PEEP-induced changes in lung volume in acute respiratory distress syndrome. Two methods to estimate alveolar recruitment

Abstract

Purpose: Lung volumes, especially functional residual capacity (FRC), are decreased in acute respiratory distress syndrome (ARDS). Positive end-expiratory pressure (PEEP) contributes to increased end-expiratory lung volume (EELV) and to improved oxygenation, but differentiating recruitment of previously nonaerated lung units from distension of previously open lung units remains difficult. This study evaluated simple methods derived from bedside EELV measurements to assess PEEP-induced lung recruitment while monitoring strain. Methods: Prospective multicenter study in 30 mechanically ventilated patients with ARDS in five university hospital ICUs. Two PEEP levels were studied, each for 45 min, and EELV (nitrogen washout/washin technique) was measured at both levels, with the difference (D) reflecting PEEPinduced lung volume changes. Alveolar recruitment was measured using pressure-volume (PV) curves. High and low recruiters were separated based on median recruitment at high PEEP. Minimum predicted increase in lung volume computed as the product of DPEEP by static compliance was subtracted from DEELV as an independent estimate of recruitment. Estimated and measured recruitments were compared. Strain induced by PEEP was also calculated from the same measurements. Results: FRC was 31 ± 11% of predicted. Median [25th-75th percentiles] PEEP-induced recruitment was 272 [187-355] mL. Estimated recruitment correlated with recruited volume measured on PV curves (q = 0.68), with a slope close to identity. The DEELV/FRC ratio differentiated high from low recruiters (110 [76-135] vs. 55 [23-70]%, p = 0.001). Strain increase due to PEEP was larger in high recruiters (p = 0.002). Conclusion: PEEPinduced recruitment and strain can be assessed at the bedside using EELV measurement. We describe two bedside methods for predicting low or high alveolar recruitment during ARDS. © jointly held by Springer and ESICM 2011.