Editor, Postoperative pulmonary complications are frequent in patients with perioperative severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and are associated with high mortality.1,2 However, limited data are available regarding asymptomatic patients with a positive SARS-CoV-2 Viral RNA reverse transcriptase–PCT (RT-PCR) in nasopharyngeal swabs who undergo emergency surgery. To clarify this point, we conducted a retrospective chart review of all consecutive patients aged above 18 years who underwent general anaesthesia for an emergency surgical procedure between 15 March and 30 April 2020 at Henri Mondor University Hospital, Paris, France. These patients (PCR+ group) were matched in a 1 : 2 or 1 : 3 ratio to historical patients (control group) according to age, sex, type of anaesthesia and indication for surgery. We defined as ‘historical period‘ the period before the end of January 2020. The first cases of coronavirus disease-2019 (COVID-19) in France were diagnosed between 9 January and 22 January 2020. One control patient underwent surgery on 11 January 2020 and had not travelled recently. We considered that all our control patients had not been exposed to SARS-CoV-2. We excluded symptomatic patients (defined by the presence of one of the following signs: cough, fever, anosmia, gastrointestinal symptoms, dyspnoea, acute respiratory failure, myalgia, sputum production and headache)3 and those who underwent regional anaesthesia or ambulatory care. End-points of the study were postoperative 30-day mortality, all-cause postoperative complications (including all complications until discharge from hospital), ICU admission rate, operating room occupation time, and hospital length of stay. Complications were classified according to the Dindo--Clavien classification.4 Continuous variables are presented as median [IQR], categorical variables as number (%). Analyses of postoperative outcomes relied on mixed effects regression models to account for the correlation between matched patients, using logistic regression for binary outcomes and linear regression for continuous outcomes. Associations between outcomes and SARS-CoV-2 status were expressed as odds ratios (OR) for binary endpoints or beta regression coefficients for continuous endpoints with their 95% confidence intervals after accounting in multivariate analysis for comorbidities and ASA physical status. All analyses were performed using Stata 16 (StataCorp, College Station, Texas USA). The level of significance was set at 0.05. The study design was approved by the Institutional Review Board (IRB 00010254 -- 2020 -- 105). A total of 642 patients underwent surgery with general anaesthesia between 15 March and 30 April 2020. Twenty-four (3.7%) asymptomatic patients with a preoperative positive SARS-CoV-2 RT-PCR (PCR+ group) and 55 control patients who underwent surgery and general anaesthesia between February 2019 and January 2020 were included. Seven patients with preoperative positive RT-PCR were matched to three control patients. Patients underwent orthopaedic (25% of cases), urology (25%), vascular (21%), oncological, plastic and reconstructive (12.5%) operations, and other (16%) procedures (see Supplementary Table, https://links.lww.com/EJA/A474). There was no significant difference between PCR+ and control groups considering baseline features used for matching and regarding demographics (Table 1). Underlying comorbidities were present in 87.5% (n=21). No significant differences were observed between the groups regarding anaesthetic management that was performed according to the published guidelines during the COVID-19 pandemic.5 The median duration of operating room occupation was longer in the PCR+ group (197.5 [IQR 115 to 250] vs. 137 [90 to 168] min, P = 0.03), mainly because of the postanaesthesia care of positive SARS-CoV-2 RT-PCR performed in the operating room. Table 1 - Patients’ characteristics and outcome RT-PCR+ groupn=24 Control groupn=55 P Patient's characteristics Age (years) 70 [60 to 82] 71 [62 to 81] Male sex 16 (66.7) 33 (60.0) BMI (kg m−2) 24.2 [22.5 to 26.5] 25.3 [23.0 to 27.2] ASA physical status 1 or 2 13 (54.2) 31 (56.3) ≥3 11 (45.8) 24 (43.6) Outcome 