Increased incidence of post-operative respiratory failure in patients with pre-operative SARS-CoV-2 infection

Objective: While studies have reported increased post-operative pulmonary complications with SARS-CoV-2 infection, many are limited by use of historical controls or focus on less severe respiratory complications. We characterized the association between pre-operative SARS-CoV-2 infection and post-operative respiratory failure (PORF). Design and setting: This was a single center retrospective cohort study in New York City between March 14–June 14, 2020. Patients: Exclusion criteria were age < 18-years, obstetric procedures, absence of SARS-CoV-2 PCR testing, and pre-operative respiratory failure. A total of 778 patients met criteria, of which 87 had SARS-CoV-2. Measurements: The primary outcome, PORF, included inability to extubate for ≥24 h or unplanned re-intubation within 5 days. Multiple exposures were measured including SARS-CoV-2 infection 4 weeks before or 5 days after surgery. Multivariable logistic regression was performed to adjust for pre-operative hypoxemia, oxygen use, and pneumonia as well as tachycardia, gender, Charlson Comorbidity Index (CCI), Surgical Mortality Probability Model (S-MPM) index, and peri-operative blood transfusion. Main results: SARS-CoV patients had higher CCI (P = 0.007) and S-MPM scores (P = 0.02). The incidence of PORF was 16% versus 7% in uninfected comparators (P = 0.001). Amongst infected individuals, 39% exhibited symptoms of COVID-19 and PORF was more common in these patients compared to asymptomatic individuals (26% vs. 9%, P = 0.04). Adjusted analysis revealed increased odds of PORF with infection (OR 2.8, 95% CI 1.2–6.2). This persisted even when adjusting for probable mediators such as pre-operative hypoxemia. Infected patients also demonstrated increased adjusted odds of 30-day mortality (OR 3.5, 95% CI 1.4–9.1). Conclusions: Detection of SARS-CoV-2 infection within 4 weeks before or 5 days after surgery is associated with increased odds of 5-day PORF and 30-day mortality. This supports delaying elective surgery, but questions remain regarding the applicability of this recommendation for asymptomatic patients needing urgent or semi-urgent procedures such as oncologic surgery.