UMEM Educational Pearls - By Mark Sutherland

When managing a hypotensive patient who may have some element of cardiogenic shock, it has long been debated whether it is better to start an inodilator like dobutamine, and use a true vasopressor like norepinephrine to offset the vasodilation, or start an inopressor like epinephrine.  Currently, this is largely a practice pattern issue, with different providers and specialties tending to make different choices (in my anecdotal experience, medical intensivists tend to do norepi+dobutamine, whereas cardiac surgeons and intensivists tend to use epi).  

Banothu et al recently studied this question in children with "cold" septic shock (they do not specify how this was defined) and found quicker time to resolution of shock with norepi+dobutamine vs epinephrine.  It should be noted that this was a secondary outcome, was a small study, was in children (who I'm told are not just little adults), and no difference in mortality or patient oriented outcomes was found.  However, this is a good opportunity to review what is known on this topic:

-A small RCT in Lancet 2007 by Annane et al found no difference

-A very small RCT in Acta Pharmacologica Sinica 2002 by Zhou et al suggested norepi-dobutamine has favorable effects on gastric mucosa and tissue oxygenation relative to epi or dopamine

-A small RCT in Intensive Care Medicine 1997 similarly suggested that oxygenation in the splanchnic circulation may be better with norepi+dobut than epi.

Take Home: There is very limited evidence in either direction when choosing between an inodilator + vasopressor (e.g. norepi + dobutamine) vs single inopressor (e.g. epi) strategy for a hypotensive patient in which inotropy is desired.  There is some weak evidence that norepi + dobutamine may be better for maintaing gut oxygenation and may resolve shock faster.  Personally, I would weakly recommend norepi + dobutamine over epinephrine, but continuing to follow provider preference and go with the agent(s) you're most comfortable with is also very reasonable.  If using the inodilator/vasopressor combination, it is recommended to titrate the vasopressor (e.g. norepi) to MAP and inodilator (e.g. dobutamine) to a measure of cardiac function such as CO/CI.  

Show References

Although extubation has historically been the purview of critical care, as ED lengths of stay continue to worsen, and as we see more and more rapidly reversible respiratory failure (e.g. opioid overdose), it is valuable for ED providers to be facile in extubating patients.  In addition, a longstanding debate in critical care has revolved around the proper device to extubate patients to, specifically: regular nasal cannula (NC) vs high flow nasal cannula (HFNC) vs noninvasive positive pressure ventilation (NIPPV).  Although data are mixed, the literature suggests extubation to HFNC or NIPPV may reduce risk of reintubation, esspecially in patients at a high risk of reintubation, but doesn't show a clear difference between HFNC and NIPPV.  

Hernandez et al recently conducted an RCT in two Spanish ICUs looking at HFNC vs NIPPV upon extubation for high risk patients.  NIPPV was associated with a lower reintubation rate (23%) as opposed to HFNC (39%).  Hospital LOS was also shorted in the NIPPV group, but no other differences were observed.  

It should be noted that this study, and pretty much the entirety of this literature base, is in ICU patients.  In fact, in this study, patients were excluded if they were intubated less than 24 hours.  Generally speaking, patients with shorter intubation tend to be lower risk for reintubation and other post-extubation negative outcomes, so I would use caution extrapolating this too much to the ED.  Unfortunately however, there is very limited literature to guide ED extubation practices.  

Bottom Line:

1) Know how to assess readiness for extubation and consider extubation in the ED if they meet  criteria

2) For patients at higher risk of reintubation (older, sicker, CHF, COPD, obesity, airway issues) who you are considering extubating, you may wish to extubate them to Noninvasive Positive Pressure Ventilation, even though there is little solid literature showing best practices in terms of post-extubation respiratory support in the ED.

Show References

Needless to say, therapeutics for COVID-19 pneumonia have been controversial.  From hydroxychloroquine to ivermectin to remedesivir to steroids to bleach (sorry, but it had to be said...),  it depends on who you ask whether medications make a difference in COVID, how much of a difference, when they should be given, and what the correct dose is. 

