Tasty Morsels of Critical Care

Welcome back to the tasty morsels of critical care podcast. Thanks to the wonders of preemptive rasburicase (which i have a tendency to associate with Rasputin at this point) I don’t think this as common as it used to be. But that is wild speculation on my part as someone who’s only ever looked after cases in the single digits. The basic premise here goes something like this. Chemotherapy does exactly what it’s meant to do and kills tumour cells. These cells break down releasing amongst other things the purines that make up various components of the cells. Our bodies have a well established pathway for dealing with an influx of purines and metabolises them to uric acid. Uric acid is weakly soluble and excreted at a certain rate by the kidneys. The problem comes in that the metabolic processes for converting purines to uric acid is much more efficient than the process to remove uric acid from the body. Phenomena like this are widespread in the body and put to good use with the long history of happy hours and shots as a means to bring about inebriation before the pesky liver can catch up. Interestingly primates are unique in not being able to convert uric acid to the harmless compound allantoin. It’s hard to see what evolutionary advantage our ancient ancestors gained in dropping the enzyme needed but then I can’t imagine beer and steaks were big on the diet of the average primate at that stage. So in TLS we have production of uric acid far in excess of our ability to excrete. The uric acid then tends to precipitate out of solution into crystals at this stage that have a tendency to crystallise out in the a) joints and b) renal tubules. We now have a combination of failing kidneys, oliguria, hyperkalaemia from the worsening acidosis and the K release from dying cells. Enough of the pathophys, what context do we see this in Large tumour mass or bone marrow involvement burkitt’s ALL/AML Small cell lung cancer also appears commonly on lists. There is something called the Cairo definition that can be used to diagnose it. I’ll spare the details but nice to know it exists. Obviously it’s common in the days post chemo but it can actually happen spontaneously. I mentioned rasburicase earlier because this has changed things significantly.  When we give rasburicase we’re providing that enzyme that all the other mammals have that we dropped. This metabolises uric acid to the fairly harmless allantoin, eliminating our dependance on the glacial process of excretion. This drug can be given prophylactically in the right context so we never see the TLS develop. The other drug you may remember about uric acid from med school days is allopurinol. This blocks production of uric acid (surely a good thing I hear you cry) but merely diverts it into the production of xanthine, a somewhat problematic compound that we’re also not particularly well adapted to get rid of. Rasburicase is  a recombinant product that seems fungal in origin (so you can make antibodies against it) and also is a reported cause of MetHb. It’s pricy but so is €2000/day in ICU on CRRT. Of note CRRT will fix all the electrolyte and metabolic issues that are seen in TLS and is a good option for treatment. References and rationalisations: Deranged Physiology IBCC

Welcome back to the tasty morsels of critical care podcast. Critical illness in pregnancy is remarkably rare given the somewhat bonkers system for reproduction that we seem to have evolved over the past million or years. Improved care of complex cardiac conditions has led to improved survival of women with an ability to procreate and pose what can only be described as challenging conditions for their treating physicians. What can we expect to see in terms of cardiac disease? Well, most commonly we’re going to see ACS. As the pregnant population continues to age and with RF and smoking and DM still common we can expect to see pregnant woman with CAD. Pregnancy itself with its bump in plasma volume, reduced Hb and increased cardiac output is like one long exercise stress test. Some unique features to consider in pregnancy is spontaneous coronary artery dissections. This can occur outside of pregnancy but due to collagen softening effect of the hormones it can occur throughout pregnancy. While interesting and exam worthy you’re still going to see STEMI on the ECH and have to go to the cath lab so perhaps it doesn’t change much of what we do. Valve disease provides the most interesting part of obstetric care. Thankfully this is typically known in advance and can be somewhat planned for. There is a WHO/ESC classification of cardiac disease from 1-4 (with 1 being no detectable increase risk to 4 being pregnancy contraindicated). There’s a lot to a classification like this so I’ve tried to summarise it to the following points. You can expect the worst outcomes in Stenotic lesions Pulmonary hypertension Eisenmenger’s (an irreversible pulmonary vascular disorder secondary to congenital heart disease) Mechanical valves Severe LV dysfunction (or a prior peripartum cardiomyopathy) Aortas so big you could drive a bus through them References: Oh’s Manual Chapter 66 ESC Guidelines on the management of cardiovascular diseases during pregnancy  

