unless they’re super sick)
• the species is a non commensal form of candida
• culture from wound or burn biopsy

Risk factors for invasive candidiasis can be summed up as “being proper sick” but they can be somewhat more nuanced to include things like invasive lines, some kind of intra abdominal perforation or major burns or some horrible immunosuppression.

We make the diagnosis here using a variety of tools. The easiest being when it pops up in blood cultures on a sick patient or commonly it can grow from drain fluid in some abdominal disaster, particularly the higher up the GI tract the disaster occurs. Having a tissue diagnosis with a named species with sensitivities makes treatment fairly straightforward.

There are a variety of named scoring systems that can help you estimate risk of an invasive candidal infection that rejoice in the names like the “Nebraska criteria”, the “Candida score” and the “Paphitou” score. However I don’t see anyone using them in clinical practice.

More commonly we’re likely to be faced with a deteriorating septic patient with RF for candidiasis and we empirically start an antifungal. The issue now comes with stopping or supporting the ongoing use of the antifungal. Here tests like beta-d-glucan probably have a role. BDG is a cell wall component of all fungi so it certainly is not specific for candida but if it’s through the roof in a sick patient then coverage with an antifungal is probably reasonable. On the other hand if it’s undetectable and the indication for the antifungal was a bit soft in the first place then maybe it’s time to stop. The sensitivity is somewhere in the 75-85% range so I would not use it to rule out the super sick, super high risk patient.

When it comes to treatment we have to consider source control. If it’s coming from a belly full of collections then you need to drain the collections. Once source control is out of the way we now face a variety of choices. Ultimately if we have grown a candida then we will eventually get sensitivities and we can tailor therapy appropriately but all that takes time and we need to start something before then.

In general we turn to the echinocandins, drugs like anidulafungin and caspofungin. These are work horses in the ICU with a low side effect profile and importantly cover the vast majority of candida likely to grow. They also have nice action against biofilm embedded bugs which is very handy given the number of line related candida infections we might see. Fluconazole is of course an option but it struggles when it’s something other than candida albicans hence I see it more often used as a deescalation agent. The other agents like amphotericin and voriconazole will work in general for candida but are generally not needed.

In terms of guidelines, the 2012 European microbiology guidelines have Caspo or Anidulafungin as first line for documented invasive candidiasis. The IDSA in 2016 suggests fluconazole in stable patients and then caspo or anidulafungin if unstable.

Finally there is often great anxiety and hand wringing regarding the eye balls when someone grows a candida in the blood. Every time we grow candida in the blood we need to twist the arm of some poor eye doctor to come and do an exam for us. Indeed candida end ophthalmitis is a well described and devastating metastatic complication of candida but the eye balls have the useful characteristic of being easily accessible to clinical examination. If the eyes are red and nasty then be worried. If not then maybe you’re OK. This is supported by a 2019 JAMA Ophthalmology SRMA that suggested that a routine exam was useful in 3/7000 patients and made the gentle suggestion that maybe the IDSA should rethink their guidance please and thank you.

If you do happen to find candida in the eye ball then you need to rethink your antifungal strategy as our beloved echinocandins just don’t penetrate that well and you should be reaching for fluconazole, or flucytosine or indeed some kind of needle injection of stuff into the eye.

Reading:

Oh’s Manual 73

LITFL.com

Breazzano MP, Day HR Jr, Bloch KC, Tanaka S, Cherney EF, Sternberg P Jr, Donahue SP, Bond JB 3rd. Utility of Ophthalmologic Screening for Patients With Candida Bloodstream Infections: A Systematic Review. JAMA Ophthalmol. 2019 Jun 1;137(6):698-710. doi: 10.1001/jamaophthalmol.2019.0733. PMID: 30998819.

IDSA 2016 Guidance

And I missed it in the prep but of course IBCC has an excellent post that covers a lot more than I squeezed in here and is honestly, compulsory reading

Welcome back to the tasty morsels of critical care podcast.

Today we look at anaphylaxis. Oh’s Manual 67 forms the basis for most of this. In many ways this is fairly straightforward. You give adrenaline and they get better. However once you get into the pathophys behind it you start to realise why immunologists are much smarter than you are.

Oh starts with a few nice descriptions of what it is: an acute multi system disorder with release of multiple mediators from mast cells and basophils. Usually this is a hypersensitivity reaction mediated by IgE to a foreign substance. The term anaphylactoid is reserved for non IgE mediated reactions. (These still involve mast cell degranulation)

The current guideline accepted definition goes like this

serious, life threatening generalised or systemic hypersensitivity reaction that may or may not be immune (ie IgE) mediated

While accurate this doesn’t really help me very much in knowing what it looks like.

A better description for what it looks like goes as follows:

Anaphylaxis is highly likely in the following scenarios

1. acute onset with skin or mucosal involvement plus at least one of respiratory compromise or reduced BP.
   • Rash and shock or rash and SOB means anaphylaxis
   • the term mucosal involvement can mean the tongue and pharynx but also the rest of the GI tract. Let’s say you eat a peanut get tummy cramps, diarroeah and a BP of 60 systolic. No wheeze, no rash – this is still anaphylaxis.
2. reduced BP after exposure to a known allergen for that patient.

The pathophysiology is mostly beyond me but the predominant features are mainly histamine mediated and there is release of serotonin and tryptase in addition.

The precipitants can be split up into a number of classes

• Food
  • peanuts, crustaceans and fish, sometimes persisting into adulthood.
  • Interestingly there is there is food associated, exercise induced anaphylaxis where exercise 2-3 hrs post the allergen causes the reaction
•  Drugs are incredibly common.
  • penicillins and cephalosoporins are the commonest antibiotics
  • NMBs such as rocuronium are an important bogey man in anaesthesia and may occur without prior exposure. If you only ever use it in the critically ill then it’ll be a while before you pick up a serious case of it appears to me.
  • NSAIDs, Aspirin
  • Radiology contrast
  • Protamine
  • Chlorhexidine is a common and often overlooked cause
• Rare beasts like mast cell disorders like systemic mastocytosis

Tryptase gets a lot of attention (not without reason) in both anaesthesia and EM circles. Peak levels occur ~15-120 mins post event and decline predictably (T1/2 ~2hrs). This influences the timing of sampling as recommended in the UK guidelines with one sample taken as soon as possible and another within 4 hrs and then a 3rd convalescent sample either >24 hrs post resolution or at allergy clinic. It is critical to understand that this is a test useful in follow up to confirm diagnosis and should not be involved in the decision whether or not to give adrenaline acutely.

Treatment is adrenaline. The ALS guidance is all about the IM and this should continue to be the case though there is a caveat for IV adrenaline use by “experts” in quotation marks. Oh is quite pro IV infusions and certainly in the ICU and OT this is ubiquitous and generally well done in my experience. Oh mentions vasopressin, glucagon and even methylene blue as rescue treatments but UK ALS only recommends glucagon for those beta blocked and vaso for those refractory to IV adrenaline.

Another interesting point from the UK guidance is an emphasis on lying the patient flat, reflecting the rapid and massive vasodilation that might be enough to precipitate LOC in some. They are clear in that medications such as antihistamines and steroids have no real role in the routine management of anaphylaxis despite their almost ubiquitous use.

Finally they quote a 95% survival rate for anaphylaxis associated cardiac arrest and while this may be contaminated by an in hospital OT sample it’s still important to know that these patients are much more likely to have a good outcome than your typical cardiac arrest cohort.

References

Oh’s Manual 67

RCEM Learning Podcast June 2021

UK Anaphylaxis Guidelines

Tasty Morsels of EM 105, August 2017

Welcome back to the tasty morsels of critical care podcast.

This week we’ll make a fly by at part of Oh Chapter 100 looking at haemostatic failure. The understanding of the haemostatic system seems a little like the universe at times with our knowledge, and the gaps in our knowledge expanding away from us quicker than can we track. As a result this summary will be necessarily superficial, brief and likely inaccurate.

The traditional understanding of haemostasis was centred on circulating coagulation factors, the modern understanding holds that coagulation takes place mostly on the surface of activated or damaged cells. However, I think it still holds true that the core of haemostasis involves vascular constriction, platelet plugging with fibrin clot formation to seal the deal. 

I was reared with the idea of two pathways towards clotting in the old fashioned clotting cascade – the intrinsic and the extrinsic pathway. These appear to have been renamed in the interim with the intrinsic pathway (always on the left hand side in the diagrams i can visualise from memory) being renamed the contact activation pathway and the extrinsic pathway (being on the right side of the diagram) now renamed the tissue factor pathway. It seems that most coagulation activation occurs via this latter tissue factor pathway where the protein tissue factor is exposed by tissue injury beginning the process. Platelet plugging (or primary haemostasis) results with subsequent fibrin formation by the action of thrombin on fibrinogen, to complete the process. (secondary haemostasis)

Though to say the process is now completed is to neglect the presence of native anticoagulant processes which to be fair are rarely measured. This consists of several types of inhibitory proteins such as protein C and S to give a couple of well known examples. These act as a sort of brakes on the system to ensure everything is held in balance.

We have various tests we can apply to the haemostatic system but the more you look into this the more you realise you’re not measuring one thing but the test really reflects the activity of potentially multiple factors and an abnormal result cannot simply be used as a surrogate for any given coagulation factor but instead reflects the blended result of upset of multiple different factors.

Platelet number is relatively easy to measure but tells us very little about platelet function which can be affected by things like vWF or the aspirin they took 3 days ago. Platelet function is a little bit of a holy grail that we have not quite nailed down yet.

Prothrombin time measures the extrinsic or tissue factor pathway and is most obviously prolonged in people on vitamin K antagonists such as warfarin where it is expressed in the INR. In the critically ill it can also reflect vitamin K deficiency or deficiency of any number of factors in that pathway say from liver dysfunction. You can distinguish between vitamin K dependant causes using an Echis time. This fascinating test involves adding snake venom from Echis carinatus multisquamatus, an Asian viper to a sample of blood. If the prolonged INR is vit k dependant it should normalise, whereas if it is factor deficient it will remain prolonged.

The APTT is felt to reflect the intrinsic or contact activation pathway. it is most commonly used for measuring heparin activity (which is a post in itself) but if the APTT is raised in in the absence of heparin then it might reflect factor deficiency (eg one of the thrombophilias) where it should correct to normal with a mixing study where normal plasma is added. A long APTT in the absence of heparin may also reflect the presence of some kind of inhibitor with the commonest example being the lupus anticoagulant.

Of course we have the viscoelastic assays which are likely a more functional assay of global haemostasis than our traditional tests but they remain fairly niche in their use in critical care (as opposed to anaesthesia) in Ireland so far. They are due their own post in due course.

References

Oh’s Manual Chapter 100

Data Interpretation in Critical Care Medicine

Welcome back to the tasty morsels of critical care podcast.

This week we’re looking at the other ACS, the surgical ACS, the old abdominal compartment syndrome. This is common, especially in the surgical population and does not always immediately jump to the front of our cerebral hemispheres when it should do. Believe it or not there is a World Society of abdominal compartment syndrome that has been on the go since 2004 and you can become a life time member for free if you want.

They even have a set of consensus guidelines published back in 2013 that provide a wonderful template for an exam answer even if they aren’t supported by the highest level evidence.

So a few basics to start with. The measurement should be done by attaching a properly calibrated and zeroed pressure transducer to a port on the urinary catheter following an installation of no more than 25mls sterile saline in the supine position at end expiration. This is considered the reference standard though I do remember concocting some Mcgyver style manometer on a needle thing years and years ago.