Dexamethasone, however, ala the RECOVERY trial, is one of the relatively few therapies supported by the majority of the literature and guidelines, and generally is recommended when respiratory support is required for COVID-19 pneumonia.  Further add to this that steroids for ARDS is a long-running point of critical care controversy (e.g. DEXA-ARDS, Meduri, etc), and all you need to say to an intensivist is "how much steroid should I give this patient?" and you can walk away and come back 10 minutes later to find them having not noticed you had ever left.

Wu et all did a fairly small (n=107) single-centered RCT looking at dexamethasone 6 mg daily vs dexamethasone 20 mg daily for COVID-19 requiring O2.  There are several notable limitations to this study, but in short it did NOT add support to the notion that higher dose dexamethasone is a good thing for COVID-19 pneumonia.  In fact, the 20 mg group trended towards worse outcomes.  Small sample size, single-center, limited follow up, variable use of biologics between the groups, and failure to investigate intermediate doses between 6 and 20 are all significant limitations of this trial. Of note, DEXA-ARDS, which was conducted before COVID (2013-2018), looked at 20 mg x 5 days followed 10 mg x 5 days and DID find a significant benefit, as well as pretty darn good NNT and p values (and was a higher quality trial), so in my opinion it is also not unreasonable to use DEXA-ARDS dosing if the patient meets moderate-severe ARDS (P:F < 200) criteria, even though of course DEXA-ARDS was before COVID and Wu et al slightly contradicts it. 

When faced with a very sick COVID-19 pneumonia patients many intensivists will do either RECOVERY or DEXA-ARDS dexamethasone (with relatively limited basis to choose one vs the other), and some will do Meduri protocol methylprednisolone (1-2 mg/kg/day).  Relatively few nowadays will omit steroids unless there's a contraindication.

Bottom Line: It probably remains a good idea to give dexamethasone to your COVID-19 pneumonia patients with hypoxia, but you can probably stick to RECOVERY (see reference below; 6 mg daily x 10 days) dosing as opposed to higher doses.  If they're REALLY sick (P:F < 200), consider DEXA-ARDS (20 mg x 5 days followed by 10 mg x 5 days) dosing.

We previously posted on the COCA trial, which looked at empiric calcium administration in cardiac arrest.  They studied 391 adult Danish cardiac arrest patients.  The immediate and 30 day outcomes showed no benefit, and in fact strongly trended towards calcium being WORSE than placebo.  This article provides the 6 month and 1 year follow up data.  Surprise, surprise... calcium is still not looking good.  

At 6 months survival non-significantly favored the placebo group, and at 1 year it significantly favored the placebo group.  Neurologic outcome for those who survived was also no better, and perhaps slightly worse, in the calcium group. 

Importantly, the trial excluded patients with "traumatic cardiac arrest, known or suspected pregnancy, prior enrollment in the trial, adrenaline prior to possible enrollment, and clinical indication for calcium at the time of randomization."

Bottom Line:  The evidence continues to not support the routine empiric administration of calcium in cardiac arrest.  Patients in whom there is an indication to give calcium (e.g. known ESRD, suspected hyperkalemia, etc) are excluded from these trials, and should likely still receive empiric calcium, but in undifferentiated cardiac arrest you can probably skip the calcium.

Show References

Although it is well-documented that there is no true "maximum" dose of vasopressor medications, further blood pressure support as doses escalate to very high levels tends to be limited.  As such, debate has raged in Critical Care as to when is the "right" time to start a second vasoactive medication.  The VASST trial (Russell et al, NEJM, 2008) is considered to be the landmark trial in this area, and found a trend towards improvement with early addition of vasopressin to norepinephrine, but no statistically significant difference, and may have been underpowered.  