Welcome back to the tasty morsels of critical care podcast. As an EM trainee doing intensive care I will confess that I struggle to work up the enthusiasm to cover a nutrition guideline. With my deepest apologies to all the great ICU dieticians I have worked with, I just struggle to get excited over this stuff. Part of the reason for this is that our dieticians are so good I have cognitively outsourced the nutrition to them and have scraped together a knowledge basic enough to get me through the exam. All that being said it’s worth being aware that there are some major pieces of guidance out there and you would do well to have a passing correspondence with them. I would highly recommend attending an excellent lecture given by your ICU dietician which is just possibly where this entire summary came from. This is certainly not intended to be a comprehensive review of ICU nutrition but a few pearls from the guideline. Firstly nutritional assessment. This is the first step. ESPEN immediately embarks on a new cold war with their American colleagues by rejecting something like the NUTRIC score as definition of critical illness malnutrition stating nothing has been particularly well validated as yet. They do stress the importance of assessing for malnutrition with some risk factors including BMI<20>40 malnourished burns sepsis prolonged stay frailty Depending on definition ~50% might be malnourished in ICU They helpfully split critical illness into phases with regards to nutrition, which is probably the most useful thing I took away from the guideline. The acute phase has an early period of instability and pressor use. Usually ~ 1-2 days when the body is very catabolic This is followed by a late period ~ day 3-7 where there is muscle wasting and stabilisation of metabolic abnormalities. Finally there is the late phase/chronic/rehab as the patient becomes more anabolic According to this concept don’t rush to feed new admissions with increasing pressors or acute bleeding or life threatening hypoxia/acidosis. As with all ICU nutrition EN is preferred over PN. There are a few other points to take away from the guidance Refeeding syndrome ~35% of ICU patients at risk of this drop in PO4 after starting feeds (drop of 0.65) prevention is typically IV or enteral thiamine and reducing rate of feeding. Also important to remember other sources of kCal in the ICU propofol dextrose citrate References: Singer P, Blaser AR, Berger MM, Alhazzani W, Calder PC, Casaer MP, Hiesmayr M, Mayer K, Montejo JC, Pichard C, Preiser JC, van Zanten ARH, Oczkowski S, Szczeklik W, Bischoff SC. ESPEN guideline on clinical nutrition in the intensive care unit. Clin Nutr. 2019 Feb;38(1):48-79. doi: 10.1016/j.clnu.2018.08.037. Epub 2018 Sep 29. PMID: 30348463.  

Welcome back to the tasty morsels of critical care podcast. Just to be clear up front we’re going to be talking about taking the lid off for stroke, NOT taking the lid off for TBI. I will get to TBI but they are really quite different topics. This is not something that is done commonly but if you work in a neuro centre you will see it happen. We’re talking about the malignant MCA syndrome here. This is somewhat poorly defined but should reflect a large area of ischaemic tissue in the MCA distribution. The diagnosis of stroke here is not going to be subtle and expect at presentation a patient with a dense hemiplegia and almost definitely some early signs of stroke on CT that will rapidly develop.  These people do not have malignant MCA syndrome at presentation (so this is not really an ED diagnosis) but large areas of ischaemic tissue over time tends to do what it does best – which is – swell. Consciousness will become impaired when it swells and remember that impaired consciousness is unusual in most strokes. The swelling is where the real badness comes in. As Monroe and Kellie have taught us the skull is a fairly fixed box and once you’ve squeezed out the venous blood and the CSF then the pressure rises very fast and bits of brain try to squeeze through orifices amongst the dura they were never designed to cross. Death follows rapidly. You can prevent this brain shifting, herniating phenomenon by taking the lid off and allowing the MCA territory of the parenchyma to expand out of the skull like bread dough rising of a tin during a nice long rest. There is a surprisingly good evidence basis for this intervention with three trials all published within a few years of each other in what seemed to be a skull removing replay of any of the Football European Championships of recent years. Brief summaries follow: DESTINY 2007 this was a German multicentre RCT including 30 pts there was a 90% v 50% survival favouring decompression DECIMAL 2007 this was a French multicentre RCT including 38 pts they looked at the mRS as the primary outcome and found that those with decompression had much better neuro outcomes than those who didn’t but this was almost all due to the fact that those who didn’t get decompressed died in massive numbers as an example this trial found a 50% absolute risk reduction in mortality HAMLET this was a dutch multicentre RCT including 64 pts. The difference here was that they allowed up to 4 days to have their surgery with prior trials only allowing up to 2 days they found that it still reduced mortality (again by ~50% absolute) but when done late did not seem to reduce long term morbidity Who should you do this in? This is by no means a straight forward question but some useful numbers to take home would be  <60 yrs old within 48 hrs of onset of the stroke other things to consider include lives are clearly saved saved but almost everyone has moderate/severe disability following it the craniectomy must be at least as big as the margins of the stroke References: LITFL Tasty Morsels of EM 089 DESTINTY DECIMAL HAMLET  

