Aug 232019

Jon-Emile S. Kenny MD [@heart_lung]

I had some dreams, they were clouds in my coffee, clouds in my coffee …”

-Carly Simon


A recent letter to the editor in Critical Care posed a rather provocative question – ‘Are we really preventing lung collapse with APRV?’  The authors cited a case report of esophageal manometry used in conjunction with airway pressure release ventilation [APRV] in an obese patient with severe, extra-pulmonary ARDS.  As the very short expiratory time in APRV [i.e. T-low] is an essential element to reclaim lost lung in ARDS, allegations that lung collapse will continue to confound the intensivist employing APRV is an existential threat to this ‘less-conventional’ mode of ventilation.  Can we conclude from the case report that decreasing T-low from 0.7 seconds to 0.1 seconds actually aggravated lung collapse?  From the data presented in the case report, is there a potentially better way to titrate T-Low?

What They Did

As above, this was a case report in an obese patient with severe extra-pulmonary ARDS.  The compliance of the patient’s respiratory system [i.e. lung and chest wall together, Crs] was 50 mL/cmH20.  The chest wall compliance [Ccw] was 29 mL/cmH2O, while the patient’s lung compliance [Cl] was 21 mL/cmH2O.

The patient’s APRV pressure settings were a P-High of 30 cmH2O and a P-Low of 0 cmH2O.  The time at P-High and P-Low – T-High, and T-Low, respectively – were the independent variables.  T-Low started at 0.1 s and increased by increments of 0.1 to 0.7 s; accordingly, T-High was variable from 5.9 to 5.3 s relative to the increased T-Low as the total cycle time was fixed.

Each setting was recorded for 2 min and the following variables were noted: airway pressure, esophageal pressure, inspiratory and expiratory trans-pulmonary pressure [Ptp], expiratory flow, and percentage decay of expiratory flow from peak expiratory flow at the end of the T-Low.

The authors did not perform an end-expiratory hold [i.e. the end of T-low] to confirm the amount of auto-PEEP in the patient described in the case report.  They did, however, present the amount of auto-PEEP using an end-expiratory hold in a separate patient on APRV to illustrate the method.

What They Found

The authors found that at the most rapid T-low [i.e. 0.1s], the trans-pulmonary pressure [i.e. the airway pressure minus the esophageal pressure] was the most negative.  Consistently, as T-Low was lengthened [i.e. from 0.1s to 0.7s], the trans-pulmonary pressure was less negative.  In other words, the trans-pulmonary pressure suggested the most collapse when T-low was 0.1s; additionally, when the T-low was the longest [i.e. 0.7s], there appeared to be less collapse, but collapse nonetheless.  Thus, all of the trans-pulmonary pressures at the end of T-low were negative.  The percent of peak expiratory flow rate for these different T-low times ranged from 87% at 0.1s to 75% at 0.7s.

In the additional patient who had an end-expiratory hold maneuver done at the end of T-low, the trans-pulmonary pressure appeared negative while the actual measured trans-pulmonary auto-PEEP was positive.  Of note, during the course of the end-expiratory hold, both the airway and trans-pulmonary gradually trended up upwards – more on this below.


The fact that shorter expiratory times [i.e. more rapid T-lows] led to increasingly negative trans-pulmonary pressure – suggesting more collapse – really grabbed my attention.  The theory underpinning APRV is that the short T-low results in auto-PEEP which recruits lung and therefore prevents collapse.  How could this result be so contradictory?  My hypothesis is that the measured end-expiratory trans-pulmonary pressures were under-estimated [i.e. too low]; in fact, I suspect that they were all probably quite positive and that their negative values were a function of caveats when measuring both auto-PEEP and esophageal pressure.  More simply, the measured airway pressures were lower than what they actually were and the esophageal pressure was higher than it actually was – this combination resulted in spuriously low trans-pulmonary pressure.  But why do I think that?