Partly as a result of VASST, the pendulum has tended to swing towards maximizing a single vasoactive before adding a second over the past decade.  The relatively high cost of vasopressin in the US has also driven this for many institutions.  However, more recently a "multi-modal" approach, emphasizing an earlier move to second, or even third, vasoactive medication, is increasingly popular.  Although cost is often prohibitive for angiotensin-2 given controversial benefits, many now advocate for targeting adrenergic receptors (e.g. with norepinephrine or epinephrine), vasopressin receptors (e.g. with vasopressin or terlipressin) and the RAAS system (e.g. with angiotensin 2) simultaneously in patients with refractory shock.  A recent review by Wieruszewski and Khanna in Critical Care (see references) outlines this approach well. 

Bottom Line: When to add a second vasoactive medication (e.g. vasopressin) for patients with refractory shock after a first vasoactive is controversial and not known.  Current practice is trending towards earlier addition of a second (or third) agent, especially if targeting different receptors, but there is limited high-quality evidence to support this approach.  Many practicioners (including this author) still follow VASST and consider vasopressin once doses of around 5-15 micrograms/min (non-weight based) of norepinephrine are reached.

Show References

How to set the correct PEEP remains one of the most controversial topics in critical care.  In fact, just on UMEM Pearls there are 55 hits when one searches for PEEP, including this relatively recent pearl on PEEP Titration.  

A recent Systematic Review and Network Meta-Analysis looked at existing trials on this issue.  They found that:

1) Higher PEEP strategies were associated with a mortality benefit compared to lower PEEP strategies

2) Lung Recruitment Maneuvers were associated with worse mortality in a dose (length of time of the maneuver) dependent fashion.

This fits with recent literature and trends in critical care and bolsters the feeling many intensivists are increasingly having that we may be under-utilizing PEEP in the average patient.  

Bottom Line: As an extremely broad generalization, we would probably benefit the average patient by favoring higher PEEP strategies, and avoiding lung recruitment maneuvers.  Do keep in mind that it is probably best to continue lower PEEP strategies in patient populations at high risk of negative effects of PEEP (e.g. COPD/asthma, right heart failure, volume depleted with hemodynamic instability, bronchopleural fistula) until these groups are specifically studied.

Show References

   There are several well known medications that we tend to give by default during cardiac arrests.  It seems like for each of them, every few years someone does an RCT to see if they really help anybody, and we're all disappointed by what they find.  Well... prepare to be disappointed again, I'm afraid.

   These Danish authors randomized 391 patients in cardiac arrest to either calcium or saline (given IV or IO).  They gave 2 doses of either calcium chloride or saline, with the first dose being along with the first epi dose.  Primary outcome was ROSC.  They also looked at modified Rankin at 30 and 90 days.

  The trial was stopped early for harm.  Now, we all know the dangers of interpreting studies that were stopped early, but this doesn't look great for calcium.  19% of the calcium group had ROSC compared to 27% of the saline group (p = 0.09).  Percentage of patients alive, and with favorable mRS at 30 days also both favored the saline group (although also not statistically significantly).  By the way, of the patients who had calcium levels sent, 74% in the calcium group, vs 2% in the saline group, were hypercalcemic.  Whether that had anything to do with the outcome, we may never know.

Bottom Line:  Is this saying that calcium hurts patients in cardiac arrest?  Maybe... but I don't think this is high quality enough data to draw that conclusion.  At the very least, however, just giving everyone in arrest calcium is probably not terribly helpful.  If you have a reason to give it (known severe hypocalcemia, recent parathyroid surgery, suspected hyperkalemia, etc) then go for it, otherwise you can probably focus your resus on more important things.

Show References

The debate around post-arrest management recently has revolved around whether therapeutic hypothermia should go cold, or LESS cold.  But what if we went MORE cold?  While recent TTM trials have compared temps such as 33 to 36 and 33 to 37.5 or less, a recent trial called CAPITAL CHILL looked at 34C vs 31C.  There is a solid physiologic basis for cooling post-arrest patients, so do they do better if we lower their temp even further?  Maybe we're not going cold enough with 33?