30-day mortality 4 (16.7) 2 (3.6) 0.06 Postoperative complications (Dindo--Clavien classification) As a continuous variable 1 [0 to 4] 0 [0 to 1] 0.004 Regression coefficient 1.14 (0.40 to 1.87) Adjusted regression coefficient 1.04 (0.34 to 1.75) ≥3 8 (33.3) 7 (12.7) 0.04 Odds ratio 3.43 (1.07 to 10.9) Adjusted odds ratio 4.33 (1.03 to 18.12) Complications Thrombotic complications 1 (4.2) 0 0.3 Hemorrhagic complications 1 (4.2) 3 (5.5) 1.00 Infective complications 8 (33) 5 (9.1) 0.02 Pulmonary complications 6 (25.0) 1 (1.8) 0.003 Covid-19 pneumonia 4 (16.7) 0 Operating room occupation time (min) 197.5 [115 to 250] 137 [90 to 168] 0.03 Postoperative ICU admission 3 (12.5) 6 (10.9) 1.00 Length of hospital stay (days) 7 [4 to 20] 6 [3 to 15] 0.56 Values are median [IQR], mean (95% CI) or n (%). ASA, American Society of Anesthesiologists; ICU, intensive care unit. Bold value was considered to be statistically significant. The 30-day mortality was similar between the groups [n=4 in the PCR+ group, n=2 in the control group, OR = 7.08 (95% CI, 0.84 to 59.4), P = 0.06]. Postoperative complications were higher in the PCR+ group (1 [0 to 4] vs. 0 [0 to 1] in the control group, P = 0.004). Severe postoperative complications (Dindo--Clavien class ≥3) were higher in the PCR+ group [33.3 vs. 12.7% in the control group, OR 3.43 (95% CI, 1.07 to 10.9) P = 0.04]. All complications occurred before postoperative day 30. Pulmonary complications were higher in the PCR+ group (25 vs. 1.8% in the control group, P = 0.003). All patients in the PCR+ group were investigated with preoperative chest computed tomography (CT) before surgery. Lung abnormalities were observed in 12 patients (50%); bilateral distribution of ground glass opacities with or without consolidation in posterior and peripheral lungs were the predominant patterns encountered, and reticular patterns were observed in two patients. In the PCR+ group, four patients (16.7%) developed postoperative COVID-19 pneumonia. They all had CT scan abnormalities in the preoperative period. Patients without CT scan abnormalities in the preoperative period did not develop pulmonary complications, including COVID-19 pneumonia, in the postoperative period. As a result, the presence of lung abnormalities on the preoperative CT scan was associated with a higher risk of developing pneumonia (50% in patient with PCR+ and lung abnormalities vs. 0% in patients with PCR + without lung abnormalities, P = 0.01). No correlation between the extent of the CT scan abnormalities and the development of pneumonia was observed. Our results are consistent with the study of Lei et al.,6 highlighting that asymptomatic patients developed pneumonia shortly after elective surgery. Moreover, two recent studies focusing on surgical SARS-CoV-2-infected patients both reported higher mortality and complications.1,2 Our rate of 25% respiratory complications would be more plausible, considering the exclusion of nonurgent procedures and locoregional anaesthesia in our study. Although large cohort studies are lacking, it is of importance to consider that the incidence of postoperative pulmonary complications was more than twice the usual incidence. We may hypothesise that combining inflammatory responses because of the virus and general anaesthesia with mechanical ventilation of the lungs could lead to an explosive inflammatory response.7 Half of patients with CT abnormalities had postoperative pneumonia whereas no patient without CT abnormalities developed pneumonia. Despite the fact that a CT scan remains debatable in the decision making-process, it seems to provide a reliable tool to identify patients at risk of complications. The routine use of a CT scan for SARS-CoV-2-positive patients undergoing surgery could be useful in identifying high-risk patients, and therefore improving management and outcome. In conclusion, our study supports the fact that asymptomatic patients requiring general anaesthesia with a positive SARS-CoV-2 RT-PCR are more likely to develop postoperative complications. A tailored and multidisciplinary approach, including a preoperative CT scan, should be implemented in order to adopt the optimal strategy.