Welcome back to the tasty morsels of critical care podcast. Pneumocystis Jriovecii Pneumonia, the infection formerly known as Pneumocystis Carinii Pneumonia The official change in name from Carinii to Jirovecii was in the late 1990s to emphasise the distinct organism that infected humans and named after a chap called Otto Jirovec. it seems that there was just one “i” when the name first changed as Pneumocystis was at that time thought to be a protozoa. once conclusively shown to be a fungal infection heavyweights from the International Code of Nomenclature for algae, fungi, and plants (the ICNafp) weighed in and it became officially Jirovecii with “i”s at the end. All of this is from what I picked up from Wikipedia on a Sunday afternoon. The main use of this information will come when some smug consultant corrects one of the team when they say PCP instead of PJP and then to rescuse the juniors from their shame you can bust out the ICNafp declaration on the two “i”s and arise victorious over said smug consultant. At least that’s the way it went in my head anyhow. Back to some actual medicine then. You can split pneumocystis into two contexts clinically the poorly controlled or first presentation HIV patient these guys tend to have a fairly slow and insidious presentation of breathlessness over weeks but not severe enough to present to hospital when they do present expect hypoxia and probable CXR changes a small number will have PTX in this group probably have a higher burden or organism overall the immunosuppressed patient this could be steroids, solid organ transplant or a haem malignancy often more like a typical acute respiratory infection with fever, cough, SOB Both groups will have the now infamous ground glass appearances on imaging. Diagnosis can be tricky and from an ICU perspective they’re probably sick enough to justify empiric treatment based on context without waiting for all the results. Usual stuff like CXR, CT and labs yadda yadda yadda… Commonly we order BDG (a common component of fungal organisms) but it’s not clear how good a test it is to rule in or rule out. LDH, while commonly ordered is even less useful. A BAL is probably going to be needed and on this you have a few options silver staining immunofluresence (which is probably most sensitive) PCR (which is good but can’t distinguish carriage from infection. however if -ve then you’re probably in the clear) There’s a  reasonable guideline from the ATS in 2019 which covers this and is worth a read Management wise the key spinal level reflex when this is considered is cotrimoxazole. This should be at high dose which you should check in a book or with a pharmacist. While pneumocystis is fungal we can’t use our usual anti-fungals like echinocandins or azoles (the caspofungins and fluconazoles) as pneumocystis has cholesterol in its walls and not ergosterol. For those intolerant of cotrimoxazole some words to consider mentioning in an exam setting would be clindamycin, primaquin and pentamidine but in reality there is expertise and pubmed to help with that question. If they’re in ICU then the cotrimoxazole should be given with a chaser of steroids as this has a mortality altering effect. References: Fishman JA, Gans H; AST Infectious Diseases Community of Practice. Pneumocystis jiroveci in solid organ transplantation: Guidelines from the American Society of Transplantation Infectious Diseases Community of Practice. Clin Transplant. 2019 Sep;33(9):e13587. doi: 10.1111/ctr.13587. Epub 2019 Jul 1. PMID: 31077616. Oh’s Manual Chapter 73 LITFL CCC IBCC      

Welcome back to the tasty morsels of critical care podcast. Today we’re talking about chylothorax, not something I’ve not had the joy of managing as yet. Chyle is produced through digestion and absorption of emulsified fats from the small intestine. It consists of triglycerides, white cells, and these wonderful things called chylomicrons. Chyle is not generally meant to be in the thorax unless it’s simply passing through in the fragile (and frequently mistaken by surgeons) thoracic duct. The chyle makes it way up the thoracic duct to drain into the venous system. This is all well and good till the thoracic duct is standing on the street corner minding its own business, when it is injured by any number of causes such as: surgery (though I’m sure they’ll blame anaesthesia) malignancy trauma infection (filiariasis and TB) sarcoid amyloid