Do-Not-Resuscitate Orders in Patients With Acute Hypoxemic Respiratory Failure Due to COVID-19: Practices, Predictors, and Outcomes in a Retrospective Study

Background Do-Not-Resuscitate (DNR) orders were commonly used in patients with severe Coronavirus Disease 2019 (COVID-19), with variation across different healthcare settings. We evaluated the practices of DNR orders in patients with severe COVID-19 at a tertiary care hospital in Saudi Arabia. Methods This retrospective cohort study evaluated patients with acute hypoxemic respiratory failure due to COVID-19 who required ICU admission in Riyadh, Saudi Arabia, between March and December 2020. We compared patients with DNR orders during ICU stay to those without and described their clinical condition on the day of DNR order implementation. Results Among 470 patients (median age 62.0 years, 348 (74.0%) were males, 207 (44.0%) used high-flow nasal cannula and/or noninvasive ventilation without intubation, and 263 (56.0%) needed intubation), 30.6% (N = 144) received DNR orders after a median of 10 days in the ICU. Patients with DNR orders were older and had more comorbidities. On the DNR order day, 126 patients (87.5%) were on invasive mechanical ventilation (median fraction of inspired oxygen, 0.80), 84 (58.3%) were on vasopressors, and 42 (29.2%) were on renal replacement therapy. The median Sequential Organ Failure Assessment (SOFA) score was 13. On multivariable logistic regression analysis, age (odds ratio (OR) per one-year increase, 1.052) and intubation (OR, 3.757) were associated with DNR orders. The hospital mortality rate was significantly higher in patients with DNR orders (133/144 (92.4%) vs. 61/326 (18.7%), p < 0.0001). Conclusions DNR orders were commonly used in patients with severe COVID-19 during the first wave of the pandemic, typically after one week of ICU admission. Key predictors were older age and persistent organ failure.