The first clue came from the respiratory maneuver they did not perform on the subject of the case report, but rather an additional patient – shown for illustrative purposes.  When the hold at the end of T-low was performed, the initial trans-pulmonary pressure was negative, but rapidly became positive.  Most importantly, the airway pressure gradually trended upwards throughout the hold maneuver.  I was so intrigued by this, that I used my favourite digital tracing software and estimated that the airway pressure in the case report gradually rose by at least 5 cm H20 throughout the hold.  This nicely exemplifies the concept of ‘occult’ auto-PEEP – illustrated in figure 2 of this exceptional review on auto-PEEP by Marini.  Briefly, occult, auto-PEEP occurs when auto-PEEP is ‘trapped’ behind closed airways – this often occurs in the lower lobes, and especially in obese patients [as in the case report] where the pleural pressure is greater at the bases - leading to small airways collapse.  The resultant pre-mature airway closure leaves the patient with distended alveoli with high levels of auto-PEEP that is ‘hidden’ from the central airways and therefore not recorded by the ventilator.  The hallmark of hidden auto-PEEP is that the airway pressure gradually trends upwards during an expiratory-hold maneuver as the gas hidden behind the closed airways gradually leaks out and raises the central airway pressure.  This is exactly observed in the case report.  In summary, the true pressure in the airways was likely much higher, but it was just hiding from the ventilator.

A significant and additional clue comes from examining the waveforms presented in the case report.  Using the digital tracing software, it was also apparent to me that the trans-pulmonary pressure at the end of P-High gradually fell from a P-Low of 0.7s to about 0.4s and then started to rise again.  What does this mean?  Consider an analogy with conventional volume control; if PEEP is increased and plateau pressure decreases [i.e. the driving pressure falls], it suggests that lung has been recruited and that the compliance has improved.  If the P-High trans-pulmonary pressure is considered analogous to the plateau pressure, its reduction as T-low decreased suggests that pulmonary compliance was improving, probably as a consequence of auto-PEEP recruiting lung.  Indeed, this may represent a novel way to titrate T-Low in APRV – simply decrease T-low until the trans-pulmonary pressure at the terminus of P-High reaches its minimum [see figure].

Figure: A novel way to titrate T-Low. This figure depicts 7 different P-High-to-P-Low releases at different T-Low durations. As T-low is decreased [to cause auto-PEEP] leftwards along the x-axis, the trans-pulmonary [y-axis] pressure at the very end of P-High [red dots] gradually falls to a minimum at 0.4s of T-Low and then rises again. The falling trans-pulmonary P-High implies improving lung compliance and may occur at a higher % decay of peak expiratory flow target than commonly recommended [e.g. often 75% is targeted]

Lastly, mediastinal weight adds a fixed load to the esophagus and thus increases esophageal pressure even though the change in esophageal pressure likely represents an ‘average’ change in pleural pressure.  Some have advocated simply subtracting 3 – 7 cm H2O from the absolute value obtained by esophageal manometry; this was not done in the case report.

In summary, the airway pressure was likely much higher than measured [because of occult auto-PEEP], and the esophageal pressure lower [because of mediastinal weight].  Accordingly, the observed trans-pulmonary pressure at the end of T-Low was arguably much lower [i.e. more negative] than its true value.  Indeed, the trans-pulmonary pressure at the end of T-Low was likely always positive.  Thus, titrating T-Low based on its absolute trans-pulmonary pressure is problematic.  Instead, a better indicator for T-Low optimization may be the minimum value of the trans-pulmonary P-High as shown in the figure above.



Dr. Kenny is the co-founder and Chief Medical Officer of Flosonics Medical; he also the creator and author of a free hemodynamic curriculum at




Get our weekly email update, and explore our library of practice updates and review articles.

PulmCCM is an independent publication not affiliated with or endorsed by any organization, society or journal referenced on the website. (Terms of Use | Privacy Policy)


APRV and Esophageal Manometry: a new way to titrate T-Low?