Clinically you would expect to see a pleural effusion (mainly unilateral) with milky white pleural fluid coming out of it if you decided to drain it. In terms of the timescale related to surgery it’s often a few days lag before the effusion is clinically apparent. From an exam point of view you might be asked how you might distinguish this white milky fluid from some propofol that had been infusing into the chest for a few days from a misplaced central line (true story!). You could say there would be high triglycerides which would be true but wouldn’t help you to distinguish it from the milk of amnesia, propofol. Trigs can also be low in a fasted patient which these guys suffering from the complications of surgical misadventures often are. The best answer is to get your lab to do electrophoresis on the fluid looking for chylomicrons. Management wise it seems that if there’s a surgical cause then non operative management is an option and if it’s a non operative cause then surgery might be your fix. There does appear to be a broad variety of surgical and interventional options many of which seem to be basically tying the ends off. From an ICU perspective we’re going to be more interested in the non surgical options, which are variations on a theme of reducing chyle flow through the injured duct. You can do this by altering their nutrition to focus on medium chain fatty acids which are apparently easier absorbed through the portal venous system and don’t need the thoracic duct. You could skip the gut completely and use TPN. Finally you could try our all purpose secretion drier upper – octreotide. As you can imagine for a condition so rare there is not a great deal of high level data so don’t be surprised if you don’t see a consistent approach. References: LITFL CCC Deranged Physiology  

Welcome back to the tasty morsels of critical care podcast. One would think that the term drowning should be relatively straightforward, however there have been a plethora of descriptions including wet drowning, dry drowning and near drowning. These have all been officially retired and the Utstein guidance on the matter describes drowning as “a process resulting in primary respiratory impairment from submersion or immersion in a liquid medium”. If you happen to have been immersed in a liquid medium with respiratory impairment but survived then the term non fatal drowning seems to be acceptable. As a brief tangent, given that this is going to be a shorter than usual podcast, the “Utstein  guidelines” refer to a 1990 ESC meeting regarding cardiac arrest at Utstein Abbey on the island of Klosteroy in Norway. Knowing that the significant output of Augustinian Abbeys is often excellent beer then I suspect that members of the ESC committee probably spent quite a lot of time immersed in a liquid medium themselves. More usefully the pathophysiology of drowning could be described as follows: when under the water we sensibly engage in some voluntary breath holding, this in combination with the cold induces a reflex bradycardia and vasoconstriction As those of us who have attempted to drown infants as medical treatment treated SVT in babies will attest to, this reflex is much stronger in tinies than adults The profound reflex in kiddos may be one of the reasons they have better outcomes than adults. They get cold before they drown and get cardiac arrest which is probably protective. (as opposed to have a cardiac arrest and then get cold which is the more common mechanism) once the breath holding is broken water will hit the cords and you might get some laryngospasm but eventually water enters the lungs despite being physiologically interesting there does not seem to be any substantial difference in salt vs fresh water drowning We often get excited about the perceived filth of the water as an indication for antimicrobials but in reality all outdoor water sources that you might drown in will have bugs in it. Pneumonia will occur in about 15% and perhaps one bug to remember would be aeromonas. Treatment is pretty much as expected in a cardiac arrest scenario but it would seem prudent to pay rather more attention to the airway than than the pads. There may be a role for early bronchoscopy to remove debris or more interestingly to acquire images for your next NEJM publication showing the world’s first sea horse in a bronchus. Of note Oh’s Manual quotes a 45% survival to discharge with cardiac arrest from drowning which is markedly higher than other forms of cardiac arrest but I suspect this refers to the swimmer witnessed to get into trouble and is quickly removed to shore.   References: Oh Chapter 82 Deranged Physiology LITFL CCC