Bottom Line: No, 31C is not better than 34C for post-arrest patients.  This study compared death and poor neurologic outcome at 180 days with 31 and 34C targets for post-arrest patients, and found no difference (in fact the 31C group did slightly, but not significantly, worse on the primary outcome, and worse on a few secondary outcomes).  

While debate remains for 33 vs 36 vs afebrile, the literature does not currently support consideration of temps below 33.  

Show References

The much anticipated REMAP-CAP trial was epublished ahead of print July 12th in Intensive Care Medicine.  It was an RCT investigating four antiviral strategies in critically ill adults with COVID-19: lopinavir-ritonavir, hydroxychloroquine, a combination of the two, and no antiviral therapy (control group).  

Despite the hype around protease inhibitors, hydroxychloroquine, and other unproven therapies in COVID (lookin at you next, Ivermectin...), all three strategies had WORSE outcomes than placebo.  They all decreased organ-support-free days (all reaching statistical significance), which was the primary outcome.  They also all led to longer ICU time, longer time to hospital discharge, and reduced 90 day survival.  Not only does this study show no benefit, it shows fairly convincing signs of harm to these therapies.

Bottom Line: Protease inhibitors (e.g. lopinavir-ritonavir) and hydroxychloroquine are unproven therapies for critical COVID-19 infection, and are not recommended.  Providers should focus on interventions with demonstrated benefit, most notably steroids and good supportive/critical care.  

Show References

Norepinephrine is widely considered the first-line vasopressor for patients in septic shock.  Vasopressin is often added to norepinephrine in patients requiring escalating doses, but when to add vasopressin, and what exactly the benefit is (as opposed to just further titrating up the norepinephrine) remain unclear.  Given the limited evidence for a patient-oriented benefit and the increasing cost of vasopressin, some centers are becoming more judicious in the use of vasopressin.  A systematic review in AJEM October 2021 examined the literature on early (< 6 hours of diagnosis) addition of vasopressin to the management of septic shock patients, compared to either no vasopressin or starting it after 6 hours.

Improved with early vasopressin: Need for renal replacement therapy (RRT; secondary outcome)

No difference: mortality, ICU length of stay, hospital length of stay, new onset arrhythmias

Bottom Line: When, and if, to start vasopressin in patients requiring escalating doses of norepinephrine remains controversial.  Based on the prior VASST trial, many providers will start vasopressin when norepi doses reach ~ 5-15 mcg/min (approx 0.1-0.2 mcg/kg/min), but there remains limited data to support this practice, and either starting vasopressin or continuing to titrate the norepinephrine as needed are both reasonable approaches in most patients.

Show References

Two items from the recent INSPIRATION trial UMEM pearl were very well pointed out by our own Dr. Michael Scott and require clarification.  Thank you to all our readers for their close attention, and please know that we always appreciate you reaching out with questions/comments.  

• Dosing Correction - The "standard-dose" prophylactic dosing of enoxaparin in this trial was 40 milligrams/day.  Please excuse the error in the prior post stating 40 mg/kg/day (we will revise the post).  Standard dosing of enoxaparin for DVT/VTE prophylaxis was a flat 40 mg/day, and was not weight based.

• Major Bleeding - While the difference in major bleeding (2.5% vs 1.4%) was relatively small, this endpoint DID NOT meet non-inferiority.  In other words, the study appeared to detect a statistically significant difference in major bleeding between the dosing regimens.  Given that this is a single study and there are concerns with this finding (the authors themselves describe this as "exploratory"), I would interpret this with caution, but this supports the very intuitive notion that the intermediate (higher) dose regimen of enoxaparin would be associated with more bleeding than the standard dose regimen.  

COVID-19 is generally regarded as a hypercoagulable state, and the role of pulmonary emboli and other VTE in COVID remains unclear.  As a result, how to optimally provide prophylactic anticoagulation in COVID-19 patients who are not known to have VTE has been a point of debate.  

