Abstract

Objectives: The lung-protective strategy employs positive end-expiratory pressure to keep open otherwise collapsed lung regions (anatomical recruitment). Improvement in venous admixture with positive end-expiratory pressure indicates functional recruitment to better gas exchange, which is not necessarily related to anatomical recruitment, because of possible global/regional perfusion modifications. Therefore, we aimed to assess the value of venous admixture (functional shunt) in estimating the fraction of nonaerated lung tissue (anatomical shunt compartment) and to describe their relationship. Design: Retrospective analysis of a previously published study. Setting: Intensive care units of four university hospitals. Patients: Fifty-nine patients with acute lung injury/acute respiratory distress syndrome. Interventions: Positive end-expiratory pressure trial at 5 and 15 cm H2O positive end-expiratory pressures. Measurements and main results: Anatomical shunt compartment (whole-lung computed tomography scan) and functional shunt (blood gas analysis) were assessed at 5 and 15 cm H2O positive end-expiratory pressures. Apparent perfusion ratio (perfusion per gram of nonaerated tissue/perfusion per gram of total lung tissue) was defined as the ratio of functional shunt to anatomical shunt compartment. Functional shunt was poorly correlated to the anatomical shunt compartment (r2 = .174). The apparent perfusion ratio at 5 cm H2O positive end-expiratory pressure was widely distributed and averaged 1.25 ± 0.80. The apparent perfusion ratios at 5 and 15 cm H2O positive end-expiratory pressures were highly correlated, with a slope close to identity (y = 1.10•x -0.03, r2 = .759), suggesting unchanged blood flow distribution toward the nonaerated lung tissue, when increasing positive end-expiratory pressure. Conclusions: Functional shunt poorly estimates the anatomical shunt compartment, due to the large variability in apparent perfusion ratio. Changes in anatomical shunt compartment with increasing positive end-expiratory pressure, in each individual patient, may be estimated from changes in functional shunt, only if the anatomical-functional shunt relationship at 5 cm H2O positive end-expiratory pressure is known. Copyright © 2008 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins.