ICU stories (common): Middle-aged pt w hx of COPD was brought to the ED by EMS after SOB x 2 days. No fever or chest pain. Very quickly after ED arrival, he was intubated. CXR showed hyperinflated, “COPD” lungs. Here depicted in two images:
Post-intubation ABGs showed:
BP dropped post-intubation to 55/40; propofol was started & then discontinued due to hypotension. iv fluids (2L Lactate Ringer's) were started & patient was brought to the ICU w SBP in upper 80s. Re-institution of propofol led again to hypotension. What would be the next step?
Patient came from the ED with these vent settings 👇 (RR was increased from 18 to 22 to ⬇️ the PCO2 of >100 in the first ABGs):
There was suspicion of auto-PEEP ---> presence of expiratory flow when every new breath was given. Next step was to check lung mechanics: Pplateau was 38 and PEEPtotal 25 (revealing an auto-PEEP close to 20…)
Next step?
We discontinued the patient from the ventilator for 15 sec and decreased RR to 14. Five minutes later, P plateau & auto-PEEP had decreased nicely 👇. Hemodynamics improved without fluids/pressors & despite the patient being back on propofol.
Next step to decrease auto-PEEP further?
IMHO, if you do the basic right things (⬇️ RR) and give patient time, things will continue to improve. No other change was made in the ventilator settings and, 30 min later, lung mechanics continued to improve:
Reminder: Extrinsic PEEP is set by the clinician on a ventilator while auto-PEEP is unintentional
Pepe and Marini used the term "auto-PEEP" back in 1982 👇
In the same paper, Pepe and Marini described the method of quantifying auto-PEEP by occlusion of the expiratory port
I love this paper not just because it's classic and brings physiology at the bedside, but also because it gives us a chance to glimpse at the ICU practices of the times... Please take a look at dopamine use, dopamine doses and tidal volume:
Excessive auto-PEEP can impede venous return and can also precipitate acute right ventricular failure. Volume expansion can help up to a point; the point where a dilated RV will i) be unable to eject its preload into the pulmonary circuit impair and ii) impair LV filling...
The auto-PEEP phenomenon still remains under-recognized. Sometimes, it is so dramatic that patients in PEA arrest spontaneously "recover" when they are disconnected from the ventilator circuit ("Lazarus
effect"). Especially in bad asthmatics or COPD patients that suffer PEA,
consideration should be given to simply discontinue ventilation transiently for 10-20 sec during CPR and observe the patient for return of circulation.
It would be interesting to learn what the highest autoPEEP you have seen is and if there is an easy way to determine if by going to zero “set PEEP” helps with air emptying
Pt 1 in the Pepe-Marini paper: "it was observed that systolic BP ⬆️ to 140 mmHg within 30 s of a brief IPPV interruption, whereas wedge decreased to 10-13 mmHg. Simultaneously, mean PAP & HR ⬇️. Whenever IPPV was reinstituted, HR, BP, wedge, PAP returned to their previous values
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Elderly 👴 w interstitial🫁disease was admitted -on NIV- from the ED to StepDown unit w "pneumonia": bil infiltrates/low-grade fever. PF ratio 140. Since I was in the ICU, I asked how he looked. Reply: needs high FiO2 but pulls good tidal volumes
When I saw him:
This is not an uncommon scenario: a "low" expired tidal volumes (Vt) makes the staff worried during NIV initiated for hypoxemic respiratory failure but a "high" Vt (even if accurately measured) is left unabated...
In fact, high Vt during NIV has been associated with NIV failure and poor outcomes
The American Thoracic Society (ATS) updated its recommendations on some aspects of the diagnosis and management of community-acquired pneumonia (CAP)
This guideline update focuses only on immunocompetent adult patients with a standard diagnosis of CAP
It addresses four clinically relevant questions:
The guideline underwent peer review by 16 experts (4 from ATS; 11 from Infectious Diseases Society of America (IDSA). It Is not clear to me why IDSA did not approve it. I copy from the supplement: "The IDSA chose to withdraw rather than approve the final version of the guideline"
Have you ever managed a septic patient & ordered antibiotics to be given stat only to discover hours later that they were still infusing? If yes, stay tuned
Among several controversies in sepsis management, the early administration of antibiotics
is one of the least debatable ones. Time is life in sepsis!
Even though it's hard to believe that every hr of delay in antibiotics ⬆️ mortality by 8% (this is a different topic...), it's one of my pet peeves when it takes for ever to give antibiotics to a septic shock patient
The most commonly ordered combination in 🇺🇸 hospitals (EDs/ICUs) is piperacillin-tazobactam (Zosyn in 🇺🇸) + vancomycin (V). If the usual practice is followed, giving a dose of Zosyn & a dose of vanco can take close to 3 hours. This should not be the case!
Is there anything regarding UGIB that has not been already done, studied or tweeted about? Not much, but all of the following are things I witnessed happening (or not happening…) in two UGIB cases I recently saw in the ICU
Here it begins:
1. If a patient does not have an obvious UGIB as in the clip above, she has just presented with melena & it is still unclear if this is due to upper or lower GIB, a quick & dirty trick is to check the blood urea nitrogen/creatinine (BUN/creat) ratio. There are various published
cut-offs in the literature (don’t sweat about them…) but, in general, a ratio of >30 suggests UGIB. This is due to the small bowel absorption/digestion of blood protein which is subsequently metabolized to urea. There are confounding factors, but the trick usually works well
Walking from room to room in the ICU on a Sunday morning while hoping to have a quiet shift, you notice this 👇 on a patient's monitor (60 yo, admitted, intubated for COPD exacerbation + pneumonia 3 days ago, now sedated/hemodynamically stable)
It's unfortunately hard to pretend you didn't see it, so you get an ECG praying that it will not show what you saw on the monitor... 😊
Of note, admission ECG was "ok"
Well, it actually looks more impressive on the 12-lead ECG: