Differentiating Lung From Chest Wall Mechanics Is Difficult Without Esophageal Manometry in Children With Acute Respiratory Distress Syndrome.

OBJECTIVES

Pediatric acute respiratory distress syndrome (PARDS) guidelines recommend limiting airway plateau pressure (Pplat) to 28 cm H 2 O, allowing for higher limits when chest wall compliance (C CW ) is poor since less of the pressure is transmitted to lung (transpulmonary pressure). Transpulmonary pressure depends on Pplat and the ratio of lung elastance to respiratory system elastance (E L /E RS ). E L /E RS measurement requires esophageal manometry, although it is not routinely available. We sought to determine if routinely available clinical data could reliably predict E L /E RS or changes in E L /E RS , to understand when Pplat greater than 28 cm H 2 O could be acceptable.

DESIGN

Secondary analysis of randomized controlled trial with esophageal manometry monitoring.

SETTING

Quaternary PICU.

PATIENTS: Mechanically ventilated children with PARDS.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN

RESULTS

Two hundred seven patients and 750 patient days were included. Using the first day per patient, median E L /E RS was 0.83 (interquartile range, 0.72-0.87), with a weak negative correlation with respiratory system compliance (C RS ) ( r = -0.26; p < 0.001). C RS was strongly correlated with lung compliance (C l ) ( r = 0.94; p < 0.001) and moderately correlated with C CW ( r = 0.53; p < 0.001). Multivariable analysis identified that higher C RS , younger age and peripheral neuromuscular disease were associated with higher C CW , while higher C RS was the only variable independently associated with higher C l (all p < 0.01). When trying to predict high (> 0.9) or low (< 0.7) E L /E RS , C RS was the only variable retaining an independent association: lower C RS (C RS × 10 [mL/cm H 2 O/kg × 1/10]) with high E L /E RS (odds ratio [OR], 0.70; 95% CI, 0.54-0.86; p = 0.002; area under the receiver operating characteristic curve [AUC], 0.73) and higher C RS (C RS × 10 [mL/cm H 2 O/kg × 1/10]) with low E L /E RS (OR, 1.14; 95% CI, 1.02-1.28; p = 0.017; AUC, 0.60). Change in E L /E RS from day to day was not predictable.

CONCLUSIONS

In PARDS, C RS is more strongly tied to C l than C CW . While E L /E RS is not easily predictable from clinical variables, when C RS is low, E L /E RS is generally high. Therefore, increasing Pplat above the suggested thresholds when C RS is impaired may be inappropriate without measuring esophageal pressure.