Acute Respiratory Distress Syndrome Cases Volume and ICU Mortality in Medical Patients

Crit Care Med. 2018 Jan;46(1):e33-e40. doi: 10.1097/CCM.0000000000002816.

Abstract

Objectives: To determine whether ICUs caring for higher volumes of acute respiratory distress syndrome patients would be associated with lower ICU mortality.

Design: A 9-year multicenter retrospective cohort study of prospectively collected data.

Setting: French medical ICUs.

Patients: From 2004 to 2012, acute respiratory distress syndrome cases were identified from a coding system through a regional database (Collège des Utilisateurs de Données en Réanimation).

Interventions: None.

Measurements and main results: Volume was calculated as the cumulative annual mean number of acute respiratory distress syndrome cases. Severity (Simplified Acute Physiology Score 2) and ICU mortality between categories (low, medium, and high) of acute respiratory distress syndrome cases volume were analyzed. Multivariable analysis using mixed effects models was performed to adjust for severity of illness and confounding factors. Over the study period, 8,383 acute respiratory distress syndrome patients among 31 ICUs met the study inclusion criteria. Overall, Simplified Acute Physiology Score 2 (median [interquartile]) was 58 (43-74), whereas ICU mortality was 53.7%. Severity as assessed by Simplified Acute Physiology Score 2 (median [interquartile]) was significantly higher in high-volume ICUs (> 65 acute respiratory distress syndrome per year) as compared to low (≤ 29 acute respiratory distress syndrome per year) and medium-volume ICUs (> 29-65 acute respiratory distress syndrome per year): 61 (46-77) versus 55 (41-72) and 55.0 (40-72), respectively (p < 0.01). ICU mortality was similar across the acute respiratory distress syndrome volume categories (53.6%, 54.1%, and 53.3% in low-, medium-, and high-volume categories ICUs, respectively). After adjustment for confounders, acute respiratory distress syndrome case volume was independently associated with ICU mortality (odds ratio for log-transformed volume: 0.77 [95% CI, 0.62-0.96]; p = 0.02).

Conclusions: ICUs caring for higher volumes of acute respiratory distress syndrome cases were associated with lower ICU mortality.

Publication types

  • Comparative Study

MeSH terms

  • Aged
  • Female
  • Hospital Mortality
  • Hospitals, High-Volume*
  • Hospitals, Low-Volume*
  • Humans
  • Intensive Care Units*
  • Male
  • Middle Aged
  • Paris
  • Respiratory Distress Syndrome / mortality*