Welcome back to the tasty morsels of critical care podcast. All of these posts are simplifications of much more complex topics but when it comes to the immunosuppressants used in solid organ transplant I’m wading out of my depth even more than I usually do. This summary is basically enough to get by but perhaps not enough to say anything sensible about. All of them are of course immunosuppressant but depending on the type of solid organ transplant may need more or less suppressing. Acute rejection in kidney transplant is a big deal but at least we have a backup organ with dialysis. Acute rejection in a liver might be easier managed but we have no backup option for support if the graft is rejected. All that to say that it’s complicated and there are reasons that transplant physicians are one of the few groups allowed to prescribe in our generally closed ICUs. There is a concept of “induction” chemo where the immune system is flattened at the time of antigen presentation basically telling the immune system “nothing to see here, move along…” while sneaking a new organ into place. This allows the delayed introduction of the supertoxic maintenance agents. Basiliximab is the one I’ve seen used as an antibody to the IL-2 receptor which I presume does something complicated to the immune system that I would understand even less than most things. The main drugs used in the long term fall into 2 groups calcineurin inhibtors such as ciclosporin and tacrolimus antimetabolites such as azathioprine or mycophenelate So here’s one sentence on each Ciclosporin found in a soil sample produced by a fungus (which is a common and somewhat inexplicable theme in transplant drugs) p450 metabolism (which is medical code for ask a pharmacist before starting a new med) with no renal excretion extensive side effects most importantly is nephrotoxicity TMA and PRES are also important ICU conditions on the list of bad things Tacrolimus less dependant on bile for gut absorption also p450 also nephrotoxic Azathioprine inhibits leucocyte function by interrupting purine synthesis dose dependant myelosuppression xanthine oxidase metabolism and renal excretion Mycophenelate again blocks purine synthesis 100% bioavailability and renal excretion mainly GI side effects marrow suppression increasingly replacing azathioprine apparently Steroids of course also have a big role by inhibiting T cell proliferation, T cell immunity and cytokines and suppress antibody formation. Usually used in high doses as part of induction and then tapered. Current idea is to try to minimise long term steroids using immunosuppressants above as lifelong steroids is a fate only matched by not having a transplant at all.  For completeness it’s worth knowing there a variety of unpronounceable biologic agents, either monoclonal or polyclonal antibodies with the main target as IL-2. IL-2 needed for proliferation of cytotoxic T cells which are key cells we’re trying to keep in a drunken stupor in the corner so they don’t notice the honking great new solid organ we just sneaked in. Basiliximab as mentioned above is an example of these drugs References: Oh Chapter 103

Welcome back to the tasty morsels of critical care podcast. DKA is bread and butter for critical care providers. In fact the combination of bread and butter in the absence of insulin is a core part of the pathophysiology of the disease. An exam worthy summary of the pathophys might go as follows. lack of insulin stimulates lipase which leads to the production of free fatty acids from lipid stores peripherally FFAs are converted to ketone bodies in the liver in the presence of excess glucagon ketones dissociate significantly at physiological pH, this sucks up the supply of base buffer in the body and eventually overwhelms it with unbuffered acid and acidaemia developing. Beta hydroxybutarate is a name of a ketone that you might want to mention if pressed to name one.  both the high sugar and high ketones lead to osmotic diuresis which results in water, sodium and other electrolyte loss.  It’s probably worth contrasting this with the pathophysiology of Hyperglycaemic Hyperosmolar State which used to rejoice in the onomatopoeic glory of HONK. In fact comparing and contrasting DKA and HHS seems to be a favourite exam question. HHS is characterised physiologically by  just enough insulin to prevent ketone generation but not enough to allow peripheral uptake slower onset and hyperosmolarity as a key feature HHS has impaired thirst mechanisms in context of hyperosmolarity hyperglycaemia itself is proinflammatory (and the higher degree of it in HHS suggests the higher rise in VTE) What criteria might we use to diagnose DKA. Ketone>3 GLucose>11 HCO3<15 +/- pH<7.3 The one niche novelty version of DKA is the much discussed and not particularly frequently seen euglycaemic DKA. This seems to be a feature of the SGLT2 inhibitors which is secondary in difficult pronounciation only to their actual generic names of the glifozins. These meds are going to become a much more common med given that the EMPEROROR and DAPA-HF trials have declared these drugs mortality reducers in heart failure even in those with out diabetes. Management here is pretty much as expected if you’ve been doctor for longer than 2 years. I would refer you to your hospital’s or a national guideline for the details. Some points for style include. given some insulin (though not too much) give some fluid (though the evil abnormal saline can cause its own issues) ketones are probably a better treatment target than the glucose level. look for some precipitants (even though the answer is always non compliance and even in those with pumps it’s non compliance as the tubing kinked somewhere and no one noticed) be careful with the K as this is probably the best way to kill them. a pH of 7.1 with a K of 3 is a real danger zone as both insulin and correction of acidosis will cause the K to drop fairly precipitously. finally on a logistic point it’s fairly common for these guys in Ireland to end up getting an art line for the frequent sampling but as one of my very grey and wise ICU bosses pointed out – a CVC would be infinitely more use than an art line in this scenario given that it gives access for bloods, multiple lumens for infusions and allows concentrated potassium correction. References: Tasty Morsels of EM 122 Oh’s Manual Chapter 59

