Postoperative Outcomes in Surgical COVID-19 Patients

A Multicenter Cohort Study

François Martin Carrier; Éva Amzallag; Vincent Lecluyse; Geneviève Côté; Étienne J. Couture; Frédérick D'Aragon; Stanislas Kandelman; Alexis F. Turgeon; Alain Deschamps; Roy Nitulescu; Codjo Djignefa Djade; Martin Girard; Pierre Beaulieu; Philippe Richebé


BMC Anesthesiol. 2021;21(15) 

In This Article


This study provides data on surgeries performed in COVID-19 patients and their postoperative outcomes in the Canadian province most afflicted by the pandemic.[17] We observed an important postoperative 30-day mortality of 15.9% in patients undergoing surgery with COVID-19, potentially from symptomatic patients although we could not conclude. We observed a 50% relative reduction in overall surgical activities during the pandemic in most university hospitals of the province of Quebec.

Overall, SARS-CoV-2 infected patients have not experienced many surgeries during this pandemic wave (< 0.2% of surgical cases). This observation is probably a combination of limited surgical needs in this population and a restriction to surgical care imposed on them until they recover from their infection to potentially reduce postoperative complications.[22,23] As such, they required mostly urgent minor surgery, although 36% of them required a major one. Routine preoperative testing for SARS-CoV-2 itself has been recently associated with less postoperative pulmonary complications in major surgeries, probably by allowing to postpone or cancel surgery in active cases.[24] Such intervention was not routinely applied to all surgical patients in the Province of Quebec during the first wave due to limited supply of reactants. Nonetheless, 25% of the patients suffered from a pulmonary complications, which is lower than other reported incidences.[7,8] Similarly to another cohort of COVID-19 surgical patients, we observed a small incidence of thromboembolic complications (4.5%) for a population of patients undergoing mostly urgent surgeries.[7]

Our observed 30-day mortality of 15.9% seemed to be lower than the 19.5 to 23.9% reported mortalities in other cohort studies.[7,8,16] The possible discrepancy might be explained by random variation, different patient selection for surgery or different overall perioperative care. However, our Kaplan-Meir 30-day survival probability confidence interval in COVID-19 patients was wide (from 0.74 to 0.96), suggesting that our observation was likely to not be significantly different from other ones. One group compared postoperative outcomes in 41 COVID-19 patients to 82 non-COVID-19 patients matched by surgical disease.[8] They observed a higher proportion of complications and mortality in COVID-19 patients, but non-COVID-19 patients had a better preoperative respiratory function, did not need any mechanical ventilation and were not in septic shock at time of surgery.[7,8] We observed a relatively comparable mortality between suspected patients (16.7%) and symptomatic COVID-19 patients (23.1%), suggesting that postoperative mortality may be secondary to the presence of an active infectious process at surgery. Such observation could also be explained by false negative COVID-19 results in suspected patients, potentially aligning their outcome to those of COVID-19 patients.[25] Overall, like other groups, we observed a moderately high 30-day postoperative mortality that was higher than expected in non-cardiac surgery (1–4%).[26,27]

Surgical care should not be overlooked during a pandemic. Even though mobilizing surgical ward and operating room resources to care for SARS-CoV-2 infected patients, population's surgical needs have to be fulfilled.[15,28,29] Compared to the previous year, more than 22,000 patients over 9 hospitals in the province of Quebec did not receive timely surgical care during our period of observation. In one centre (centre #1), 8 out of 420 patients hospitalized for COVID-19 needed surgery (data not reported previously), while 3500 surgical cases were cancelled during this period. In the greater Toronto area hospitals, all surgical activities were reduced by as much as 37 to 70% for both inpatient and outpatient surgeries during the SARS pandemic in April 2003.[30] During the first wave of the COVID-19 pandemic, we observed a reduction in surgical activity of 50% over 3 months in 9 participating centres, although the impact is probably beyond what we observed. Thus, a significant backlog of surgeries will have to be undertaken while the health system is still stressed by the pandemic.[23] Real-time system utilization indicators should be further developed and applied during a pandemic to help adapt surgical elective programs within system overwhelming prevention strategies as well as "COVID-19 free pathways" to prevent cross-contamination.[31,32] Allocating care to COVID-19 patients is paramount, but this should not be done at the price of over restricting care in surgical patients to ensure proportional resource allocation across all population subgroups.

Our study has limitations. Its main limitation is the small sample size. While this highlights the limited number of surgeries performed in this population, it precluded any regression-based quantitative analysis on determinants of poor postoperative outcomes. We did not have the power to quantitatively measure association with either the presence of symptoms in COVID-19 or the presence of an active infection compared to our controls. Also, our comparators are biased, since "suspected" patients were by definition sick patients and "patients who had recovered" were fit enough patients to undergo an elective surgical procedure. Since we included patients tested positive for SARS-CoV-2, we could have missed untested patients, especially asymptomatic ones. However, this was probably limited since all patients having compatible symptoms or any risk factor for SARS-CoV-2 based on Québec Public Health Agency guidelines were tested prior to surgery. Our observations must therefore be interpreted as descriptive and exploratory. On the other hand, we included patients from many centers, providing relative generalizable results. Despite these limitations, we were able to draw a perspective of the surgical care during the current pandemic in Québec, Canada's hardest hit province with almost half of the national COVID-19 cases.