The INSPIRATION trial looked at 600 patients admitted to academic ICUs in Iran, and compared what is often-referred to as "intermediate-dose" prophylaxis (in this case 1 mg/kg daily of enoxaparin) to standard dose prophylaxis (40 mg/day of enoxaparin).  The study utilized a combined endpoint of venous thromboembolism, arterial thromboembolism, need for ECMO, or mortality.  As a reminder, composite endpoints can skew results.  However, the dose and type of anticoagulant chosen is similar to many academic centers around the world, and pharmacy guidelines often recommend providing this type of "intermediate-dose" prophylaxis in COVID-19 patients, sometimes based on clinical status, d-dimer or other coagulation-related patient-data.  As with many things with COVID-19, this practice is based on limited data.

There was no significant difference between groups in the primary outcome (45.7% in intermediate ppx group vs 44.1% in standard group), and while safety outcomes were similar (major bleeding in 2.5% in the intermediate ppx group vs 1.4% in standard group), the intermediate regimen failed to demonstrate non-inferiority to the standard regimen for major bleeding.

Intermediate vs standard-dose ppx was similar in this study with a small, but statistically significant increase in major bleeding in the intermediate-dose group.

Bottom Line: Although this study had methodologic flaws and there are external validity concerns, the INSPIRATION trial supports the notion that standard dose (e.g. 40 mg/g/kg/day enoxaparin) and intermediate-dose (e.g. 1 mg/kg/day enoxaparin) VTE prophylaxis are equivalent in critically ill COVID-19 patients who do not already have a known VTE in terms of preventing negative VTE outcomes.  Intermediate-dose may be associated with increased bleeding.  As more critically ill patients require ED boarding, the dose of VTE prophylaxis may remain controversial, but the need to start it remains an important consideration.

Show References

Optimal oxygenation targets and the possible, theoretical, benefits of hyperoxygenating critically ill patients have long been points of controversy.  Multiple studies have suggested harm in pursuing aggressive hyperoxygenation amongst critical patients with various conditions ranging from myocardial infarction to sepsis to neurologic conditions.  In addition, oxygen toxicity is a known mechanism causing ARDS.

The HOT-ICU trial adds to the list of arguments against hyperoxygenation, by looking at 2928 ICU patients on high levels of supplemental oxygen and targeting a paO2 of 60 mm Hg (low oxygen group) vs paO2 of 90 mm Hg (high oxygen group).  There was no difference in mortality, or other significant difference in outcomes.

Bottom Line: A lower paO2 goal of 60 (correlates to an O2 sat of 90%) is noninferior to a higher paO2 goal of 90 (O2 sat of approximately 96%).  When titrating oxygen, targeting a pulse ox of 90-96% is reasonable in critically ill patients.  Be sure to include an upper limit on the sat goal, beware an O2 sat of 100%, and titrate down supplemental oxygen when the spO2 is above goal, as the paO2 may be dangerously high.

Show References

N-acetylcysteine (NAC) is well known as the accepted antidote for acute acetaminophen (tylenol/paracetamol) overdose and is well studied for this indication.  While the literature base is not nearly as strong in other causes of acute liver failure, NAC is increasingly used in these scenarios as well.  In the emergency department in particular, the cause of fulminant hepatic failure is often not known.  NAC may have some protective benefit in non-acetaminophen acute liver failure.  Existing data do not show a mortality benefit to NAC in non-acetaminophen acute liver failure, but do show improvement in transplant-free survival.  The AASLD guidelines (last revised in 2011) do not comment on NAC in non-acetaminophen acute liver failure.  A common practice is to continue NAC until the INR is < 2 and AST/ALT have decreased at least 25% from their peak values.  

Patients in fulminant liver failure should also be strongly considered for transfer to a center that does liver transplant, if presenting to a non-transplant center.  The King's College criteria is the most commonly used prognostic score for determining need of transfer to a transplant center, but in addition to calculating a King's College score providers should generally consider consultation with a transplant hepatologist for any fulminant liver failure patient to discuss the risks/benefits of transfer for transplant evaluation.