There is a grading system for the degree of abdominal hypertension with the higher grades being higher pressures. Important to note that these pressures are in mmHg and if you’re using some old school manometry method then you may need to use the appropriate conversion factor. A normal intra-abdominal pressure is usually in the 5-7mmHg range in the critically ill as a reference.

Intra-abdominal hypertension is one thing and it is worth looking for but it is key to note that a high pressure on its own does not equal ACS. To be ACS you need a pressure >20mmHg and associated organ dysfunction with a good example or organ dysfunction being AKI.

Finally in terms of definitions you can split ACS into primary and secondary with primary being the intra-abdominal catastrophe where the surgeons find it hard to get the wound closed. Secondary causes are more likely to be related to medical illness with overly aggressive fluid resuscitation (otherwise known as iatrogenic salt water drowning) leading to oedematous abdominal structures and high pressures.

Once the diagnosis is made then the guideline has split the interventions into 5 categories summarised neatly in a table reproduced in the show notes. I’m not suggesting these are the ideal way to manage ACS but they are certainly reproducible in a viva or written type exam and certainly covers all the bases.

 

You can split up the interventions and categories as follows

1. evacuate the intraluminal contents. NG tubes on drainage, enemas and laxatives to clear things out from the bottom end. Removing contents from the abdominal cavity will lower the pressure. Colonic decompression with endoscopy is at the extreme end of this and I’ll confess I’d probably plump for neostigmine well before that.
2. evacuate intra-abdominal collections. If there’s a 10cm rim enhancing lesion on the CT 10 days post surgery then that needs dealt with not just for source control reasons but also as another adjunct to lowering the pressure
3. improve the abdominal compliance. This is something we’re kind of used to in intensive care as it involves deep sedation and even paralysis. If muscular tone is worsening things then get rid of it
4. optimise fluid resuscitation. Time to unleash your diuresis jedi or even maximize your ultrafiltration and get the patient in a -ve balance.
5. optimise perfusion. here they move into the concept of abdominal perfusion pressure which they define like CPP as MAP-Abdominal pressure.

These are all very nice and should all be reflected upon and followed when appropriate in your ACS patient. But it is critical that you don’t forget the all important step that comes at the bottom of this algorithm – if all else fails open the abdomen. In reality I’ve only ever seen this done in surgical cases where the belly is already open as part of the surgery and like a middle aged man trying to fit into an old suit it turns out you just can’t squeeze all the contents back in. One of the commonest examples would be the open emergency AAA repair where the large retroperitoneal haematoma just makes it too hard to close fascia and skin – they often return with the abdomen still open and I give thanks to the surgical gods above that they have done so. In the era of the EVAR these patients are much more at risk of ACS as the belly never gets opened.

References:

Kirkpatrick et al, Intra-abdominal hypertension and the abdominal compartment syndrome: updated consensus definitions and clinical practice guidelines from the World Society of the Abdominal Compartment Syndrome. Intensive Care Medicine 2013

Welcome back to the tasty morsels of critical care podcast.

Oh dedicates an entire chapter, number 88 to CBRN issues. While not commonly seen you can rest assured that critical care will be expected to turn up and manage these patients if an incident does occur so it is something we need to know fairly well. I will cover the content of chapter 88 here but I will borrow from a lecture I gave at the EuSEM conference back in 2019 when we used to actually go to conferences. Ultimately you’re going to get an edited selection of the available nasty agents.

These agents can be damaging, incapacitating or lethal and can come in persistent (liquids, particles etc) and non persistent (eg gases or volatile substances) forms. We can be exposed through a whole variety of routes from inhalation to the skin to ingestion.

There is a real risk of health care provider exposure and illness so you need to be sure that you are safe in caring for the patient.

When it comes to approaching these I found the CBRN chain of survival to be useful. This was published in 2019 by the Paris fire brigade and is an excellent illustration of the response. There is an image of this in the show notes but to run through it verbally, remembering that these are all meant to be performed pre hospital.

1. spot decontamination – Initial decontamination is wiping with paper roll and disrobing. this is likely to remove 80% of decontamination
2. early toxidrome recognition eg if its organophopshates
3. early antidotes
4. extensive decontamination
5. transport to hospital.

In terms of agents there is quite the cornucopia to choose from but we’ll begin with the chemical ones, and nerve agents such as sarin are up first. Sarin acts as an organophosphate resulting in overwhelming cholinergic activity due to inhibition of acetylcholinesterase. The toxidrome here is on the wet side with wet lungs, wet pants and salivation and diarroeah with small pupils and a side course of apnoea. The knee jerk response should be bucket loads (literally) of atropine to dry things up and some pralidoxime to get the breathing going by resurrecting the inhibited AChE, hopefully enough to enable muscular contraction working again.

When it comes to biological agents we’re just going to look at anthrax for today. Most famously brought to world attention in the days following 9/11 where packets of white powder where mailed to various parts of the US government, some of which actually contained anthrax.

Bacillus anthracis is the bug, a gram+ve rod, usually found in grass eating animals (hence the leather industry exposure) but it is also big in the IDU population in recent years. It causes mainly skin disease in humans but also GI and respiratory disease, usually with <48 hrs incubation period.

Cutaneous anthrax is usually painless with a red papule developing into an ulcer with septicaemia in 20%. Animal exposure is the obvious risk here.

Inhalation anthrax is more what we are worried about which can present as a flu like illness with cyanosis, sub cutaneous oedema and mediastinal widening from lymph nodes which can cause a haemorrhagic mediastinitis. This is unfortunately fairly non specific so you’ll like diagnose this from a culture off the skin lesion. Person to person airborne transmission does not occur so you don’t need a hazmat suit to manage them. Management is good old fashioned antimicrobials inc cipro, penicillin and clindamycin in combination.

Ionising radiation no doubt deserves its own post but to give an ultra summary here. We can get exposed in a few ways eg by radioactive waves such as gamma rays which is what happened to Bruce Banner. This type of radiation can usefully be blocked by something like a big concrete wall. Beta particles are a type of high speed electron emitted by radioactive decay. They penetrate tissue but can be blocked by relatively simple things like wood. If anyone saw the TV series Chernobyl – the extensive skin burns that many of the firefighters received were mainly from beta particles. Alpha particles consist of 2 protons and 2 neutrons bound together. Produced as part of alpha decay of a radioactive substance. They can only pass a a few cm through the air and are easily blocked even by skin and are effectively non toxic unless inhaled or ingested. They are at least 20 times more ionising than gamma or beta. The polonium 210 involved in the murder of Alexander Litvinenko involved alpha particles and the Lancet article on his death is a compelling read.

If your patient was exposed to something like pure gamma or X-ray radiation then there’s little role for decontamination as the radiation as passed through them and out the other side. They are no longer radioactive and the damage is done. If however there are concerns for beta or alpha particles involved (Chernobyl is a good example again) then the patients are likely to be covered externally with radioactive dust containing these ongoing radioactive particles and they may well have inhaled or ingested more particles meaning that their blood and secretions likely contain ongoing radioactivity. A geiger counter can help in assessing external exposure to these things.

Assessing the dose of radiation to the patient is important prognostically. There’s a couple of ways to do this but if you were to actually remember two things then:

• how quickly the lymphocyte count drops
• time to first emesis, the earlier you puke the more likely you’re going to die

a dose >5 gray puts you in the high risk for death.

References

Oh’s Manual 88

Criminal Poisoning Lecture form 2019

Tasty Morsels of EM numbers 87 and 88

Welcome back to the tasty morsels of critical care podcast.

It is with trepidation that I approach any topic that involves the negative feedback loops of endocrine control as I really struggle to keep it all straight in my head, but today I’m going to try and cover the basics of calcium in the critically ill from Oh Chapter 63.

We’ll start from some very basic physiology. 99% of the calcium is held in the bones. Of the calcium not in the bones, most of the rest is in the cells. So as an important starter, the serum level of calcium does not tell us much about overall levels in the body. In the plasma itself, 50% is ionised, 40% is plasma bound and 10% is chelated to various anions. There is a large gradient between the ionised calcium in the plasma and the tiny ionised fraction in the cells. Finally the ionised fraction is the active bit and the determining factor of endocrine regulation.

At a level somewhat below medical student level, calcium is controlled as follows. A low serum Ca stimulates PTH, this in turn stimulates osteoclastic activity and renal reabsorption. It also stimulates renal production of calcitriol (an active metabolite of vitamin D) which encourages gut absorption of calcium. Calcitonin produced by the thyroid acts as a kind of PTH antagonist dampening things down a bit if the calcium level gets too high.

Now let’s turn to hypercalcaemia, a fairly common diagnostic issue in the ICU. The major causes to consider are:

• Malignancy (both from bony mets and probably more commonly from PTHrP that mimics PTH in raising the Ca)
• High calcium following hypoCalcaemia (eg pancreatitis and recovery from rhabdo)
• primary hyperPTH
• granulomatous diseases such as sarcoid

Management of high calcium involves 2 mains steps

1. increasing urinary excretion, either with fluids alone or fluids and loop diuretics. Though the addition of loops is somewhat controversial as loops themselves have been known to cause HyperCa.
2. reduction in Ca resorption from bone. This can be done with bisphosphonates working by inhibiting osteoclasts. Or by calcitonin which inhibits osteoclasts and reduces renal reabsorption of calcium. Steroids can be helpful by reducing gut absorption and inhibiting the inflammatory production of calcitriol. Finally on the list is denosumab which as the name suggests is a monoclonal that prevents bone resorption

Hypocalcaemia on the other hand produces quite a prolonged differential including

• citrate (low ionised, high total)
• massive transfusion (which is just citrate again)
• high phos
• sepsis (multiple mechanisms including PTH suppression)
• thyroid or neck surgery
• low vitamin D (poor sun exposure, low intake, malabsorption, liver disease, renal failure)
• drug induced (bisphosphonates, propofol, EDTA, ethylene glycol, protamine, gentamicin)
• MRI contrast can cause a form of pseudo hypocalcemia  (niche and irrelevant..)

Many of the above are associated with alkalosis and indeed a separate differential can be formed for HypoCa in the context of acidosis

• AKI
• tumour lysis
• rhabdomyolysis
• pancreatitis
• ethylene glycol
• hydrofluric acid

Management of hypocalcaemia is both straightforward in that you just replace the calcium but at the same time desperately complicated in that you need to treat the cause which is often incredibly challenging.

A final mention is warranted of Vitamin D in critical illness given that it has been spread across the literature in recent years and is currently subject to lots of ongoing work. It is thought to have what is best described as “pleiotropic effects”. It is also similar to cortisol in that serum levels probably don’t reflect cellular activity. Either way there is no clear role for it in the ICU as yet.

References

Oh’s Manual Chapter 63

Welcome back to the tasty morsels of critical care podcast.

Condensing all of “inotropes and vasopressors” into a single 5 minute podcast is of course doomed to fail but that’s never stopped me before. The main reference for this is Oh’s Manual Chapter 92 by John Myburg who is known to me  for describing adrenaline as “God’s own inotrope” and in the same lecture describing dobutamine as “the Devil’s semen”. i have also heard him say with regards to fluid choice in the ICU that you can give any cat’s piss if you like as long as you do it carefully.

The chapter begins with a brief discourse on some of the physiology noting that ~ 20% of blood volume is held in the large conducting arterial vessels, meaning that the majority is held in the smaller vessels and venous structures. This larger venous proportion is often referred to as the unstressed volume. I think of it like the lazy river in a swimming pool, slowly meandering it’s way back to the RV while the arterial side is the flumes that you weren’t allowed on till you were 7 years old and you always had some unconscious fear that you’d enter in and never leave again. But that’s enough about my childhood.

Blood in this lazy river of unstressed volume returns on the venous side along a gradient from something called the mean systemic filling pressure (MSFP) to the lower right atrial pressure (RAP). Maintenance of a lowish CVP will therefore aid venous return.

