3 RCTs on ARDSnet protocol/standard vs. higher PEEP

  • Lower PEEP (i.e. PEEP of 5 means FiO2 30%)
  • There was no mortality benefit nor change in duration of mechanical ventilation
  • Higher peep arm -> Improved P/F ratio, compliance, but higher mean airway pressures and higher plateau pressures
  • Briel meta analysis in 2010 looking at the three trials
    • Found when you look at mod ARDS -> benefit w/ higher PEEP arm
      • But, then the opposite trend was occurring with the mild ARDS (“ALI” during that era)
      • Discordant effects depending on the severity of illness
    • Take home: if you have real ARDS, you should be on “higher” PEEP

Driving pressure -> Amato 2015

  • Plat-PEEP=driving pressure
  • This one found that increasing PEEP did not (in-and-of-itself) improve mortality

ART trial JAMA 2017 -> Recruitment maneuver to max compliance

  • Lower 6mo mortality in control arm/non-recruited group (“low PEEP”) <- but had higher driving pressures than recruitment arm?? I didn’t know this
  • Recruitment arm/intervention showed high PEEP with best compliance was harmful maybe b/c higher mean airway pressures
  • Harm from intervention??
    • Was the real harm from the hypoxia or hypoventilation in the recruitment maneuver <- pts became unstable during this recruitment
    • Recruitment maneuver was long time -> something like 60 secs

Dreyfuss rat study on transpulmonary pressures -> high vol = ARDS and die…is it the tidal vol or the high transpulmonary pressures that lead to death and ARDS?

Remember, transpulmonary pressure = airway pressure – pleural pressure

  • Group 1 -> ventilated w/ a highTV -> deemed high vol/high pressure rats
    • ARDS and died
    • High transpulmonary pressures…
  • Group 2 -> iron lungs w/ same TV as group 1 ->  b/c iron lung (airway pressure is neg) -> high vol/low pressure rats
    • ARDS and died
    • This group brings up the huge transpulmonary pressures/P-SILI topic I think…
  • Group 3 -> rubber band around chest wall -> low vol/high pressure
    • No ARDS; no die
    • I.e. low transpulmonary pressure
    • Seems like the low transpulm pressure is the thing that matters–potentially more than low TV…

*not sure what year this three-group rat study by Dreyfuss was…

Talmor 2008 looking at transpulmonary pressure-guided vs. ARDSnet low PEEP arm

  • Looked at esophageal pressure-guided transpulmonary pressure and found trend towards decreased mortality
  • The transpulmonary pressure-guided (i.e. esoph balloon) arm, the PEEP was higher
    • Argument is are you just doing the high vs. low PEEP trial…
    • Kicker -> in the esophageal balloon group they started to see positive end-expiratory transpulmonary pressures
  • Looked at the end-expiratory transpulmonary pressure

EPVENT 2 -> mod/sev ARDS

Control is high PEEP arm

Intervention is arm with PEEP titrated to end-expiratory transpulmonary pressure

  • He gave example to first place the balloon in the right spot
    • Test by pushing on the stomach -> should get the same rise in the airway pressure as you get in the esophageal pressure
      • If you’re in the abd; the “esoph pressure” (which is really abd pressure in this case) would rise out of proportion of airway pressure
  • Prelim data showing PEEP is roughly the same in the two arms <- this is important b/c you’re not doing a high-PEEP vs. low-PEEP trial

Random commentary on bladder pressure <- seeming to make argument that could inform what your transpulmonary pressure: using bladder pressure as rough guide in conjunction w/ PEEP measurement

  • If your PEEP is 12 and bladder P is 16; you prob have neg transpulmonary pressure


Recruitment improves compliance -> theorized to improve outcome

  • Multiple trials that showed no benefit -> Guerin 2004, Taccone JAMA 2009
    • But, the Gattinoni meta-analysis of 1,867 pts showed mortality benefit in pts with P/F < 100

  • Recent Guerin PROSEVA Investig NEJM 2013 RCT showed that P/F <150 showed that 16h/day proning w/ mortality benefit (16% vs. 33%) without increase in complications
    • Thought to be 2/2 increasing compliance of the lung?? <- remember, that above analysis of JE Kenny’s diagram shows how driving pressure decreases 2/2 proning (b/c chest wall compliance decreases w/ proning; but, overall compliance of resp system increases?)
  • Problem with proning is that there were three neg trials before this PROSEVA trial showed benefit <- furthermore, the mortality of 16% is super low…signal or random noise/chance??
  • Gave us an example of strong negative inspiratory efforts creating large, neg transpulmonary pressures <- loses the benefit of PEEP


  • Papazian ACURASYS NEJM 2010 -> 340 pts with mod/sev ARDS
    • Group randomized to 48h of Nimbex w/ small benefit of 90-d mortality <- But not statistically significant difference
    • No difference in lung parameters -> other than rate of PTX was lower
    • Low PEEP for both arms of the study
    • Once they adjusted for the P/F ratios, Pplat, and SAPS II there was a significant difference in 90-d mortality with benefit in Nimbex

  • The new ROSE study of the PETAL network will look at Nimbex, but will give both groups high PEEP -> huge sample size of 1,400 pts
    • Physical therapy will do a formal evaluation of neuromuscular weakness
    • We’ll know the answer to NMB after this ROSE trial
    • Levitt’s hunch is that the trial will be negative trial b/c of clinicians excluding pts/biasing selection prior to enrolling pts <- b/c anyone they thought would benefit from paralysis would’ve already gotten it and those thought to not benefit would have already been decided to not get it. Therefore, left over enrollees are diluted…


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s