Perioperative Morbidity and Mortality of Patients With COVID-19 Who Undergo Urgent and Emergent Surgical Procedures

Anne Knisely, MD; Zhen Ni Zhou, MD, PhD; Jenny Wu, BS; Yongmei Huang, MD, MPH; Kevin Holcomb, MD; Alexander Melamed, MD, MPH; Arnold P. Advincula, MD; Anil Lalwani, MD; Fady Khoury-Collado, MD; Ana I. Tergas, MD, MPH; Caryn M. St. Clair, MD; June Y. Hou, MD; Dawn L. Hershman, MD; Mary E. D'Alton, MD; Yolanda Ya-Chin Huang, MD, PhD; Jason D. Wright, MD


Annals of Surgery. 2021;273(1):34-40. 

In This Article


A total of 468 subjects, including 36 (7.7%) with laboratory confirmed COVID-19 were identified (Table 1, Supplemental Table 1, The number of daily surgical procedures performed declined from 62 on March 17, 2020 to 15 on April 15, 2020. Among those with COVID-19, the diagnosis was confirmed in 55.6% preoperatively and in 44.4% postoperatively. Before the availability of preoperative COVID-19 laboratory screening, 7.7% (1 of 13) of cases were diagnosed preoperatively. After the availability of laboratory screening, 65.2% (15 of 23) of cases were diagnosed preoperatively (P = 0.0008) (Figure 1).

Figure 1.

Total emergent and urgent surgical procedures from March 17 until April 15, 2020. Triangles represent COVID-19 patients diagnosed postoperatively, stars represent COVID-19 patients diagnosed preoperatively.

Patients diagnosed with COVID-19 were more often non-White (P = 0.04) and Medicaid recipients (P = 0.003) compared to those without COVID-19. Overall, 22.2% of those with COVID-19 were over 70 years of age compared to 25.2% of subjects without COVID-19 (P = 0.68). Compared to the controls, subjects ultimately diagnosed with COVID-19 more commonly had underlying COPD (8.3% vs 3.9%; P = 0.13) and coronary artery disease (30.6% vs 12.0% P = 0.003). The majority of procedures were classified as urgent (vs emergent) in both cohorts (P = 0.19) General surgical procedures followed by gynecologic procedures were the most common procedural types for both cohorts. Patients with COVID-19 were more commonly classified as ASA class 4 (22.2% vs 9.5%) or 5 (5.6% vs 0.23%) compared to those without COVID-19.

Serious perioperative complications were noted in 58.3% of COVID-19 positive patients compared to 5.6% of COVID-19 negative patients (P < 0.0001) (Table 2). Individually, cardiac arrest (16.7% vs 1.2%; P < 0.0001), shock (13.9% vs 0.9%; P = 0.0002), respiratory failure (33.3% vs 2.6%; P < 0.0001), pneumonia (50.0% vs 2.8%; P < 0.0001), acute kidney injury (22.2% vs 3.5%; P < 0.0001) and acute respiratory distress syndrome (ARDS) (8.3% vs 0%; P = 0.0004) were all more common among subjects with COVID-19. Similarly, the rates of intraoperative transfusion (16.7% vs 6.3%; P = 0.02), postoperative transfusion (22.2% vs 7.9%; P = 0.007), and ICU admission (36.1% vs 16.4%; P = 0.004) were more frequent in those with COVID-19. The rate of reoperation (2.8% vs 3.9%; P = 0.36) did not differ between the cohorts. Patients diagnosed with COVID-19 required oxygen postoperatively more than COVID-19 negative patients (50.0% vs 15.1%; P < 0.0001). The perioperative mortality rate was 16.7% among those with COVID-19 compared to 1.4% in the controls (P < 0.0001).

In an unadjusted model the risk ratio for ICU admission among patients with COVID-19 was 2.20 [95% confidence interval (CI), 1.27–3.79] but was not statistically significant in a multivariable model (aRR = 1.34; 95% CI, 0.86–2.09) (Table 3). In adjusted models, COVID-19 infection was associated with an increased risk of serious complications (aRR = 7.02; 95% CI, 4.96–9.92) and death (aRR = 9.29; 95% CI, 5.68–15.21). In addition to COVID-19 status, ASA classification was significantly associated with ICU admission and serious complications.

Using the Chinese centers for disease control classification of disease severity, among the COVID-19 positive patients, the severity of disease was classified as mild in 16 (44.4%), severe in 6 (16.7%) and critical in 14 (38.9%). Within the COVID-19 positive cohort, symptomatic patients were more likely than asymptomatic patients to experience a severe complication (94.1% vs 26.3%; P < 0.0001) and in-hospital mortality (23.5% vs 10.5%; P = 0.03). When stratified by ASA classification, COVID-19 positive patients were at increased risk of severe complications regardless of preoperative ASA category (Supplemental Table 2 and Table 3, The risk ratio for death for COVID-19 positive compared to COVID-19 negative patients was 16.69 (95% CI, 4.08–68.26) for ASA 3 patients and 6.30 (95% CI, 0.70–56.35) for ASA class 4–5 patients (Supplemental Table 3, Similarly, when stratified by urgency of the procedure, COVID-19 patients who underwent both urgent and emergent procedures were at increased risk for serious complication. The risk ratio for death for COVID-19 positive compared to COVID-19 negative patients was 55.00 (95% CI, 17.22–175.66) for those who underwent urgent surgery and 3.40 (95% CI, 0.68–17.03) among those who underwent emergent operations. There were no statistically significant differences in the rates of serious complications or mortality between patients diagnosed with COVID-19 preoperatively compared to those diagnosed postoperatively.

These findings were robust in a series of sensitivity analyses. In a matched analysis, matches were identified for 97.2% of the COVID-19 positive patients (Table 4). The matched cohort was well balanced for ASA classification, procedure type, urgency of the procedure, sex, and age. The proportion of serious complications was 57.1% in COVID-19 positive patients compared to 14.3% in the COVID-19 negative cohort (P = 0.0006; OR = 8.50; 95% CI, 2.02–75.85) whereas the mortality rate was 17.1% versus 5.7%, respectively (P = 0.025; OR = 6.73; 95% CI 1.22-infinity). The 1 COVID-19 patient that was not matched experienced a serious complication but was not admitted to the ICU and was discharged from the hospital alive. Based on the ACS NSQIP Surgical Risk Calculator the predicted probability of serious complications in the COVID-19 patients was 12.4% whereas the observed risk in the cohort was 58.3%. Similarly, the predicted probability of death was 4.9% whereas the observed mortality rate was 16.7%. These findings were largely unchanged in another sensitivity analysis in which the window for COVID-19 positivity was limited to detection within 14 days of surgery (Supplemental Table 4,