In terms of improving cardiac output, autotransfusion of this unstressed volume (and increasing preload to LV) is the easiest and quickest and most effective way of improving CO.

Altering vascular tone and cardiac output can be done through a variety of systems:

• the adrenergic system
• renin angiotensin, aldosterone system (RAAS)
• Vassopressinergic
• Glucocorticoid
• Local systems such as nitric oxide and endothelin

Finally the determinants of cardiac output are stroke volume and heart rate. Heart rate in particular is easy to measure and causes issues at either end of the spectrum. When it’s too low the oveerall CO is too low, at some point it’s too fast, impairing cardiac and coronary filling and hence impairing stroke volume.

Pretty much all vasoactive medications have the same end point – that is the release, utilisation or sequestration of intracellular calcium. There are various methods to get there, many of which are cAMP dependant,  but calcium is the end point.

First off our beloved catecholamines. There are typically our first line in the fight against MAP<65. We have a fairly bewildering range of options available to us all with their own nuances.The nuances stem from the variety of catecholamine receptor biology we have evolved over the millenia. We know the basics of α and β but these can be extensively sub divided further in forms that only reinforces how little I understand about medicine despite over 20 years studying it.

For exam purposes I find having a rudimentary understanding of the differences between α and β stimuli is useful. Following a β receptor stimulus, there is increased cAMP while following an α receptor stimulus something called phospholipase C is engaged.

Tachyphylaxis is a common clinical phenomenon with the adrenergic drugs and reduced receptor density, sequestration and enzymatic uncoupling are all part of down regulation. Of note steroids act as pressors probably by increasing receptor sensitivity to catecholamines.

Both adrenaline and noradrenaline are predominantly β in action at lower doses with the α effect coming in at higher doses. Pretty much all the synthetic catechols are β in action with the obvious exception being phenylephrine as a pure α.

Myburgh is keen to make the point that at the doses we use, the catecholamines have no effect on arterial tone and CABG or vascular grafts, which is a frequent concern of our surgical colleagues who are understandably somewhat precious about their grafts and anastamoses. However it seems that when they go iscahemic it’s not the use of catecholamines is to blame but rather the severity of illness that requires the use of catecholamines. It is acknowledged that necrosis is common but more likely due to microthrombosis from sepsis rather than vasoconstriction from pressors.

Next let’s look at the phosphodiesterase inhibitors. Milrinone, enoxamone and levosimendan are all in this bucket.  They work by non-receptor mediated inhibition of PDE, ultimately resulting in increased cAMP. They are thought to be unique in that they may improve lusitropy (ability of heart to relax). All come with potent vasodilation so expect to have to crank up your pressor to compensate. They do not seem to suffer from tolerance and tachyphylaxis. It’s reported that they can inhibit platelet aggregation but unclear how significant this is in real life.  A major downside is their prolonged half life and dependance on working kidneys for excretion.

Vasopressin is another obvious category to discuss but I’ll save that for its own entry.

References

Oh’s Manual 92

Ashley Miller’s short video on MacroCirculation physiology

 

 

Welcome back to the tasty morsels of critical care podcast.

Today we cover an incredibly common inpatient issue – hypnatraemia. We’ll often find 1 or 2 of these in our high dependency unit at any given time, mainly due to the requirement for frequent testing of Na levels that seems beyond the remit of normal ward level care. The approach I describe here is neither comprehensive or especially robust but it is how I approach it. Caveat emptor and all that.

The over bearing demyelinating elephant in the room in hyponatraemia is the risk of osmotic demyelinating syndrome (the pathology formerly known as central pontine myelinolysis). If we correct the Na too fast will our patients end up with a severe brain injury? This is rare but is a very real phenomenon.The brain is actually quite good at adapting to sodium levels that have lowered over a few days or weeks. Hence why the slow developing sodium of 120 often causes minimal or no symptoms. However once the patient is in this adapted state (as mentioned this probably is after a few days at a minimum) then a rapid return to baseline sodium can cause ODS. By contrast a rapid drop in sodium, eg over a few hours drinking litres of unnecessary water during a marathon, is poorly tolerated but the plus side is it can be corrected fairly rapidly without harm.

Most of the hyponatraemia we see admitted through the ED will be hypoosmotic hyponatraemia. The bucket here will include heart failure, cirrhosis, SIADH, tea and toast and beer potomania. I’m going to put these common ones to one side for a minute and look at some of the niche exam ones.

For example, i said hypoosmotic hyponnatraemia there, so presumably there could be an isotonic and a hypertonic verison. There is indeed. The isotonic hyponatraemias are usually from spurious results. For example, when you have high lipids (super high, like high enough to cause pancreatitis high) or high proteins (eg high paraproteins like myleoma) the measurement method can underestimate the sodium. You can work this out by always sending a serum osmolality. If this is normal but the Na is 125 and your calculated osmolality is low, then you have an isoosmotic hyponatraemia. You should then check the lipids and the protein. Hypertonic hyponatraemia is another strange beast. This time the tonicity is high from something else such as high glucose or mannitol drawing water from cells into plasma. Again a mix of clinical context and a serum osm will help you out here.

Let’s go back to the bread and butter (or should i say the “tea and toast”) hyponatraemia, the hypotonic or hypoosmotic hyponatraemia. Context as always will give you lots of clues, if the patient has consumed nothing but beer for weeks then the likely causes is beer potomania. If the patient has a new cancer then SIADH is high up your list.

I confess I lean heavily on the approach you can see on Deranged Physiology and have Alex Yartsev’s flow diagram saved on my phone and i look at it almost every time i’m trying to work this out. The first test (assuming you’ve confirmed this is hypotonic hyponatraemia) in this algorithm is urinary osm, the question you are asking here is whether the kidneys are doing what they’re meant to be doing in the face of a low sodium. A normal sane and functioning kidney will try and lose water to conentrate the plasma in order to bring the sodium back up to normal, in other words the kidney should be producing a dilute urine with a low osm. Next step is to check the concentration of sodium in this dilute urine. If the kidney is doing what it should be doing it should be holding onto to all the sodium it can and urine sodium should be low. The problem here is too much water, not enough solute. Think, beer potomania, tea and toast, and polydipsia. If the urine is dilute but the sodium is high then you know something has gone wonky in the kidney itself, typically AKI or resolving ATN.

On the other side of the algorithm we have a concentrated urine, in other words, a high urine osm. The kidney is holding onto water and concentrating the urine. This may be a very sane and sensible response by the kidney if you are frankly hypovolaemic from eg gastroenteritis. The kidney also gets tricked by a few conditions into thinking its hypovolaemic, things like CHF or cirrhosis where the kidney itself just mightn’t be being perfused very well.  In this scenario you should have a concentrated urine with a high osm and a low urinary Na as the kidney holds onto Na for all its worth in an effort to maintain effective circulating volume. On the other hand you might find a concentrated urine with a high osm but a high sodium also. This tells us that the kidney is handling water reabsorption OK but has lost the run of itself when it comes to regulating sodium. Something may be strong arming the kidney into losing more sodium than it should, like thiazides or an external actor like ADH, in this case it would be inappropriate ADH, hence the syndrome of inappropriate ADH. In addition a lack of steroid (and in particular the mineralocorticoid part) or a dodgy thyroid may cause the kidney to lose sodium when you shouldn’t. Or of course this scenario could be due to intrinsic renal disease.

So that’s 8 or 900 hundred words running through the deranged physiology algorithm and you can imagine that simply looking at the algorithm would probably be a better use of your time so go do that.

Next time we’ll have a look at how we might manage hyponatraemia

Reading

Deranged Physiology – Wonderfully titled ” A Lazy Man’s Classification”

Oh Chapter 95

Welcome back to the tasty morsels of critical care podcast.

In a further scandalous departure from Oh’s Manual today we’re going to look at a chapter of verified Irish Critical Care legend, Martin Tobin’s huge mechanical ventilation textbook. I have made it through about 5 chapters of this beast and it is undoubtedly comprehensive.

Anyhow, today will be nitric oxide, covering Tobin’s Chapter 61.

Nitric oxide is a colourless, odourless gas that exists in the atmosphere at anywhere between 10 and 500 parts per billion (emphasis on the billion here). Oddly it exists in quite high concentration in the nasal sinuses where it has been found at concentrations up to 30ppm which is in the therapeutic range. It seems that it has some kind of antimicrobial role here in the snot factory. NO is generated by the enzymatic actions of the practically named nitric oxide synthase enzyme.

How does it work in the lungs? Well the basic principle is that when NO reaches an alveolus it encourages more blood to flow past it. Thus it improves matching of ventilation with perfusion otherwise termed as improving V/Q matching. The hope is that in doing so it will also divert blood away from non ventilated alveoli reducing shunt. In a wonderfully patient oriented mechanism it becomes inert as it traverses the alveolar-blood membrane thus taking care of its own clearance and preventing systemic side effects.

There are a few potential indications for nitric

• Persistent pulmonary hypertenison of the newborn where it seems to substantially reduce the need for ECMO.
• RV dysfunction, particularly surrounding cardiac surgery or for primary graft dysfunction in lung transplant
  • it has been definitively shown to reduce pulmonary artery pressures but not mortality
  • in this scenario it can be used at a much higher dose than ARDS at up to 100ppm
• ARDS
  • usually at a fixed dose of 20ppm
  • better in primary pulmonary ARDS (eg pneumonia, aspiration) rather than secondary causes (such as pancreatitis)
  • no study has found definitive benefit
  • it can be used as a brief trial at 20ppm in refractory hypoxaemia and you’re looking for a ~20% increase in the PaO2. Only about 50% respond.

Beyond the hassle of setting it up and getting all the plumbing on the vent right, are there any issues with giving it? Well, glad you asked there are indeed a few concerns.

• Nitric Dioxide (NO2) is the evil cousin of NO that can cause pulmonary oedema and haemorrahge. Can be formed with high concentrations of NO and O2 together and the machine monitors for it
• there is often rebound hypoxaemia after withdrawal probably related to suppression of native NO synthase. Sometimes it can feel like you’re stuck on homeopathic doses of NO trying to wean it off safely.
• MetHb can be an issue as a byproduct of NO binding to Hb at the alveolar membrane. This is mentioned but is not something I have seen in real life.
• it is thought to impair platelet aggregation, however clinically a variety of anti and pro coagulant effects have been noted
• there is a definite association between AKI and NO in the ARDS trials of unclear mechanism

References

Tobin’s Principles and Practice of Mechanical Ventilation

LITFL

Welcome back to the tasty morsels of critical care podcast.

Today we’re going to talk about a fairly rare and niche issue in critical care – gas embolism. The venerated stone tablets of Oh’s Manual do not mention it in any great detail but my alternative and go to textbook has been Irwin & Rippe’s weighty tome/deadweight, and chapter 177 here is dedicated to gas embolism syndromes.

This can quickly be split into venous and arterial emboli and we’ll start with the venous side

There are a few causes for this and a reproducible short list might include

• Surgical causes
  • sitting awake craniotomy the classic case where air gets into a venous sinus
  • any procedure with an open vein
  • it is worth noting that when this is searched for in surgeries that are considered high risk it is very common to find signs of air in the venous circulation on doppler. It is a testament to the lungs as a protective filter that it doesn’t cause more issues
• Traumatic causes
  • a stab wound to a big vein
  • major thoracic injury
• Procedural
  • CVC being the most obvious one we think of
  • it has been reported with peripheral IVs and even epidurals (though I can’t quite get my head around the mechanism there)

Ultimately there needs to be some kind of pressure gradient between the atmosphere and the vein in question hence why position and vigorous respiratory effort are risk factors as both induce a pressure gradient and rapid flow toward the right atrium. The air can get trapped anywhere between the entry point and the pulmonary circulation and it is then obstruction of flow through and out of the right heart that causes all the drama.