Welcome back to the tasty morsels of critical care podcast. In the main we get excited about systolic dysfunction. We obsess over the ejection fraction with numbers like EF of 12% being reproduced recurrently in handover sheets. But this is to neglect what constitutes the vast majority of the time of cardiac cycle in those with sensible heart rates. It has become clear that a lot of cardiac dysfunction occurs during this phase of relaxing and we neglect it at our peril. In terms of terminology heart failure with preserved ejection fraction is the preferred term. This gets abbreviated to the wonderful HFpEF which always makes me think of Hufflepuff from Harry Potter. This is in contrast to HFrEF – heart failure with reduced ejection fraction. From a physiological perspective diastole goes from AV closure to MV closure. It consists of the following phases following AV closure IVRT – Iso Volumic Relaxation Time. Where the muscle is relaxing but the volume as yet remains unchanged Early diastolic filing – when the MV pops open and the suction effect of the relaxing LV pulls blood from the LA into the LV.  Diastasis – where the nothing really happens and they all pop out for a quick smoke Atrial contraction – the atria gives its puny little ‘kick’ as an attempt to justify its musculature MV closure While we talk about diastole as cardiac relaxation it is actually an active and passive process. Active relaxation begins towards the end of systole continues until end of early filling. Late filling (from end of e wave to end of a wave) is largely passive. From our point of view we’re looking for it in patients who look like they’ve got heart failure but maybe don’t have an obvious history of it or the cardiac PoCUS looks fairly normal. We should think of it more in Hypertensives Older patients Those with certain echo signs such as raised right sided pressures, big left atria and things like abnormal E/e’ (which sadly have nothing to do with optimus prime)(of note I misspoke on the audio here and talked about a high e’ being the problem – soz…) Typically the diagnosis is made with echo and there is a comprehensive guideline document endorsed by the American and European echo societies with lead author Nagueh from 2016 but increasingly it seems that many of the numbers and criteria in that document may not transfer well to the critical care population. This has relevance to the ICU population in 2 major ares Sepsis. EF does not correlate with particularly well with mortality in sepsis though diastolic dysfunction does It may be  due to under filling and the reduced diastolic filling due to the ubiquitous tachycardia in sepsis. This raises the tantalising possibility of beta blocking the patient on 40mcg/min of norad which as yet can only be described as physiologically interesting but unsupported fluid resus here is very tricky as too little bad and too much is bad with no room for error Weaning failure cardiac is a common cause of failure at extubation. You pull the tube, lose afterload reducing effect of the PEEP and your patient gets reintubated with pulmonary oedema.  it seems that B lines and filling pressure estimations with echo are more predictive of failure than systolic dysfunction or diaphragmatic dysfunction on ultrasound This is all very interesting but we are left with the typical conundrum that advancing the understanding of a particular disease process often brings about – we have no proven treatments. Beta blockers and BP control remain important in the OP population but it is difficult to know how they apply to the ventilated septic ICU patient. From physiological reasoning it may be wise to maintain sinus rhythm maintain a normal preload a low afterload and a slowish rate References and justifications: Deranged Physiology McClean et al, Advanced Critical Care Echo Sanfilippo F, Scolletta S, Morelli A, Vieillard-Baron A. Practical approach to diastolic dysfunction in light of the new guidelines and clinical applications in the operating room and in the intensive care. Ann Intensive Care. 2018 Oct 29;8(1):100. doi: 10.1186/s13613-018-0447-x. Erratum in: Ann Intensive Care. 2018 Nov 6;8(1):106. PMID: 30374644; PMCID: PMC6206316. Karrowni W, Chatterjee K. Diastolic heart failure: the current understanding and approach for management with focus on intensive care unit patients. J Intensive Care Med. 2014 May-Jun;29(3):119-27. doi: 10.1177/0885066612453131. Epub 2012 Jul 10. PMID: 22786981.