Incidence of noninvasive ventilation failure and mortality in patients with acute respiratory distress syndrome: a systematic review and proportion meta-analysis.

BACKGROUND

Noninvasive ventilation (NIV) is commonly used in patients with acute respiratory distress syndrome (ARDS). However, the incidence and distribution of treatment failure are unclear.

METHODS

A comprehensive online search was conducted to select potentially eligible studies with reports of the rate of NIV failure in patients with ARDS. A manual search was also performed to identify additional studies. Data were extracted to calculate the pooled incidences of NIV failure and mortality. Based on oxygenation, the severity of the disease was classified as mild, moderate, or severe ARDS. Based on etiologies, ARDS was defined as being of pulmonary origin or extrapulmonary origin.

RESULTS

We enrolled 90 studies in this meta-analysis, involving 98 study arms. The pooled incidence of NIV failure was 48% (n = 5847, 95% confidence interval [CI]: 43-52%). The pooled incidence of ICU mortality was 29% (n = 2363, 95%CI: 22-36%), and that of hospital mortality was 33% (n = 2927, 95%CI: 27-40%). In patients with mild, moderate, and severe ARDS, the pooled incidence of NIV failure was 30% (n = 819, 95%CI: 21-39%), 51% (n = 1332, 95%CI: 43-60%), and 71% (n = 525, 95%CI: 62-79%), respectively. In patients with pulmonary ARDS, it was 45% (n = 2687, 95%CI: 39-51%). However, it was 30% (n = 802, 95%CI: 21-38%) in those with extrapulmonary ARDS. In patients with immunosuppression, the incidence of NIV failure was 62% (n = 806, 95%CI: 50-74%). However, it was 46% (n = 5041, 95%CI: 41-50%) in those without immunosuppression.

CONCLUSIONS

Nearly half of patients with ARDS experience NIV failure. The incidence of NIV failure increases with increasing ARDS severity. Pulmonary ARDS seems to have a higher rate of NIV failure than extrapulmonary ARDS. ARDS patients with immunosuppression have the highest rate of NIV failure.