100mls is considered to be a fatal “dose” in case you were wondering.

Paradoxical emboism is a very real concern here. 1 in 5 of us is walking around with a PFO  and generally it’s not a problem as the slightly higher pressure in the LA vs the RA ensures that the little flap of tissue sits closed. However in the scenario of raised right heart pressures (say when there is 50 mls of air trapped in the RV) then the PFO can blow open and air can enter the left sided circulation converting the scenario from a venous gas embolism to the as yet unmentioned arterial gas embolism situation.

In terms of making the diagnosis the context is everything – what was the situation when the haemodynamic collapse occurred? Did they rip out their vascath while sitting upright attacking staff with an IV pole prior to their cardiac arrest?

There is apparently a mill-wheel murmur that has been described that i suspect is useful only as an answer in an exam. In reality this will probably be a tricky echo diagnosis or a slightly late and embarrassing CT diagnosis. Hopefully you’ll have nailed it at the context stage and proceded to treatment.

Treatment involves stopping any further gas entering the circulation which will likely involve clamps on vascular devices or sticking a finger on the hole in the vessel. The ninja move that is needed next is to lie the patient on their left side in the hope that the air lodges in the RV apex and allows an unobstructed conduit from RA through RV and on to PA. Next try and get a catheter into the RV and aspirate some air. This always make me think of some form of reverse John Travolta/Uma Thurman in Pulp Fiction precordial syringe stab but in reality it’ll be a CVC placed in a rush.

100% O2 is considered a good idea as physiologically you might replace the insoluble nitrogen in the apex of the RV with pure metabolisable O2 but this didn’t really pan out in PTX and I suspect is useful only in theory. This is another time to suggest hyperbaric oxygen but again, in reality, taking a critically ill dying patient to a single person compression chamber seems like a recipe for disaster but perhaps it is something that is done.

Let’s turn to the arterial side.

Overall the venous side will tolerate gas somewhat better than the arterial side, with the lungs acting as a giant filter however once it gets big enough you can obstruct your RV and the usual RV spiral of death beings. Air in the left side is much less problematic haemodynamically but neurologically is devastating.

The list of causes is similar to those for venous gas embolism as many of them can shunt through a PFO In addition some important ones to note include

• cardiac surgery
• angiography
• traumatic injuries like a bronchovenous fistula will rapidly entrain blood into the LA and LV
• decompression sickness
• carotid endarterectomy

The pathophys is one of simple obstruction to flow in the brain but there is a corresponding endothelial injury that has its own consequences

The treatment here is much less exciting as all we can really do is give 100% O2 as a strategy and consider HBO. And by HBO i mean hyperbaric oxygen not the TV channel.

References:

Irwin & Rippe’s Intensive Care Chapter 177

 

 

Welcome back to the tasty morsels of critical care podcast.

With extreme brevity we are going to try and cover Oh’s Manual Chapter 38 on respiratory monitoring. This is something of a hodge podge i must admit. I’ll start by looking at the lung mechanics section.

The first useful point is that the pressures recorded by the vent are usually interpreted as reflecting the compliance of the lungs when in fact they often are significantly impacted by the chest wall, most commonly in the case of obesity. The oesophageal balloon is probably our best way round this as it gives a fairly accurate surrogate for pleural pressure. There is some data supporting the use of an oesophageal ballon to guide PEEP titration by calculating transpleural pressure (PEEP – oesophageal pressure) and typically you end up on a higher PEEP in the obese patient than you would have otherwise. This is all theoretical for me as these devices have not been available to me throughout my training and the outcome data for their use has not been supportive but all the smart ventilator people talk about it when they give lectures.

Talking of PEEP, one of the things we have to look for is intrinsic PEEP, sometimes called auto-PEEP, sometimes called dynamic hyperinflation. Intrinsic PEEP adds to the work of inspiration, with additional effort needed to overcome the additional +ve pressure within the lungs. We most commonly measure this with the end inspiratory expiratory hold (as Dean points out, I misspoke and of course means end expiratory hold). The number generated here is the static intrinsic PEEP which should be distinguished from its cousin – dynamic intrinsic PEEP.

Dynamic intrinsic PEEP instead reflects the pressure change needed to initiate inflation of the lungs. Oh describes a way to measure it while pointing out its complextity as you need oesophageal and even gastric balloons to appropriately quantify it. All this is to say dynamix intrinsic PEEP exists and it’s tricky to measure.

in the same chapter we have patient-ventilator asynchrony and we are bemoaned for missing it. Oh describes this as most commonly caused by a failure of triggering and notes that the best way to pick it up is to look for deflections on an oesophageal balloon that are not followed by a breath from the vent. When you don’t have a balloon then you’re stuck with looking at the patient and the vent waveforms. Note deflections in the pressure waveform are much less sensitive for asynchrony than changes in the flow waveform.

Autotriggering, defined as initiation of breath without patient effort, comes in a few flavours with, cardiac oscillations and hiccups being well known examples.

One important concept to think of is that of matching the length of inspiration from the vent with the length of inspiration that the body wants. This is described as mechanical inspiratory time vs neural inspiratory time. Where the mechanical time is shorter than the neural time then the body is not getting the breath it wants either in terms of the duration or the flow of gas it wants. As a result the body triggers the next breath very shortly after the completion of the first. Occasionally the mechanical breath can be longer than the neural breath and as a result the lungs are passively inflated rather than assisted or supported.

Following on from this we have a few methods of measuring neuromuscular function. Firstly we have the P0.1 or the airway occlusion pressure. (I had this wrong, the P0.1 and airway occlusion pressure are distinct entities.) The P0.1 measures the negative pressure generated in the first 100ms of inspiration. In the spontaneously breathing patient on a support mode this gives you some idea of the work of breathing or respiratory drive. A normal value is somewhere in the 1-5cmH20 range and I have used this as a means of checking if the patient is working harder than I’d like in a PS mode.

The Toronto group has recently published on using the occlusion pressure as an alternative method that measures the “swing in airway pressure generated by respiratory muscle effort under assisted ventilation when the airway is briefly occluded”. You can see why they use detalPocc as a surrogate. Personally I would just call it the occlusion pressure but that is of course not very specific. This is done on someone breathing in a spontaneous mode. An expiratory hold is applied randomly and the negative pressure deflection as the patient inhales against a closed valve is measured. This number is the occlusion pressure. In their study they used the oesophageal balloon and some methodological wizardry to see how this occlusion pressure weighed up agains things like Pmus (the pressure generated by the respiratory muscles) and  the transpleural driving pressure. Overall they conclude that the occlusion pressure is a good surrogate for the presence of badness in these numbers while not being able to quantify them. This has relevance as we commonly sit back and relax when a patient goes on a PS mode and the peak and driving pressures go from bad numbers to good numbers. However there may well be adverse “dynamic lung stress” that is going on unmeasured unless you use this nifty little move to look for it. I suspect many of us do this automatically by eye balling the patient and their sterno cleido mastoids and their overall work of breathing. Even looking at how negative the swing on the pressure trace is before the vent kicks in is some kind of surrogate for this. What we do about this I am not entirely sure but I am now worried about patients even when they’re in a pressure support mode. Thanks a lot Toronto…

Another interesting way of monitoring NM function is to look at the electrical function of the diaphragm. A modified NG tube is placed with a sensor that picks up electrical activation of the diaphragm and uses this to trigger the vent. I have used this a grand total of once and found it intriguing but of little use in the n = 1 case series.

If you were still looking for other things to place in an SAQ on monitoring the vented patient then a mention of diaphragmatic ultrasound is worth throwing in. Basically you can visualise the diaphragm fairly well and use its observed thickening as a way of predicting various things about ventilated patients. It remains as far as i can tell a fairly niche application of PoCUS.

Finally I would like to point you to a site called RTMaven.com that comes out of the group in Toronto.  This has some wonderful respiratory monitoring calculators with some nice videos on how to perform the particular moves on our beloved servo-i vents. Highly recommended.

References

Oh’s Manual Chapter 38

Telias I, Damiani F, Brochard L. The airway occlusion pressure (P0.1) to monitor respiratory drive during mechanical ventilation: increasing awareness of a not-so-new problem. Intensive Care Med. 2018 Sep;44(9):1532-1535. doi: 10.1007/s00134-018-5045-8. Epub 2018 Jan 19. PMID: 29350241.

Conti G, Antonelli M, Arzano S, Gasparetto A. Measurement of occlusion pressures in critically ill patients. Crit Care. 1997;1(3):89-93. doi: 10.1186/cc110. PMID: 11094467; PMCID: PMC137221.

RTMaven.com

Bertoni, M. et al. A novel non-invasive method to detect excessively high respiratory effort and dynamic transpulmonary driving pressure during mechanical ventilation. Crit Care 23, 346 (2019). Chen, L. et al. Potential for Lung Recruitment Estimated by the Recruitment-to-Inflation Ratio in Acute Respiratory Distress Syndrome. A Clinical Trial. Am J Resp Crit Care 201, 178–187 (2019).

Welcome back to the tasty morsels of critical care podcast.

This is part of Oh’s Manual Chapter 77 on head injury and we covered ICP monitoring before in number 20.

A key principle here is cerebral perfusion pressure or CPP. This is easily calculated at the bedside as MAP-ICP. Of course this is only easy if you actually have an ICP monitor.  The CPP should probably be around 60-70 which assuming your patient is unconscious means that the ICP is probably north of 20mmHg and therefore your MAP should be around the 80-90mmHg range. If you actually have a monitor then you can be much more scientific about your MAP target. A moment’s thought about the complexities and unknowns of cerebral blood flow and perfusion in the acutely head injured patient should hopefully make you realise that targeting a CPP is an incredibly blunt tool for so fine tuned an instrument, but given the lack of anything better and strong recommendations from international guidelines then we better stick to it.

Oh splits TBI into two phases which I find useful as a concept even if I still find it unclear when a patient transitions from one phase to the next.

1. the early hypo-perfusion phase: after a bump to the noggin there is a rise in pressure from extra axial lesions or oedema that necessitates a bump in your MAP to keep CPP in the right range.
2. the hyperaemic phase: about a quarter will have this at around day 3-7 and it seems ill defined but it’s suggestive you can lower your MAP targets here a bit

Overall the basic bundle of interventions in the ICU for ICP include:

• sedation – this is needed for the tube but also it reduces metabolic rate considerably and reduces ICP. There is no clear advantage of one agent over another although the ileus inducing agent thiopentone is useful in refractory settings
• SUP – this is one of the groups particularly at risk of GI bleeding
• keeping the head up 30-40 degrees can promote venous drainage and indeed you’ll see ET tubes taped to the cheeks rather than circumferential ties in many neurosurgical units.
• DVT prophylaxis should be given – the timing of which is a topic in itself and frequently delayed longer than it should be. The phrase used typically is when the appearances on CT are “stable” whatever that means but it probably does include the absence of fresh bleeding at the very least.
• CO2 should be in the normal range and same goes for oxygenation. Now is not the time for permissive hypercapnoea and so for your patients with head injury and ARDS this can be really challenging

The BTF guidance is the key guideline document that should be referenced and it is well worth a read for any exam candidate and indeed any clinician dealing with TBI. The following summarises some of the headlines.

