Optimal Positive End-Expiratory Pressure Therapy in Infants and Children with Acute Respiratory Failure1

Abstract
Positive end-expiratory pressure (PEEP) has become a mainstay in the treatment of hypoxemic acute respiratory failure (ARF). Whereas PEEP improves arterial oxygen tension by decreasing intrapulmonary shunting, it may also impair cardiac output and hence decrease systemic oxygen transport. Inasmuch as optimizing oxygen transport is a goal of therapy in ARF, we sought to determine if the level of PEEP that results in maximal oxygen transport could be estimated from measurements of compliance of the respiratory system (Crs) or PaO2. We studied the effects of PEEP application on cardiorespiratory parameters in 15 children who required mechanical ventilation for ARF. Static Crs, PaO2, central venous and arterial blood pressures, indicator dilution cardiac index (CI), and oxygen transport were determined at 0, 3, 6, 9, 12, and 15 cm H2O PEEP. PaO2 increased significantly at PEEP levels ≥9 cm H2O (p < 0.001), while CI fell by 15% between 0 and 15 cm end-expiratory pressure (p < 0.02). Crs and oxygen transport did not change significantly with increasing levels of PEEP. The level of PEEP resulting in maximal oxygen transport ranged from 0 to 15 cm H2O, and in all patients it corresponded to PEEP of best CI. At levels of PEEP above that associated with maximal oxygen transport, CI and oxygen transport fell significantly, while PaO2 continued to rise. No relationship between Crs and oxygen transport was observed. In our normovolemic patients with ARF, neither PaO2 nor Crs predicted PEEP of maximal oxygen transport. The decrease in CI at high levels of PEEP that prevented improvement in oxygen transport could not be detected by routine clinical monitoring of heart rate or vascular pressures.