Bottom Line: While not as strongly indicated as it is in acute acetaminophen induced liver failure, NAC should be considered in both non-acetaminophen liver failure and liver failure of unknown etiology.  In addition, strongly consider consultation with a transplant hepatologist in any case of fulminant hepatic failure.

Show References

Another week, another COVID-19 study...

On August 12th, the Metcovid study was e-published ahead of print in Clinical Infectious Diseases.  This was another study looking at steroids in COVID-19 pneumonia, this time performed in Brazil.  Metcovid was a parallel, double-blind, randomized, placebo-controlled phase IIb clinical trial which enrolled 416 patients at a single academic center for the evaluation of methylprednisolone (MP; 0.5 mg/kg BID x 5 days) vs placebo.  As with all COVID studies, Metcovid has some significant limitations, and some equivocal findings.  However, Metcovid was largely in line with RECOVERY and other trials looking at steroids in COVID-19, which lends it some face validity.  Metcovid found no significant difference in the primary outcome (mortality at day 28), but did find a difference in mortality in patients over 60 years old (a post-hoc analysis).  Metcovid was probably underpowered (sample size was based on a 50% reduction in mortality), and did have a very small trend towards reduced mortality in the MP group (37.1% vs 38.2%, p=0.629).

Bottom Line: 

• Steroids (methylprednisolone 0.5 mg/kg BID x 5 days in this case) may have some mild benefit in severe cases of COVID-19 pneumonia, especially in patients who are elderly or have more aggressive inflammatory responses (as measured by CRP here).  
• Steroids in COVID-19 may be associated with some theoretical downsides like reduced viral clearance, but are relatively safe.  Main side effect is the well known hyperglycemia induced by corticosteroids.
• When using steroids in COVID pneumonia, both to stick with the evidence and for theoretical pharmacologic reasons, it may make sense to use dexamethasone or methylprednisolone, as these medications have a higher glucocorticoid:mineralocorticoid activity ratio.  It is hypothesized that using high mineralocorticoid steroids (like cortisone or hydrocortisone) may lead to increased water retention, which could be deterimental in ARDS.  This is purely theoretical.
• There was a signal towards harm in younger and less sick patients in this study, and it probably remains prudent to reserve steroids for older, sicker COVID-19 pneumonia patients, similar to the RECOVERY trial.

Show References

As has been previously noted, the white blood cell count is "the last refuge of the intellectually destitute."  However, within a CBC (especially if a differential is obtained), there is information that can sometimes be of value.  One measure, which was noted before COVID but has come under increasing attention in the current pandemic, is the Neutrophil-To-Lypmhocyte Ratio (NLR).  Because physiologic stress typically causes the Absolute Neutrophil Count (ANC) to increase and the Absolute Lymphocyte Count (ALC) to decrease, the ratio of the two values (NLR = ANC/ALC) should increase when the body is under stress.  Similar to the WBC however, it should be noted that ANY source of physiologic stress can cause abnormalities of the NLR, and thus this is not limited strictly to infectious etiologies.  

With that caveat in mind, the NLR can sometimes be a clue to the degree of physiologic stress the patient is under.  As lymphopenia is a frequent finding in COVID, the NLR has come under particular interest in the setting of COVID and appears to have prognostic value in COVID+ patients.

It should be kept in mind that inflammatory stressors (e.g. sepsis) are likely to disproportionately raise the NLR relative to noninflammatory stressors (e.g. pulmonary embolism), so a septic patient with an NLR of 10 might not be all that ill, whereas a PE patient with an NLR of 10 may be sicker.  As with any single lab, and particularly one so nonspecific, there are no hard and fast cutoffs, and the NLR has to be interpreted in the context of other clinical data (it is very much possible to have a high NLR and not be that sick, or to have a low NLR and be sick... this is only one datapoint and does have pitfalls associated with it).  As a rough guide however, a Pulmcrit post by Josh Farkas from 2019 suggested the following interpretation of the NLR:

1-3: Normal

6-9: Mild stress (e.g. uncomplicated appendicitis)

9-18: Moderate stress, may be associated with critical illness

>18: Severe stress, commonly associated with critical illness

The post (see references below) provides an excellent overview of NLR, further information on the uses and pitfalls of NLR, and several additional sources on the subject.  It's a very worthwhile read.  

Bottom Line: The Neutrophil-To-Lymphocyte Ratio (NLR = ANC/ALC) is one indicator of the degree of physiologic stress, and may be used in conjuction with other clues to determine how sick your patient is.  

Show References

As the debate regarding the pathophysiology and ventilator mechanics of COVID pneumonia rages on, it is important to have a method to evaluate the distensibility of patients' lungs so that we can minimize lung injury.  It has been well shown that both under- and over-distention lead to acute lung injury and inducing or worsening ARDS.

One method to find the "best" level of PEEP is through the PEEP titration test (also called a Driving Pressure titration test).  High Driving Pressure (DP), which is equal to Plateau Pressure - PEEP, has been shown to be associated with lung injury, and the minimal DP obtainable for a given patient while still meeting ventilatory goals is often an objective in the ICU (common DP goal is < 15 cm H2O).  A PEEP titration is optimally done on paralyzed patients, although it can be used on sedated or very calm patients as a "best guess" approximation.  It will not work well on agitated patients or those participating heavily in their ventilation.  Be sure not to do this if you are not authorized to make vent changes, and always make sure to coordinate appropriately with your RT.

To perform a PEEP titration:

*Consider placing the patient on square waveform VC, as this will also allow evaluation of stress index (if patient is not participating).  This can be skipped if not evaluating stress index

1) Make a table for yourself on a piece of paper where you can record PEEP, Plateau Pressure, Driving Pressure, Blood Pressure, and SpO2.

2) Write down the initial PEEP, BP, and SpO2.  Clearly document for yourself that this is the initial PEEP, so you do not inadvertantly leave the vent on different settings at the end.  Perform an inspiratory hold to measure a plateau pressure.  Fill in DP by using the equation DP = Pplat - PEEP

3) Change the PEEP.  You can either increase or decrease.  If you have a suspicion that the patient is over or under distended, go towards optimal distention, but if unsure it is ok to guess.  Usually we go by increments of 2 cm H2O.  Wait about 20-30 seconds on the new PEEP.

4) Measure a new plateau pressure and calculate a new DP.  At each step, write down the BP and SpO2 as well to ensure you are not generating decreased cardiac preload or derecruitment/hypoxia (keep in mind that due to pulse ox lag, you may not see hypoxia for up to a few minutes).  

5) Repeat at a few different PEEP levels.  Typically in more unstable patients who may not tolerate aggressive vent changes you may only want to check 2-3 levels of PEEP.  In more stable patients or if concern for ongoing lung injury is high, you might check up to 5-6 different levels of PEEP.  Please note that some COVID ARDS patients are so unstable that they will not tolerate any derecruitment, and this manuever should not be used in those patients as they could desaturate during the titration.

Once you have all of your data, consider changing to whichever PEEP level gives the lowest driving pressure.  Keep in mind that while data from a PEEP titration can be very useful, it is only one data point and should be considered in combination with blood pressure, volume status, CXR findings, habitus, FiO2 weaning, and other factors.  PEEP titrations should be reperformed periodically (usually daily in most semi-stable ICU patients, more often in unstable patients).  it is also recommended to write a note in the chart with your initial vent settings, data from the titration, and settings upon termination of the titration -- and call your RT if you changed the vent settings.

Bottom Line: PEEP titration (aka Driving Pressure titration) aims to identify the PEEP level where (PPlat - PEEP) is minimal and may help reduce risk of ongoing lung injury in ventilated patients.