Decompressive craniectomy

• a bifrontal cranitoyomy is not recommended to improve neuro outcomes, which is a way of acknowledging the important DECRA trial.
• if done then a large frontotemporal craniectomy is recommended
• of note these were published before the results of the RESCUE-ICP were available

Hypothermia

• there is a substantial literature and a compelling physiologic argument that cooling would be good
• however a sequence of trials have put this to rest with the key ones being
  • Eurotherm 2015
    • RCT ICP>20, 400 pts
    • small impact on ICPs
  • POLAR 2018
    • 500 pts RCT
    • no difference, again little impact on ICP also. There was a decent increase in pneumonia here.

Hyperosmolar

• mannitol effective controlling high ICP but reserve it for those who look like they are near herniating
• importantly they have no recommendation on hypertonic saline. Doesn’t mean it doesn’t work just that not enough data yet to sell it over mannitol

CSF Drainage

• an EVD can be useful to drain CSF (and lower pressure) in the early period

Steroids

• no, no and no again…

Seizure prophylaxis

• not recommended for preventing late epilepsy (which is an obscure answer to a question no one asked as we use it typically to prevent early seizures)
• phenytoin is recommended to reduce early (within 1 week) seizures when “benefit outweighs risk” (though early seizures not shown to worsen outcomes)
• not enough evidence to recommend keppra over phenytoin but you can guess what happens in real life

Tagged onto the end of this post on ICP management I’ve included recommendations from a separate BTF guidance that covers when to operate on TBI. In reality this should maybe have been covered first as managing an expanding extradural with mannitol and thio is doing it all wrong. These particular guidelines can be printed out and rolled up and used as a stick to beat your neurosurgeons into operating. But i jest.

SDH

• clot >10 mm, >5mm shift, regardless of GCS
• those with less severe numbers but low GCS or unequal/fixed pupils should also get surgery

EDH

• Conservative: <30cm3, <15mm depth, <5mm shift and GCS>8
• Surgical: all >30cm3 should be evacuated.

Parenchymal TBI

• any >50cm3
• >20cm3 and shift >5mm with GCS 6-8

Posterior fossa

• most of these get evacuated (due to mass effect or reduced GCS)

Depressed fractures

• open and depressed >thickness of the skull generally get surgery (though not all)

In reality I find this guidance is viewed somewhat loosely by our neurosurgical colleagues and given the level of evidence it’s based on they may well be right but we do run into the usual neurocrit care dilemma of self fulfilling prophecy.  Managing these patients with medical therapy when they should have had surgery inevitably results in bad outcomes.

In closing, I realise I have neglected to mention the 3 tiered approach to ICP management where you provide therapies in a step wise fashion but alas i have ran over time already.

References:

Oh’s Manual 77

BTF Guidance

BTF Surgical Guidance

 

 

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

Oh’s Manual chapter 104 has a decent chapter on the intensive care aspects of lung transplant. A lot of this stuff will end up fairly centre specific but there is a good general overview to be had here.

Lung transplants do reasonably well overall (most likely due to careful selection) with a median survival of 7 years and 80% alive at one year. Donation after cardiac death is becoming an increasing source of donor organs and the lungs seem relatively tolerant of the short warm ischaemic time (though kidneys are still the best overall).

Who might be a candidate for lung transplantation? The generic phrase is “advanced, non malignant lung disease, for which there is no alternative therapy” However a list of common potential indications might go as follows

• advanced non malignant lung disease without alternative
• COPD/Alpha1 antitrypsin deficiency
• Pulm fibrosis (single lung Tx still done for this sometimes)
• CF
• Bronchiectasis
• Pulmonary vascular disease

The other major component of patient selection is how well they’re going to tolerate the overall burden of surgery, intensive care and the long term prospects of immunosuppression. As a result candidates are typically fairly stable with severe disease and generally admitted from home unlike the liver transplant patient who may well get their transplant in the midst of multi-organ failure. There are exceptions such as using VV-ECMO as a bridge to transplant akin to an LVAD as a bridge to transplant but these should be considered as truly exceptional.

The potential donor is typically split into one of two categories

• standard donor
• high risk (marginal) donor

These are distinguished by obvious factors such as age and smoking history and the state of the donor lung at the time of procurement.

As someone without an anaesthesia background, I have little insight or involvement with the procurement process but I do find the recent development of ex-vivo perfusion devices to be fascinating. These are now recently developed for the lungs which allows donor lungs to be perfused and even optimised prior to implantation.

The surgical technique used is typically a sequential single lung transplant, one at a time with individual bronchial anastamoses rather than a complete “en-bloc” technique with an anastamosis at the trachea. If the native lungs are good enough to tolerate single lung ventilation then it can be done without bypass but the second lung is obviously dependant on immediately effective perfusion and gas exchange in the first implanted lung to allow the second lung to be implanted. This is certainly outside the realm of intensive care but remains a somewhat magical act that my surgical and anaesthetistic colleagues perform in the operating room.

Post op is where intensive care is subbed onto the field to take over the physiology of the fairly roughed up transplant patient. There are a range of potential issues to cover that are probably best split up by system.

From a respiratory perspective the grafted lungs are denervated and therefore there isn’t the usual cough and mucociliary clearance at least in the early phases. The general idea is to wake and wean and extubate but ~15% need prolonged ventilation.

Single lung transplant can cause a lot more issues as the two lungs can often have markedly different respiratory dynamics and will experience the pressures from the ventilator very differently. The same discrepancy between lungs is also experienced by the right heart which continues to pump expecting the long standing increased PVR it’s used to and now once blood gets into the pulmonary trunk it is faced with either an obstructive, high resistance PA on one side and nice low resistance system in the other PA. This can end up with the new lung receiving a disproportionate amount of the cardiac output that can flood the lung. Indeed the right heart is one of the major causes for concern overall here.

Immunosuppression typically comes with some early basiliximab and ultimately an antimetabolite and a calcineurin inhibitor.

In terms of infection, prophylaxis plays a big role as the lungs are more exposed to the manky, dirty outside world than most of the other transplanted organs. You will typically have some sputum cultures from the donor already growing before the transplant so that can help guide your antimicrobials. PJP prophylaxis will be needed but there’s no great rush. CMV is needed for all but as usual mismatched donors/recipients have the highest risk.

Primary graft dysfunction typically presents as worsening gas exchange, decreased compliance and alveolar infiltrates. Ischaemia reperfusion injury is a common cause. It is non immune mediated and an important cause of early mortality and causes long term morbidity. It’s worth noting that it’s a recognised indication for VV-ECMO.

Other things to look for in the early stages are anastamotic ischaemia (the bronchial arteries are typically not anastamosed so perfusion is not as good as native). This occurs in ~2%. Phrenic nerve palsy is an understandable complication in ~5%. A fairly niche one is pulmonary venous kinking where one of the newly anastamosed pulmonary veins has got kinked in the process. TOE or a CTA can see this.

Acute rejection on the other hand is a delayed phenomenon with lots of changes that can look like infection and indeed both can happen at the same time. Ultimately you need a biopsy and once you find it, it’s big doses of pulsed steroids a treatment.

Post transplant lymphproliferative disorder (PTLD) can occur in 5-10% of lung transplants (much higher than other solid organ transplants). Usually this is with EBV infections with B cell proliferation which is unopposed as the T cells are suppressed by the immunosuppression.

References:

Oh’s Manual chapter 104

LITFL

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

This week: serotonin syndrome. Much talked about, very common in an exam, quite rare in real life. It doesn’t really get much coverage in Oh’s manual so the following is prepared from the hodge podge of resources listed at the end.

In normal circumstances we produce serotonin from tryptophan. There are a bunch of complicated pathways and mechanisms that are as usual more beyond my comprehension as opposed to beyond the scope of this article. Serotonin does masquerade under its alternate name of 5-hydroxytriptamine which in turn has bred a whole range of receptors of which most relevant is the 5HT2A receptor.

The other piece of the puzzle here is monoamine oxidase (MAO). This enzyme is responsible for metabolism of serotonin. So when we give drugs that inhibit MAO we can also get into trouble

So in summary 5HT (pseudonym for serotonin) stimulates neurons through a variety of receptors but the 5HT2A receptor is the one that gets us into bother with serotonin syndrome. We can get into bother by having too much 5HT stimulating the 5HT2A receptor or we can get into bother by not getting rid of 5HT with MAO when we need to. Clear as mud I’m sure.

This is all very interesting but it’s clear from the now somewhat dated 2005 NEJM review article that this piece of knowledge is only the tip of the iceberg in terms of its pathophysiology. Though rest assured the tip of the iceberg is likely to be enough for exam purposes.

In terms of clinical presentation it’s clear that it exists as a spectrum from mild to ICU level. A common mnemonic used is CAN

• CNS dysfunction
• Autonomic disturbance
• Neuromuscular effects

To flesh this out in bullet point form, these are the type of features we’re looking for:

• onset <24 hrs
• big pupils
• tremor
• clonus (esp spontaneous)
• hyperreflexia
• fever
• autonomic issues
• rapid resolution within 24 hrs with treatment
• hyperactive bowel sounds (you may snigger at the inclusion of BS here but in terms of distinguishing this from its examinable partner, NMS, the presence of hyperactive bowel sounds may be one of the few if only indication to listen for them)

The proposed algorithm for diagnosis here is the Hunter criteria. Named for the lush area of NSW where they make nice semillion wine and have the Hunter Valley Toxicology service that named this a few years ago. It is a step wise algorithm that in the limited science available claims a 95% specificity for diagnosis. This is one of those algorithms (like the CAM-ICU) that refuses to stay in my brain so invariably I end up looking it up. Sadly this is not an option in an exam.

It is important to remember that this algorithm only applies in the presence of a serotonergic agent. There is, it seems a quite large number of agents to populate this list. Some of them are obvious with the clues in the name like Selective Serotonin Reuptake Inhibitors (SSRIs) or Monamine Oxidase Inhibitors (MAOI). There are however, a few unusual ones such as linezolid (which acts as an MAOI), fentanyl (which is serotonergic) and usual tox favourites such as cocaine. Tramadol is on the list, which is unsurpising given that it won 1st prize at the recent Toxicology awards for filthiest side effect and interaction profile ever. The final two worth noting are lithium (which can increase sensitivity to 5HT) and methylene blue which sort of acts like a toxicology double agent by playing the hero in methhaemolobinaemia then stabbing you in the back with a serotonin syndrome.

Overall expect to see this in 2 scenarios

• the overdose of a single serotonergic agent – the Saturday night young person presentation
• the cumulative effect of poor prescribing in a hospitalised patient

The archetypal case of the second presentation was with Libby Zion, the young New York lady who died from unrecognised serotonin syndrome due to inadvertent combination prescribing of serotonergic agents by undersupervised and overtired junior doctors. It is thought to be the key case that led to reforming residency hours in the US from life threatening indentured servitude to the more modern underpaid and overworked trial by fire that it seems today.

Once you’ve made the diagnosis then you should fill the rest of the answer with something like this

• supportive care – usual ICU issues, you can split by organ system to cover everything. This might be a good opportunity to wax lyrically about controlling temp either with a bellevue ice bath or a CRRT circuit.
• specific/antidotal therapy

Cyproheptadine is the drug that is listed in all the books. This is an ancient 1st gen antihistamine that has an affinity for the 5HT2A receptor greater than that of serotonin. Like all talks about toxicology it provides an excellent reason to use the distracted boyfriend meme, an example of which is shown below.

