Background and aims Respiratory failure is common in immunocompromised patients. Intubation and mechanical ventilation (MV) is the mainstay of treatment but is associated with increased risk of pneumonia and other complications. Non-invasive ventilation (NIV) is an alternative to MV in a select group of patients and aims to avoid the complications of MV.
In these patients, we performed a meta-analysis on the effect of early NIV versus conventional oxygen therapy in reducing intubation rates and other important clinical outcomes.
Methods We performed an extensive online and unpublished data search for relevant studies that met the inclusion criteria. We included randomised controlled trials that used early NIV versus conventional oxygen therapy in immunocompromised patients with respiratory failure. Risk of bias and acceptability assessment were independently performed by the authors.
The primary outcome of interest was intubation and MV rate.The secondary outcomes were ICU and all-cause mortality, ICU length of stay and duration of mechanical ventilation.
Results Four studies with a total of 553 patients met the criteria for inclusion and were included in the analysis.
Patients given NIV were 38% less likely to be intubated vs. those given oxygen, RR 0.62 (95% CI: 0.42, 0.93); however, this result is significantly heterogenous. After sensitivity analysis, results showed 48% less likelihood of intubation and mechanical ventilation in the group treated with NIV, RR 0.52 (95% CI: 0.35, 0.77).
Patients on NIV had 1.18days less stay in the ICU vs. oxygen group (95% CI: −1.84, −0.5 days). There was no statistically significant decrease in all-cause mortality between the two groups, RR 0.84 (95% CI: 0.63, 1.13), but this effect is heterogenous. After another sensitivity analysis performed specifically for this outcome, results showed a 25% significant reduction in all cause mortality in patients given NIV vs. oxygen therapy, RR 0.75 (95% CI: 0.58, 0.96).
There is no difference in the duration of mechanical ventilation between groups.