Show References

Within the past few days we completed a review of complications of COVID-19, to describe what sequelae and clinical patterns, besides the obvious (URI, respiratory failure, ARDS, sepsis, etc), are noted in the literature.  This review, along with a plethora of other information focusing on critical care of the COVID-19 patient, will be posted in the next few days to http://covid19.ccproject.com/.  Below are the key points from that review:

• Acute cardiovascular complications appear to be the most common and concerning sequelae:  

                 -Acute myocardial injury (7-17% of hospitalized patients in one study),   

                 -Myocarditis (primary cause of death in 7% of COVID deaths in one study),  

                 -Arrhythmias (16.7% of hospitalized and 44.4% of ICU patients in one study), 

                  -Venous thromboembolism (incidence unknown).   

• Concerns for sudden cardiac death, even after recovery, have been raised but are not well documented in the literature.  Proposed mechanisms include respiratory compromise, myocarditis, malignant tachydysrhythmias, heart failure, and coronary plaque instability (i.e. Type 1 MI) secondary to inflammation 

• Co-infection and secondary infection rates are unknown but estimates range from 4.8% to 21%, with higher rates in sicker patients. Viral co-infection is more common than bacterial co-infection, but both may be seen. The ability to rule out COVID-19 by a positive multiplex respiratory viral panel is questionable. 

• Cytokine release syndrome and secondary HLH are both described complications, but their incidence is unknown.  The relation of this finding to purported benefits of tocilizumab (which is also a therapy for HLH) is unknown. 

• Other extrapulmonary complications are relatively typical of sepsis, such as kidney injury, abnormal LFTs, and delirium 

If anyone would like a copy of the full document, which details known complications by organ system, please feel free to email me at msutherland@som.umaryland.edu.  Thanks to David Gordon for organizing the project.

Everyone stay safe, and be sure to take care of each other, as well as our patients.

Don't forget cerebral fat embolism syndrome (FES) on the differential for altered trauma patients.  FES is typically associated with long bone fractures, but has been reported with other fractures, orthopedic reaming (i.e. aggressive orthopedic procedures), and in rare cases even with non-fracture (soft-tissue) trauma.  Typically symptoms occur between 24 and 72 hours after injury, but there have been cases both earlier and later.  Diagnosis is clinical, but MRI may be helpful, and will often show multiple cerebral white matter lesions.  It is debated whether FES is truly an embolic phenomena (i.e fat molecules traveling to and blocking blood supply of organs), or rather an inflammatory response to free fatty acids in the blood stream (i.e. more of a vasculitis type pathology).  Management is supportive care, but give these patients time as there can be favorable outcomes, even after prolonged coma.

Show References

There are few conditions that can be as dramatic or difficult to control as variceal GI bleeding in a cirrhotic patient.  It is important to be familiar with all options in these cases, from Blakemore/Minnesota tube placement to massive transfusion to when and which consultants to get involved.  In cases that are refractory or not amenable to endoscopic intervention, emergent interventional radiology consultation for Transjugular Intrahepatic Portosystemic Shunt (TIPS) may be a consideration.  In high risk cases, think about getting IR on the phone at the same time as you engage GI, in case endoscopic management fails.  Variceal bleed patients can decompensate rapidly, get your consultants involved early!

Generally accepted indications for emergent TIPS (both of the following should be true):

-GI bleeding not amenable or not controllable by endoscopy

-Cause is felt to be variceal. May also consider in portal hypertensive gastropathy

Contraindications:

-Right heart failure or pulmonary hypertension

-Severe liver failure (MELD > 22, T Bili > 3 or Child-Pugh C. In these cases TIPS may not confer a significant survival benefit)

-Hepatic encephalopathy (relative contradindication.  HE may be worsened by TIPS).

-Polycystic liver disease (makes TIPS technically challenging)

-Chronic portal vein thrombus (makes TIPS technically challenging. Acute PV thrombus is NOT considered a contraindication)

Bottom Line: In cases of variceal GI bleeding from portal hypertension, consider getting IR on the phone early to discuss emergent TIPS.

Show References