As you can imagine an early generation antihistamine is probably not the ideal agent for this these days, especially with a dosing regime of NG administration every few hours of a drug that you will probably not be able to locate after hours.  It persists, likely due to the challenging evidence base that tox has to work off. The same Hunter Valley Tox people who came up with the criteria have published their experience with chlorpromazine and olanzapine as alternate agents to cyproheptadine. There is little data to support one over the other but a bit like COVID vaccines, the best one is the one you can get. (This is a comparison that may not hold up over time…)

References:

NEJM 2005 Review article

LITFL

Deranged Physiology

Geoff Isbister et al (from Hunter Valley…) review from 2014 

One of the early Hunter Criteria papers from 2003

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

Today we’re talking a little about that most vital of gases – oxygen. This is going to be back to some very basic physiology from Oh’s Manual Chapter 28, that I probably should have learnt in medical school but honestly looking back I’m not sure I learnt anything in medical school except how to scrape by with the minimum of effort and knowledge. My post graduate career has been somewhat more enthusiastic I might add.

Oxygen is that vital substance that we need in order to conduct oxidative phosphorylation which is the body’s most efficient means of producing ATP.

For oxygen to get from, say, my left nostril, to a skeletal myocyte in my right tibialis anterior, it has to go through roughly 5 steps

1. convection of O2 to alveoli (this is ventilation)
2. diffusion through alveolar membrane
3. reversible bonding with Hb
4. transport to tissues (CO dependant)
5. diffusion to cells and organelles

Let’s run through that in a  little more detail. Oxygen is drawn in through the big bellows of the lungs and at the alveolus it meets its first real challenge – how to get across the membrane. Here, it obeys Fick’s law of diffusion, where “the rate of diffusion is proportional to both the surface area and concentration difference and is inversely proportional to the thickness of the membrane”. In other words when there’s lots of membrane that is very thin, and not very much oxygen on the other side of the membrane then diffusion is at its best. In apnoea for example the continued blood flow through the lungs draws away any O2 increasing the gradient across the membrane. O2 is drawn through the membrane and in turn more O2 is drawn from the larger airways. This is one reason why apnoeic oxygenation works so well, we see this perhaps most dramatically during the apnoea testing for brainstem death. Simply maintaining a high concentration of O2 in the airways will continue to keep a patient oxygenated even in the absence of bulk flow of gas through breathing.

Once diffused into the blood it then joins up with its best friend forever – haemoglobin. Oxygen would much rather be joined to Hb than merely dissolved in the blood. At this stage I am legally required to mention the oxygen-Hb dissociation curve. Despite years of doing this I still cannot get my the left and right shifts stuck in my head. What has stuck in my head is the intelligent adaptation of a system that encourages oxygen offload in areas of the body that are hot, full of CO2 and acidotic – for example muscles working at high load. This might be a rightward or a leftward shift, I really can’t remember but thankfully the human body seems to do it without my input so all is well.

This oxy-Hb relationship allows us to move large amounts of oxygen around the body fairly easily. However, of course it is dependant on the cardiac output to get it to where it needs to go. The amount of oxygen the pump can deliver is dependant on flow but also on the oxygen content of the blood. An Hb of 15 will carry more oxygen for a CO of 5L/min than an Hb of 10 for a CO of 5L/min. The oxygen carrying capacity of the blood combined with the CO can be put together to form the oft cited DO2. DO2 is one of those abbreviations for a physiologic concept that has an off little dot above one of the letters and superscript 2 making it altogether difficult to reproduce online without a bewildering number of keyboard shortcuts. DO2 is also best discussed with reference to its partner VO2, indeed combined you can use the term DO2:VO2 relationships in a physiology discussion on a ward round and pray no one asks a follow up question. Perhaps something actually worth knowing is that resting oxygen delivery (DO2) comes in at roughly 1000ml/min. This is the supply half of the relationship. The oxygen consumption comes in at around 250ml/min. This is the demand half of the relationship. We see this reflected in the normal venous saturations of around 65%. If the body was ran by a socially funded health service in English speaking Western Europe, then I doubt we’d see such generous reserves. Why are we going to all this effort supplying 4 times the amount of oxygen than the body actually needs in any given minute? But the body is smart, having learnt its way through natural selection that the best way to avoid a sabre tooth tiger eating you is to have a large reserve of oxygen supply in the event that you need to get away from said Tiger in a hurry. The nationalised health services that we know and love could do with a lesson in supply and demand from the perspective of DO2:VO2 relationships. But I digress.

Having navigated the rushing rivers of the circulation our intrepid hero, the oxygen molecule is nearing its final destination – the mitochondria. It has to cross the cell membrane and get to the mitochondria and again it finds benefit in Fick’s principles of diffusion in addition to the shift in the oxy-Hb curve mentioned above. In my head I pictured oxygen tensions in the mitochondria similar to that taken from my arterial blood gas. If I had been guessing in a medical school undergraduate exam (which was generally my default position) I would have said the PO2 would be in 8-10 kPa range. In reality when we have measured oxygen tensions in the mitochondria we find it to, quite staggeringly, to be as low as 0.133 kPa. Below this the mitochondria gets a little upset and tends towards anaerobic respiration. So next time the med reg is trying to get someone into ICU with a PaO2 of 7.5 kPa you can always be smug and say “call me back when it’s less than 0.133 kPa…” and hang up.

References:

Oh’s Manual Chapter 28

 

 

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

Today we’ll talk about CRRT timing. When should your critically ill patient take a spin on the green machine?

From an exam point of view this can go a few directions. You can go core physiology or you can go down the literature route. I think both are worth covering.

Oh Chapter 48 has a list of indications that include:

• volume overload
• electrolyte issues – classically refractory hyperkalaemia
• severe acidosis
• uraemic symptoms inc pericarditis
• progressive AKI (oliguia or anuria, high creat/urea)
• Oh also has temp>40 as an indication
• Dialysable toxin

These all seem fairly reasonable.

There is an interesting concept called the furosemide stress test that was first described by Chawla in 2013 and then repeated by Rewa in 2019. The concept here is that you can use the response to furosemide to predict the need for CRRT. The dose is between 1 and 1.5mg/kg of furosemide looking for >200 mls of urine in the first 2 hrs. If it’s less then wheel out the green machine and get started. Of note it is not used to determine who is ready to stop CRRT which seems to be a much slippier beast that i’ll mention at the end.

Physiology over and done with for now let’s look at the literature that looks at this question. The papers here are all classics in intensive care and a careful student of the discipline should have a passing familiarity with all of them (and I am not even going to cover all of them!)

Once upon time there was the AKIKI trial led by Gaudry and published in NEJM in 2016. This was a French study (side bar – i know the locations are largely irrelevant to the science here but the location makes it so much easier for me to remember as I can picture a polo necked french man with onions round his neck as a means to remember the paper). Anyhow. Once you had stage III AKI then you could be randomised to CRRT or wait till a solid indication like hyperkalaemia developed. They included 600 mainly medically septic and ventilated patients. Mortality was pretty much the same and of note only 50% actually ended up needing CRRT in the delayed group which is a recurrent theme worth noticing.

In the same year, Zarbock and the Germans published in JAMA the ELAIN trial. Again, randomised but single centre and enrolled patients at an earlier stage (stage 2 by KDIGO). They got 200 pts with a 40% v 50% mortality favouring early CRRT. Of note these were almost all surgery patients but the mortality benefit was big enough to attract attention with the idea that getting in early with CRRT was key.

Next enter the IDEAL-ICU trial by Barbar in 2018 which is yet another French multi centre RCT which randomised septic shock patients to CRRT or not at 12 hrs after meeting the F on the RIFLE criteria. They wanted 800 pts and got 500 pts and abandoned it due to futility with no difference between the groups. Of note again almost half in the delayed group did not end up getting CRRT.

Now enter the START-AKI trial, the most recent and potentially practice changing study in the field. This is one of the current generation of ICU mega trials where all the great and good critical trials groups get together and put out a trial to end all trials. They randomised at 170 different sites, and included patients with a stage 2 or 3 AKI to either immediate CRRT or wait till a traditional indication developed. This trial took in 3000 patients and found no benefit of the early CRRT but again noted that 40% in the delayed group never needed CRRT.

Putting this altogether it seems that we probably pull the trigger on CRRT a little early. I know when i see someone on 2 pressors that are both escalating and on stress dose steroids and not a drop of urine in 6 hrs I feel I can predict the renal trajectory pretty well and so why not just bite the bullet and get started. But it seems that we’re not that great at predicting who needs CRRT in the earlier stages and we would probably be better waiting a little longer. Which is probably good for patients and for systems overall given that CRRT is one of the most expensive and labour intensive interventions we do in the ICU.

Finally just to cover it – there are a few things that might help us know when to stop the CRRT. If we can get a creatinine clearance of some kind (even a 2 hr one) then a clearance of >15ml/min is useful as a target. A 24 hr urine output of 400mls is a number to remember but this should be >2000mls in 24 hrs if you’re using diuretics. Obviously it’s good if the actual indication for CRRT such as volume overload has been corrected before you think about taking down your filter.

References:

Oh Chapter 48

LITFL Indications for CRRT

The furosemide stress test

• Chawla, L. S. et al. Development and Standardization of a Furosemide Stress Test to Predict the Severity of Acute Kidney Injury. Crit Care 17, R207 (2013).
• Rewa, O. G. et al. The furosemide stress test for prediction of worsening acute kidney injury in critically ill patients: A multicenter, prospective, observational study. J Crit Care 52, 109–114 (2019).

AKIKI Trial: Gaudry, S. et al. Initiation Strategies for Renal-Replacement Therapy in the Intensive Care Unit. New Engl J Medicine 375, 122–133 (2016).

ELAIN Trial: Zarbock, A. et al. Effect of Early vs Delayed Initiation of Renal Replacement Therapy on Mortality in Critically Ill Patients With Acute Kidney Injury: The ELAIN Randomized Clinical Trial. Jama 315, 2190 (2016).

START AKI: Investigators, T. S.-A. Timing of Initiation of Renal-Replacement Therapy in Acute Kidney Injury. New Engl J Med 383, 240–251 (2020).

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

If you were tuning in to try and pick up some core exam level content then this is probably not it. However, it is what came up in my non-random, semi alphabetical by system, trawl through my notes that forms the basis for the order I write these in.

We’re going to spend 5 minutes on chemo agents in the ICU. Roughly hewn from the stone tablet of Oh Chapter 46 on solid tumours in intensive care.

This is perhaps not something that will be top of the list in exams and certainly for clinical practice you are not going to be prescribing any of these agents. However a passing familiarity with some of the commoner ones, (or at least the ones that are more likely to run the risk of an ICU admission) is probably worth while.

Table 46.1 has a table that covers two pages of scrolling on my browser and includes many drugs that I have never heard of or can’t pronounce or both. Thankfully the next segment has a much smaller selection of “specific chemo induced toxicities” and we’ll try and cover at least that much today.

First up – bleomycin lung injury. Bleomycin is actually an antimicrobial used in a variety of head and neck and gynae tumours and Hodgkins. The pneumonitis can occur in up to 40% (though the corresponding up to date article puts it at more like 15%) and can be fatal. There is generation of oxygen free radicals with corresponding fibrosing alveolitis that is actually worsened with oxygen therapy. If this sounds familiar, then the insecticide paraquat does a very similar thing but is, I suspect, much less useful in treating cancer. In terms of treatment it seems that just like in ARDS, the terms fibrosing alveolitis and high dose pulsed methyl pred are inexplicably connected.

Second on our list is ifosfamide related neurotoxicity. Ifosfamide is an alkylating agent used in a broad range of tumours and is well known for causing an encephalopathy in 10-20% of patients. It is a diagnosis of context and exclusion. I have heard it discussed, if not diagnosed on several occasions by oncologists with patients in the ICU and one was even given methylene blue which is a somewhat established antidote of sorts through some mechanism of MAO activity. However it turns out that guidelines from the European Society for Medical Oncology specifically recommend against giving it so maybe forget i just said that and instead cite them in opposition to any oncologists wandering into your ICU with little blue vials.

