Dilip Jayasimhan

On the note of VAP, we've written a review article. More specifically, focusing on the evidence behind strategies to prevent and manage VAP. This link gives 50 days of free access. Hope you all find it useful. #emimcc
authors.elsevier.com/a/1lbEr7K3e2...

Insulin gtt + usual management. Oral triglyceride meds going forward.

A lot of times, just good simple intensive care goes a long way.

What does be careful mean, though?

Using PSV pretty early and primarily as a ventilatory mode (not for SBT) would be usual practice in most centers in Australia & New Zealand.

Nice thread. Well summarised. Thank you. Imho- prone ventilation helps homogenise distribution of mechanical power. If on VV and on “rest” ventilation (whatever that is since no consensus on what rest is whilst ELSo guidelines allude to) then the benefit from proning is v minimal

PRONECMO used a competing risk analysis as the outcome, and hence, I do not understand the power calculations. If someone could explain this to me, I'd be grateful. From my perspective, however, I wonder if a trial with 170 patients is definitive on the effect of PP in ECMO.

A simple sample size calculation shows that to reduce mortality from 35% to 20%, you'd need to enrol 276 patients, which still amounts to a lot of patients on ECMO. Reduce that estimate of effect to 10%, and you'd need to recruit 878 patients.

EOLIA had a 60-day mortality of 35% in the ECMO group. It's unlikely, I believe, that PP would reduce mortality by 15% in this group, let alone at the levels observed in PROSEVA, given that ECMO allows a safe reduction in driving pressure, one of the benefits of PP.

Power calculations for PROSEVA estimated a 28d mortality of 60% in the control group and required enrolling 456 patients for 90% power to detect a 15% absolute reduction in mortality. The mortality in the control group was higher at 75%, with PP mortality of 38%. The true effect is likely smaller.

I think, given that it took many trials with different sample sizes and durations of proning for us to realise its effectiveness in ARDS without ECMO, it would be even harder to prove in patients on ECMO for several reasons. COI: I love to prone 🧵 #emimcc

Mostly by titrating PEEP and prone positioning. Occasionally using a brief recruitment maneuver as rescue.

As important as understanding the physiological rationale of APRV and its potential efficacy is, it is equally crucial to assess how well-prepared your system is to adopt and apply a new approach to demonstrate effectiveness. This partly explains the difficulties in performing large APRV RCTs.

There may be individuals within the system who are comfortable and able to use this mode, but it would be worthless if the system as a whole (the rest of the team) is not. Unfamiliar strategies introduce complexity, and increasing complexity may lead to increased noise in decision-making.

I work in a system or region of systems where APRV is very rarely used, ventilator alterations are minimal, and typically undertaken by doctors and nurses (with no respiratory therapists, etc.).

I don't have strong opinions about APRV, but I think the utility and effectiveness of certain strategies depend heavily on the system within which they work. #EMIMCC

#emimcc

A well written argument cautioning against misinterpreting evidence from clinical trials on angiotensin II. Currently, A-II is not available outside a clinical trial where I work. However, there are efforts to evaluate A-II in other populations (not sepsis). pmc.ncbi.nlm.nih.gov/articles/PMC...

Prophylactic Antibiotics in Adults With Acute Brain Injury Who Are Invasively Ventilated in the ICU Current evidence from randomized clinical trials does not provide definitive evidence regarding the effect of prophylactic antibiotics on mortality in patients receiving invasive mechanical ventilatio...

Although this is biologically plausible, at present, the certainty of the evidence is low. However, more trials will be conducted evaluating this question in the coming years. journal.chestnet.org/article/S001...

Other disciplines have shown the benefits of identifying disease behaviours. An example is the idea of a progressive fibrotic phenotype in ILD, which now has therapies that have shown benefit despite the underlying entities being heterogeneous (IPF, Fibrotic HP, NSIP, etc.).

I agree that clustering patients into syndromes has its limitations. The current clusters are nowhere near perfect. However, I disagree that this is not worth studying. Many patients with different etiologies of critical illness share common pathways that may benefit from a particular treatment.

I struggled to find any. I suspect a big cost to running such a trial (without industry sponsorship) is the equipment and resources.

That's a fair point. I was just picking a trial to illustrate that these therapies can be evaluated systematically rather than specifically endorsing that particular trial.

I rather think it’s action bias, easier to do something than to wait and see. People overestimate the risk of deterioration and underestimate bleeding risk.

I guess the question is how much we value shortening the illness over the small but catastrophic risk of a major bleed.

Why do we not see similar issues for other indications for thrombolysis (e.g stroke or STEMI without access to PCI)?

interesting thread with some great points here 👇

IMHO the key is having an institutional pathway about classification & management that everybody agrees on

for patients in whom the treatment pathway is clear, there is no need to involve everybody (eg, if clearly not an IR candidate, why call IR?)

Similar arguments could be made perhaps for “shock” teams

🤔

Could not agree more! The problem with pulmonary embolism is the lack of clear ownership (at least in my part of the world). #emimcc #cardiosky

We often make things more complicated than they should be.