Shanholtz – Salvage Therapy for Acute Hypoxemic Respiratory Failure

Dr. Carl Shanholtz was the man of the hour at the University of Maryland Multidisciplinary CC Conference this week, here to discuss his “Plan B – Z” for patients with refractory hypoxia despite your best efforts.  Dr. Shanholtz is one of the few PI’s of the ARDSnet collaboration and shares an amazing evidence based walkthrough of the different salvage therapies we have available to us when you must throw the proverbial kitchen sink at your dying patient.  What do you have in your back pocket?


  • General Housekeeping
    • Beware of studies where outcomes are compared against historical controls – standard of care and practices change, which will bias the data.
    • We all know to keep the plateau pressure < 30 mmHg, but there doesn’t appear to be a safe Pplat that you can ignore low-tidal volumes (Hager et al., 2005).
    • For patients with reasonable hemodynamics and ARDS, practicing a fluid restrictive strategy appears to improve outcomes.
      • Now this is a big one – remember, fluids are drugs!
      • Keep a close and constant eye on the cumulative I/O balance sheet.  You’ll be surprised at how positive your patients actually are if you don’t keep track!
      • It has been suggested that up to 10% (over 5,000) of the overall mortality from ARDS is a direct result of iatragenic hypervolemia!
  • Proning
    • Concept: Reducing the amount of atelectasis and alveolar edema at the bases as well as the posterior segments due to the weight of the patient’s heart, body tissue, gravity, etc. while lying supine for an extended period of time.
    • No statistical benefit demonstrated for all-comers with ALI/ARDS, however…
    • For patients with moderate to severe ARDS (P/F < 200) there does appear to be significant benefit to proning.
    • Guerin et al., 2013 – for patients with severe ARDS found a reduction in 28 & 90 day mortality by 50%.
  • Neuromuscular blockade
    • Concept: Reducing ventilator dyssynchrony will reduce biotrauma of lungs and minimize iatragenic inflammation/edema in lungs.
    • Technique: Cistatracurium infusion x 48 hours (Papazian et al., 2010)
    • May have some outcome benefit.
  • Inhaled vasodilators
    • Concept:  By using an inhaled vasodilator, you can improve V/Q mismatch by preferentially dilating pulmonary vasculature that’s perfusing unaffected lung – unlike an intravenous Rx that theoretically would just increase your shunt fraction.
    • Multiple studies on iNO, but does not appear to have a significant impact on outcomes for hypoxemic respiratory failure, specifically in ARDS (Adhikari et al,. 2007).
  • Extracorporeal Life Support (ECLS/ECMO)
    • Historically, ECLS technology was pretty poor – however, improved circuit technology has increased the availability, reduced blood trauma, and increased the length of support that can be provided.
    • The CESAR trial (Peek et al., 2007) is intriguing, however the significant benefit in death & disability (composite endpoint, not a primary endpoint) in the ARDS group managed with ECMO, and…. there was an asterisk.
      • (*) The treatment arm may have not actually received treatment (i.e. ECMO) – 1/4 of these patients received “standard treatment” but were merely transferred to an ECMO capable center.
      • Other confounders (treatment center variation, low tidal volume ventilation strategy, etc.)
    • But, there’s hope for all you ECMO nerds out there (like me) … let’s wait and see what happens with the EOLIA trial.
  • Airway Pressure Release Ventilation (APRV)
    • Concept: Ventilator strategy that essentially provides a high CPAP with inverse ratio ventilation.
      • Increased PEEP will maintain recruitment, prevent RACE injury, improve oxygenation.
      • Sacrificing CO2 clearance.
    • Overall, no significant clinical data to support outcome benefit with ARDS.
  • High Frequency Oscillatory Ventilation (HFOV)

Share this Post


  1. mjanderson972

    So, if I have a patient (predicted BW: 6 cc/kg~550) with severe ARDS, showing air hunger on PRVC so switched to PCV, has a Vt of 900-1000cc with ventilator set on PCV ~ 10/12. Does the data support to paralyze this patient to prevent volutrauma–just want to be sure I interrupt this correctly?

    1. John Greenwood

      Hi Matt! So I think all of us have run into this issue (on a relatively frequent basis!), awesome question. Below are some thoughts from Dr. Shanholtz:

      I’ve wrestled with this concept, but that was the point of Art Slutsky’s editorial accompanying the Papazian cisatracurium study. Essentially patients can spontaneously overventilate themselves and add to volutrauma. So there are two or three approaches to this: 1) sedate/neuromuscularly blockade them for the first 48 hr to prevent spontaneous overdistention, 2) do what we did on the ARDSnet 01 trial and if there is significant dysynchrony with marked double-triggering, increase the tidal volume by 1 mL/kg predicted body weight at a time, figuring that 7, 8, or 9 mL/kg PBW is at least better than 2 X 6 = 12 mL/kg PBW. The possible third option is, if PEEP and FIO2 are low, to figure out if the patient’s tidal volumes are so large because their lungs are healed and keep them on a ventilator liberation protocol.

      One of my criticisms of APRV is that I’ve seen some less experienced practitioners only pay attention to spontanous tidal volumes at P1 and the so-called “release volumes” whenever the ventilator cycles ignored. This is especially a problem if the patient takes a deep breath as the ventilator cycles. I’ve seen tidal volumes of 700-900 mL in some cases. I am skeptical that “release volumes” are any safer than tidal volumes, because it is still a lung volume that changes with airway and transalveolar pressure.

      Remember, one of the early ventilator-induced lung injury (VILI) studies performed by Didier Dreyfus in rats included a group ventilated with high airway pressures, but the chest and abdomen were strapped to prevent overdistention, and they had histology and fluid flux that resembled the control animals. Another group had tidal volumes delivered under negative pressure (e.g. cuirass) where airway pressure was actually negative. but lungs were overdistended to the same degree as the high pressure-high volume group, and they had histology, lung water, etc. that was just as bad.

      Hope that helps.


      Thought his comments were really interesting…

      As far as alternative strategies, Eddy Fan & Daniel Brodie talk about using ECCO2R to achieve low tidal volume ventilation and prevent VILI ( I think this concept is intriguing. Smaller catheters – basically the size of RRT catheters – and lower flow to clear CO2. Given the etiology of “Air hunger” is the result of increased dead space or need to ventilate ( perhaps this can be mitigated by extracorporeal means if traditional strategies fail… just a thought.

      Hope this helped!

      — John

  2. Matt

    Thanks for the great reply. Keep up the awesome work! Hoping that my program may jump on the FOAMed bandwagon in the future.

Leave a Comment