Coming in at number 3 is anthracycline related cardiomyopathy. There are a number of drugs in this category all ending in rubicin and seem to be used fairly widely. There are two parts to this. Firstly, there can be an acute cardiomyopathy, sometimes with arrhythmias that happens early and secondly, a more chronic cardiomyopathy that can happen months to years after the drug has been given. The mechanism has too many proposed options to be memorable but reactive oxygen species seem to have at least some role.

Last but by no means least are the immunotherapy agents. This is chemo Jim, but not as we know it. Melanoma appears to be the poster child for the agents here. There seem to be a broad variety of agents and mechanisms that fall under the term immunotherapy but the check point inhibitors are probably the most well known. The immune system has various check points as such to stop the cellular militia getting too carried away. The check point inhibtors effectively remove these check points and let the immune system go wild on whatever foreign tumour antigen it can get its hands on. Mechanistically this is genius as a therapy but as you can imagine there can be a few issues with getting the genie back in the bottle. This can affect all kinds of body systems and is best summarised in an excellent IBCC post by Josh Farkas that you can find linked in the show notes. If this podcast does nothing but direct you to better resources then it’s 5 minutes well spent.

References:

Oh’s manual Chapter 46

UpToDate articles on the relevant toxicities

IBCC Check Point Inhibitors  

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

Today we’re looking at Oh Chapter 64 covering some of the absolute basics of pre eclampsia. ICU level pre eclampsia is rare. In Ireland most of it is managed in separate obstetric hospitals by the obstetricians and the anaesthetists. And given that the definitive treatment is removing the baby from the mother, it turns out that this will typically have been done before we even get involved. Unfortunately its rarity does not get us out of having to know it very well as it is an exam favourite.

Firstly some definitions. To have pre eclampsia you’ll need to have

• new onset of hypertension with proterinuria OR
• new onset of hypertension with end organ damage (inc foetal growth issues)

Hypertension here defined as BP >140/90 and proteinuria as >300mg/24hrs or protein +ve on a dipstick.

To have eclampsia you simply need to have seizures in addition to the above.

To summarise the cause and pathogenesis, the short answer is we don’t know. The longer answer is – we still don’t know but we have lots of science to show it and the medium length, exam appropriate answer answer that I might be able to reproduce is a straight quote from deranged physiology:

[pre eclampsia is] a systemic response to placental hypoperfusion, with increased activation of the potent vasoconstrictor endothelin-1, as well as an increased sensitivity to vasoconstrictors in general, and a down-regulation of vasodilatory mechanisms such as nitric oxide synthase.

The clinical presentation of pre-eclampsia from a critical care perspective is best split into organ systems

• Cardiovascular: hypertension is the main player here. You’ll see increased SVR also
• Neurological: early signs can be headache and visual symptoms. You will see hyperreflexia if you look for it and you probably should. But with progression you’ll start seeing seizures, cerebral oedema and even ICH
• Renal: protein loss is obvious but the fancy term to pull out is renal endotheliosis which is a form of thrombotic microangiopathy or TMA. A topic that deserves its own post.
• Haematologic: low platelets but also impaired function. This might come as part of HELLP syndrome
• Hepatic: HELLP syndrome is part of this but the most dramatic complication can be hepatic rupture which as you can imagine is somewhat bleedy

While rare you can even get pre eclampsia in the post partum patient and I always liked the quip from Mel Herbert of EMRAP fame, that the usual definitive treatment of pre eclampsia is of course to deliver the baby but if it’s post partum what are we meant to do? put it back in again?

We are unlikely to be given the job of ridding the woman of the placenta so we can instead usefully occupy ourselves with supporting the various failing organ systems. The priorities here will be

• seizure prevention and treatment.
• BP control
• maintaining placental perfusion

Seizure control here should lead to a brainstem level reflex of magnesium prescription. The indication is described as eclampsia or imminent risk of eclampsia and BP is often in the 160/110 range by this stage. Recommendations are 4g magnesium over 5 mins followed by 1g/hr as an infusion. Any further seizures can be treated with another 2g of magnesium. The NICE guidance on this has a carefully worded phrase that says “Do not use diazepam, phenytoin or other anticonvulsants as an alternative to magnesium sulfate in women with eclampsia”

In terms of a level of magnesium, you’re looking for somewhere in the 2-3.5 range which is obviously much higher than we’re used to. I was once told to just keep giving the magnesium till they’re completely areflexic and then pull back. This may well get you into trouble as the lack of reflexes will be shortly followed by some respiratory insufficiency and you might find yourself having to reach for your calcium as a reversal agent to the magnesium.

Magnesium is a well supported critical care intervention with the colossal MAGPIE trial from 2002  which was a 10000 person RCT showing definitive benefit and halving the risk of eclampsia.

The recommendations for BP control from NICE are as follows

• labetalol (oral or IV)
• nifedipine PO
• IV hydralazine

References

Oh Chapter 64

Deranged Physiology

Altman D, Carroli G, Duley L, Farrell B, Moodley J, Neilson J, Smith D; Magpie Trial Collaboration Group. Do women with pre-eclampsia, and their babies, benefit from magnesium sulphate? The Magpie Trial: a randomised placebo-controlled trial. Lancet. 2002 Jun 1;359(9321):1877-90. doi: 10.1016/s0140-6736(02)08778-0. PMID: 12057549.

 

Welcome back to the tasty morsels of critical care podcast.

Today we’ll cover some key exam content, all be it not something you’re likely to run into in the ICU too often. The thyroid is a deceptive little organ, tucked in the neck, quietly secreting hormones and interfering in negative feedback loops. It usually restricts its mischief to outpatient clinics by running hot or cold on a chronic basis, occasionally hypertrophying and interfering with its more important neighbour the airway. But every now and then in a pique it decides it’s fed up of this low level mischief and uses its deeply embedded relationship with the rest of the body to wreak havoc.

We’ll split this into 2 parts, one when the thyroid goes on strike and is under active and the other when it goes bananas and secretes far too much hormone

Some basic physiology. Thyroid hormones are essential for all organ systems. The active forms are T3 and T4. T3 is generally the more active one. They are synthesised by incorporating iodine into tyrosine residues in thyroglobulin in the thyroid gland. Hence how iodine deficiency can cause a deficit in thyroid hromone. Their release into the circulation is stimulated by TSH. TSH causes endocytosis of this thyroglobulin into the follicular cells where they undergo hydrolysis into T3 and T4 which is released into the circulation. Both are highly protein bound to thyroid binding globulin.

Our first relevant condition is the wonderfully named thyroid storm. Most commonly you might see this as part of untreated Grave’s disease. It can be precipitated by the usual physiological stressors such as surgery or sepsis etc…

Expect to see (at least in an exam scenario)

• fever
• tachycardia or fast AF
• jaundice
• delirium
• heart failure
• eye signs or a goitre consistent with thyroid disease

For awareness there is a clinical prediction tool that rejoices in the name Burch-Wartofsky Point Scale. This includes most of the features listed above. It’s clear that the features listed above are fairly non specific and like always it’s likely just sepsis. But if something in the spidey sense tingles then finding undetectable TSH and high T3 or T4 should really get you going. In reality this is an incredibly rare diagnosis, one which in its fulminant form i have yet to see. Or perhaps more accurately one that i have failed to diagnose as yet. This is of course hardly surprising as it is hopefully clear by now on this podcast that I am not especially good at what i do and continue to put my appointment to my current job down as some kind of administrative error that is yet to be detected.

Once you’ve decided you’ve made the diagnosis then you’ll need a few basic principles of treatment. Firstly do a bit of resuscitation. There may well be some co existing sepsis so give some antibiotics. If they’re hypoxic give some oxygen. They may need some fluid or indeed they may be in congestive heart failure. The key is to do an assessment, this likely includes having a sneaky peak at the heart and the lungs with ultrasound. A commonly recommended treatment is propanolol to help with the tachycardia. Many patients will be hyperdynamic and tachycardic and giving a beta blocker may well be a good idea but giving a negative inotrope to someone who’s heart is a bit clapped out is generally considered bad form. The key message is to assess comprehensively and then decide.

For specific therapies, your list should include some steroids, this reduces the release of thyroid hormone from the gland. There is occasionally some coexisting adrenal insufficiency so you’ll treat that as well.  You’ll need to use something like PTU (propylthiouracil) or carbimazole in order to block new production of thyroid hormone. Good luck finding PTU at 3am. Having performed one miracle in locating PTU you are now expected to perform a further miracle and find something that sounds more like a tonic you’d buy from a wild west apothecary. This is of course “Lugol’s Solution”. Only give this once thyroid production has been blocked (recommendations suggest an hour afterward). It contains typically a bunch of iodine and will block the release of any T3/4 left in the gland.

Your next patient in your exam viva comes from the opposite end of the spectrum. Myxoedema coma. As an aside, myxoedema coma is a terrible name, the patient may not have oedema or be in a coma. Again, Farkas on the IBCC uses the term “decompensated hypothyroidism” which i think is much more descriptive and accurate. This is hypothyroidism but not as you’ve seen it before. Typical features or as Josh Farkas calls them “cognitive triggers” to consider myxoedema coma include

• neuromusclular features like reduced consciousness, delirium, slow reflexes and weakness
• hyothermia. the classic is someone found unresponsive with a much lower temp than expected for the environmental conditions
• endocrine issues like low sugars and low sodiums
• cardiovascular features such as bradycardia, hypotension and pericardial effusion
• respiratory features like alkalosis and pleural effusions
• GI issues like ileus and weight gain

Again lots of these are non specific so keep the differential broad before anchoring too early. As expected TFTs will be helpful and in general expect to find a high TSH and low T3/4

Management will involve your usual assessment and resuscitation but the specific therapy here is IV thyroid hormone. It does exist but is also hard to track down at 3am. Your ICU probably has it and you’ve typically only seen it as part of management of a potential organ donor in brain death. T4 is the one typically recommended (which will converted intracellularly to the more active T3). But in Ireland the most commonly available will be T3.

These people also need steroid, typically some hydrocortisone for the ubiquitous adrenal insufficiency. Indeed giving thyroid hormone without steroid may cause an adrenal crisis.

Reading:

Oh Chapter 61

Tasty Morsels of EM 130

The IBCC

Deranged Physiology

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

Today’s podcast is mostly taken from Oh’s manual Chapter 96 covering critical care nutrition. Something we all know and love deeply for exams and then immediately outsource to our dieticians the moment we get the chance.

Perhaps the first take home figure is 25kCal/kg/day. This has been around as a recommended energy intake since the late 90s. And like most nutritional things it’s not exactly stellar in its support from the literature. There has been a recent tread in trials towards hypocaloric feeding that have hinted towards benefit at feeding to a lower target with purported mechanisms including that higher targets suppress autophagy which is an important part of killing nasty organisms. The lower target has not panned out as yet but it is worth pointing out that we rarely actually achieve the 25kCal/kg/day that we aim for, so we are probably inadvertently underfeeding people at baseline. On the other side of this it’s clear that if we have prolonged periods of not meeting targets then patients clearly do worse.

We could, of course, measure the energy requirements the patient needs instead of a blanket recommendation for all. And there are a variety of methods for doing this. Oh describes indirect calorimetry as “a rather burdensome gold standard” and it seems all the units I have worked in have taken this on board and simply not bothered with getting the metabolic cart needed for doing it. This device, when connected to the vent allows measurement of O2 consumption and CO2 production, and can calculate energy expenditure.

The other options for estimating energy expenditure include a bunch of equations of which the Penn-State one is recommended as the best of a bad bunch and finally you can calculate using the reverse Fick method which needs a PA catheter to measure O2 consumption. I mention these only as useful options for a candidate to scribble down in answer to a question rather than perform them in real life.

