PEEP titration during prone positioning for acute respiratory distress syndrome
The present analysis highlights the paucity of data on concomitant high PEEP with
prone positioning. Given the purported overlapping benefits to regional mechanics
afforded by both interventions, these data raise legitimate concern about generalizability
of existing clinical trial data on prone positioning to patients receiving a high-PEEP
strategy. Our findings highlight the need for additional research comparing prone
positioning, high PEEP, and both in combination for treatment of moderate-severe ARDS.
The mechanisms by which high PEEP may protect against lung injury are similar to those
associated with proning. Both therapies promote more homogenous lung recruitment,
minimizing local shear stress; reduce cyclic opening/collapsing of potentially recruitable
lung units during tidal ventilation; and improve ventilation-perfusion matching. Indeed,
preclinical models have demonstrated that both proning and PEEP mitigate ventilator-induced
lung injury 31]–33]. Clinically, the PROSEVA trial demonstrated definitively better survival from moderate-severe
ARDS with prone compared with supine positioning when managed with a low-PEEP strategy
9]. A similar survival benefit has been suggested with a high-PEEP supine strategy,
compared with a low-PEEP supine strategy, for moderate-severe ARDS 20]. No multicenter randomized trial testing a high- versus low-PEEP strategy has demonstrated
definitively a survival benefit nor enrolled exclusively patients with moderate-severe
ARDS (PaO
2
/FiO
2
of not more than 200). However, a meta-analysis of the three largest trials found
improved survival with high PEEP in the subgroup of patients with PaO
2
/FiO
2
of not more than 200 20], comparable to the population enrolled in PROSEVA. Thus, the overlapping physiological
and clinical effects of proning and high PEEP raise doubt as to whether a proning
low-PEEP strategy yields a survival advantage compared with a high-PEEP supine strategy;
further studies are needed to address this question.
Moreover, whether concomitant provision of high PEEP and proning offers additional
clinical benefit over either alone is unknown. In patients with diffuse infiltrates
on chest computed tomography (CT), combination therapy improves oxygenation and reduces
intrapulmonary shunt compared with either monotherapy 34]. Prone positioning may reduce chest wall compliance 35], potentially necessitating higher PEEP to offset these changes. A recent CT study
of 24 patients with ARDS found that cyclic atelectasis decreased only when higher
PEEP (15 versus 5Â cm H
2
O) and prone positioning were applied together. Whereas each therapy in isolation
promoted lung recruitment, simultaneous proning and higher PEEP also mitigated regional
hyperinflation observed with higher PEEP during supine positioning. An individually
titrated high-PEEP strategy in combination with proning has not been studied rigorously
to date. The ideal combination therapy may require adjusting PEEP after each repositioning
to account for changes in chest wall and lung mechanics in the prone versus supine
position.
Importantly, the optimum approach to setting PEEP remains undefined. In this study,
high PEEP was estimated by using PEEP-FiO
2
titration tables of the ALVEOLI and LOVS trials. Such an “open lung†strategy was
shown recently to achieve higher levels of PEEP in patients with more lung recruitability
as measured by CT scan, whereas other methods—ExPress, stress index, and esophageal
manometry—had no association with lung recruitability 36]. Most major clinical trials to date have titrated PEEP according to severity of oxygenation
impairment by using an arbitrary PEEP-FiO
2
table with comparably lower PEEP 17], 18]. However, mechanical insults and resultant biotrauma appear to be the primary drivers
of ventilator-induced lung injury 37]. Thus, we believe that the optimum approach involves titrating PEEP to minimize mechanical
lung injury. Several mechanics-based approaches have been proposed: setting PEEP above
the lower inflection point of the static pressure-volume curve 38], 39], according to pleural pressure (estimated by esophageal manometry) to achieve a non-negative
end-expiratory transpulmonary pressure (airway minus pleural pressure) 23], 40], and according to highest respiratory system compliance 41], 42] or lowest driving pressure (airway end-inspiratory plateau pressure minus PEEP) 43], among others. Mechanics-based PEEP strategies yield different PEEP selections than
high PEEP-FiO
2
tables 36]. To maximize synergy with proning, a mechanics-based approach to PEEP adjustment
may be necessary to further improve lung homogeneity, avoid overdistension or hemodynamic
compromise, and prevent cyclic opening/collapse of potentially recruitable lung units
44]. Changes in chest wall compliance (decreased when prone) and aerated lung volume
(increased when prone) following repositioning highlight the apparent need to adjust
PEEP after each position change to optimize mechanics for lung injury prevention.
Some important limitations to the present study are of note. First, although no included
trial incorporated a high-PEEP titration strategy, the particular PEEP strategy used
differs for each included study. Only the PROSEVA trial used a PEEP-FiO
2
titration table for all patients, identical to the low-PEEP strategy employed in the
ALVEOLI trial. Second, our calculation of expected high PEEP may overestimate what
the true PEEP level would have been had a high PEEP-FiO
2
table been used. Expected PEEP was calculated by using a PEEP-FiO
2
table, although a mechanics-based approach is preferred in our view. Increasing PEEP
often improves oxygenation, permitting a reduction in FiO
2
while maintaining PaO
2
or SpO
2
(blood oxygen saturation measured by pulse oximetry) goals. High PEEP occasionally
also could contribute to hemodynamic instability, particularly in patients with high
lung compliance in whom overdistension is likelier. These limitations, which would
lower PEEP, are offset in part by our reporting the lowest expected PEEP level allowed
whenever the high-PEEP titration protocol permitted more than one PEEP setting for
a given FiO
2
. Third, it is unknown whether a clinically meaningful interaction between PEEP and
prone positioning exists, as simultaneous provision of these therapies has never been
studied in a clinical trial powered for patient-centered endpoints. Finally, it is
unclear whether high PEEP would have altered clinical outcomes in either the prone
or supine study arms, although existing data suggest a benefit with supine positioning
in moderate-severe ARDS 20].