‘Do-not-intubate’ orders in patients assisted by noninvasive respiratory support for acute hypoxaemic failure caused by coronavirus disease 2019; a systematic review and meta-analysis

BACKGROUND Noninvasive respiratory support (NIRS) has been revealed feasible solutions to cope with the massive request for ventilatory support in patients subjected to ‘do-not-intubate’ order (DNI). OBJECTIVES The aims of the present systematic review and meta-analysis was to estimate pooled incidence of DNI orders and the associated in-hospital mortality in patients undergoing NIRS for hypoxaemic acute respiratory failure (ARF) related to coronavirus disease 2019 (COVID-19). DESIGN Systematic review of observational studies and randomized-controlled trials with meta-analyses DATA SOURCES PUBMED, EMBASE, and Cochrane Controlled Clinical trials register were searched for observational studies and randomised-controlled trials from inception to the end of April 2022. ELIGIBILITY CRITERIA Inclusion criteria were: observational studies enrolling ≥50 hospitalised patients with hypoxaemic COVID-19-related ARF requiring NIRS and DNI order application. Two authors independently extracted data from enrolled investigations. Data are presented as proportions with 95% confidence interval. RESULTS Thirty-one observational studies were included for a total of 6645 COVID-19 patients undergoing NIRS, of whom 1590 received DNI orders. Among patients assisted by NIRS, a DNI order was expressed in a summary estimate of 25.4% [20.0–31.1] of the cases with a high between-study heterogeneity. The summary estimated of in-hospital mortality was 83.6% [75.3–90.7] for DNI patients and 20.0% [14.2–26.5] for full treatment patients, both with a high between-study heterogeneity. CONCLUSIONS In COVID-19 patients assisted through NIRS for hypoxaemic ARF, a DNI order was frequently issued and associated with a high in-hospital mortality TRIAL REGISTRATION Prospero registration number: CRD42021271313.