As the good book says, man shall not live by kCal alone so we need to consider what else the patient needs nutritionally. First off – protein. For normal people walking around their everyday business not attached to pressors or a ventilator the daily protein intake is around 0.8g/kg/day. In the critically ill this has been bumped up to at least 1.2g/kg/day or even has high 2g/kg/day in some of the super catabolic patients such as major burns.

For bonus points remember that patients require all the micronutrients such as vitamins, thiamine and elements that come as part of a healthy diet. Most enteral formulations will contain these and it’s really in those on parenteral nutrition that you need to stress a little more about it.

When calculating the energy intake we need to consider all the sources of intake. Energy is not just glucose but will include the protein that we give and also some of the infusions we use for example the fats in the propofol infusion (which comes in at ~1kCal/ml) and the glucose in the 5% dextrose we’re giving to correct the hypernatraemia.

References

Oh’s Manual Chapter 96

Deranged Physiology – there is a lovely section on the theoretical maximum and minimum amounts of each class needed including some lovely stuff on inuits and zero carb diets

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

GBS management can be nicely split into disease specific management and ICU supportive care.

In terms of specific treatment this is something that has a reasonable evidentiary basis at this stage. The menu on offer is similar in appearance to many of the immune mediated disasters that appear in the ICU with steroids, IVIG, plasmapheresis and more recently mabs have all been studied.

IVIG is supported by a Cochrane review of randomised data and a major neuro society guideline, which is generally enough to persuade a non specialist like me mainly because I’m a very lazy man… The theory here is the usual hand wavy IVIG one – that within the massive of pool of donors (literally 1000s) that go into a bottle of IVIG there is sufficient neutralising antibody against the dodgy ones the body has produced along with a smorgasborad of other immune modulating effects we don’t really understand. Either way it works and seems to be fairly benign, if not a touch hard on the wallet.

Plasmapheresis (which involves spinning the blood in a big machine until the plasma separates out, then removing all the plasma and replacing the volume with some 5% albumin and letting the body reproduce all the other plasma components itself) is another treatment supported by a 2012 Cochrane review of randomised trials. The mechanism here is thought to be removal of the circulating antibody that is causing all the trouble. It also takes away all the associated complement that goes with it. Plasmapheresis is often delivered in 4-6 sessions. This had always confused me – why so many sessions? Is there ongoing antibody production needing treatment sessions?  The number of sessions primarily depends on the antibody type you’re going after. IgM typically spends all its time in the vascular space so after a session or two you’ve got rid of it all. IgG (which is probably what we’re going after here) has a larger Vd as such and as soon as you’ve washed it out of the vascular space it leaks back in again and plasma levels rise again. Hence the multiple sessions.

Presumably in GBS there is no ongoing antibody production as it is one of the few times in ICU plasmapheresis that you don’t need a heavy duty immunosuppressant to go along with it to stop ongoing antibody production. With regards to immunosuppression it is important to note that steroids do not work and should not be used.

The question at this stage remains as to which expensive treatment you should choose. There does not seem to be an advantage of one over the other so I think I would go for the one that does not need a 12F catheter in my neck.

Turning to supportive treatment there are a few important principles at play. Firstly is the question of when do we intubate these folk. This can be split into two parts

1) airway protection. Especially in the Miller-fisher type or if you start seeing any bulbar issues. It ould be a real shame to start a 6 week ventilation run with a bad and preventable aspiration.

2) respiratory failure. This is probably more at the fore front of our minds. Remember this is a very quiet and non distressed type of respiratory failure (unless cough goes then it’s a noisy rattly mess) where there simply isn’t any respiratory effort to increase the work of breathing. There are a few published indications for intubation and perhaps the most useful and oft quoted for exam purposes is a VC of <15ml/kg. In reality we’ll probably intubate along our usual guidance which is whenever we damn well want, especially if another team makes a polite suggestion that given the VC is less than 15ml/kg maybe we should possibly just maybe consider taking them to the ICU.

Given that the nadir of the illness is probably a month away it would seem reasonable to tube and trache and wake them up as soon as possible

A question on supportive care for GBS could be filled with FASTHUGS IN BED but a useful point to make is to look for autonomic dysfunction affecting about 10% of GBS with simple things like hypertension but it can be profound enough to need an emergent pacing wire so be prepared.

While the whole disease seems horrific it’s worth noting that 70% are independent at a year.

References:

Oh’s Manual Chapter 58

Both Deranged Physiology and LITFL have nice tables comparing GBS with other neuromuscular disorders. A favourite exam question.

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

GBS is a clinically important diagnosis for both the emergency department and the ICU. Its rareish but common enough that you will at least suspect it often enough in the ED and many ICUs will have one long stay ICU patient with GBS every few years and a short stay one more often. It has the right mix of just enough clinical findings to diagnose from clinical exam but also has some novel test characteristics and some lovely management options and discussion points making it all very examinable.

The pathophys I was taught was that of molecular mimicry. Some foreign antigen is introduced and has a structure similar enough to the wiring insulation wrapping our nerves that the antibody produced binds not only the foreign antigen but also the native nervous tissue. This leaves the myelin and axons suspect to the full rigors of an immune system crackdown akin to the US Marshall’s office on a man hunt for Richard Kimble

Some of these pathogenic antibodies even have complicated named acronyms. Neutralising such antibodies with something like IVIG or washing it out of the blood with plasmapheresis forms the basis of treatment of the disease. The antigen associated with GBS is classically c jejuni (which probably carries a worse prognosis than other precipitants) but flu, HIV and mycoplasma are also on the list as alternatives. Vaccination was always on my list of precipitants but this seems somewhat in doubt. Given how common vaccinations are and how rare GBS is its tricky to separate out causation from background rate. Indeed the risk of GBS from the flu itself is well in excess of any potential rate with the flu vaccine.

The breakdown of the blood-nerve barrier at the point the peripheral nerves enter and leave the spinal canal allows blood plasma proteins to enter the CSF which is the rationale for the high protein on LP.

To make what was already a complex disease more complicated it is sub divided into 4 subtypes.

• AIDP (acute inflammatory demyleniating polyneuropathy)
• MFV (miller fisher variant)
• AMAN (acute motor axonal neuropathy)
• AMSAN (acute motor and sensory axonal neuropathy)

(the last two being significant for being an axonopathy as opposed to demyelination)

Now that you have heard of them you can reproduce them for exams and then forget them for clinical work as AIDP is by far most common and we treat it largely the same even if the prognosis varies from the other types.

Most of the diagnosis here is clinical with not only the history but real life bonafide examination findings actually being a useful part of the diagnosis. In the ICU the diagnosis is usually fairly easy as they come with a label of possible GBS in the referral. The ED diagnosis is much more of a challenge as they present often before the textbook findings are there. Pain in the legs with an odd or supported gait has been a common presentation in my limited experience. The weakness in the legs may be subtle at this point and often dismissed  by idiots like myself as secondary to pain. The real challenge in the ED lies in locating a tendon hammer, a somewhat mythical device in the ED. The much talked about and rarely seen Miller-Fisher variant has more dominant cranial nerves findings

So once we have identified ascending weakness and demonstrated reduced reflexes the next test we reach for is typically the LP. As mentioned above, look for a high protein and normal cell count. This finding is termed albuminocytological dissociation which was a key part in the discovery and labelling of the illness in the early 20th century. The finding is present in a deeply underwhelming 66% of those with GBS so it’s practically of no use if there’s an entirely normal LP as your clinical findings will trump everything and they may well still have GBS. That being said I’m sure neurology will find lots of random antibodies to send on the CSF so it’s still worth sending. The presence of protein increases over the first couple of weeks so a repeat LP might well be useful.

The twitchy electrode needle type tests do have specific findings depending on the type of GBS you’re dealing with but I struggle to see any relevance to what we do in critical care

Next time we’ll cover management.

References:

Oh’s Manual Chapter 58

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

We see C. Diff in the ICU in a couple of contexts. Firstly the poor unfortunate soul who starts with a benign illness and gets some antibiotics and develops a fulminant colitis and shock needing colectomy and an ICU admission. Secondly we have the frequent dilemma of the prolonged ICU patient who is collecting complications like they’re merit badges. They’ve developed new shock and there’s some diarrhoea and you’re worried about c. diff.

First off some risk factors. Firstly is antibiotic exposure and topping the list would be beta-lactams and clindamycin, closely followed by the quinolones, aztreonam, and the carbapenems.

A truncated list of patient specific risk factors might go as follows:

• age >65y
• previous GI surgery or IBD
• renal impairment
• prolonged hospitalisation esp with a C. diff contact
• gastric acid suppression (though PEPTIC trial would suggest against this is unlikely, or at least not more likely than H2RA. However I still think it’s a fair answer in an exam)
• immunocompromised and chemotherapy patients

There are a few useful principles of prevention:

• Hand washing – remember that spores survive alcohol and on surfaces, need soap and water
• Isolation
• Testing of symptomatic individuals
• avoidance/minimising antibiotics
• Limit PPI use and get NG tubes out ASAP

In terms of diagnosis we’re typically thinking of this in patients with diarrhoea. We send a stool sample and wait patiently by the computer for a result. But what are we actually testing? This will of course depend somewhat on your local lab but there are a few important principles. Most of the time we’re testing for a toxin produced by the bacterium. This can be toxin A or B or both. If both are +ve then cool cool cool you’ve probably got your diagnosis. If you’ve got one +ve and one -ve then IDSA recommends using nucleic acid amplification (or PCR if you like) as a tie breaker. The PCR is very very sensitive. And the issue is that the presence of c diff DNA in your poo doesn’t necessarily imply disease whereas the combination of symptoms a +ve toxin immunoassay and a PCR is much more compelling.

An important component of the diagnosis involves some form of mucosal assessment – especially in more severe cases. This can be some form of flexible boweloscopy device or a CT scan looking for colitis or mega colon.

Once you’ve made your diagnosis you’re ready to start some treatment. This can be very confusing as treatment is not only determined by severity grade but the recommendations have changed relatively recently (in 2018) so the stuff you learnt for your membership examinations may no longer be relevant.

The split here is into non severe, severe and fulminant so keep that in mind.

The split between non severe and severe, is defined by a WCC<15 and a normalish creatinine. This distinction would seem an important thing to memorise, however as far as I can work out, IDSA suggests identical treatment for both non severe and severe. Namely

• Vance 125mg PO QDS OR fidoxamicin
• note that metronidazole PO is no longer recommended the way it was in the mid noughties when I started.

Fulminant disease is primarily determined by shock and the presence of megacolon. This probably is something worth remembering as treatment here is significantly different. Here you give vanc 500mg PO QDS plus metro IV (which after being given IV is secreted unchanged into the GI tract via the biliary tract). For those with nasty distal disease vanc enemas can also be used which will certainly make you super popular with the nursing staff.

While rarely done, it’s really important in both clinical practice and certainly in an exam to consider taking out the whole colon. The surgeons will of course make the final decision on this but some may need a little nudge in the right direction. Potential nudges might include

• Hypotension requiring vasopressor therapy
• Clinical signs of sepsis and organ dysfunction
• WCC in excess of 50
• Lactate in excess of 5mmol/L
• Failure to improve on medical therapy after 5 days

If you’ve managed to get all this across in your SAQ then you’re flying but if you want to go for gold then you could throw in a few comments like

• faecal transplant – niche and not really considered until 3rd recurrence
• IVIG as a toxin binding agent in fulminant (no substantial support)
• there is a monoclonal antibody to bind toxin but this is really sketchy at this point

References and rationalisations

Deranged Physiology

LITFL

IDSA 2018