SURgical Outcomes in COvid Patients - the SUROCO Multicenter Cohort Study (COVID-19)

METHODS Objectives

Primary: To describe postoperative outcomes in patients with COVID-19 undergoing a surgical procedure and measure the association between presence of symptoms and postoperative outcomes.

Secondary: To describe postoperative outcomes in patients undergoing a surgical procedure after they have healed from COVID-19 and measure the association between time from first positive test and postoperative outcomes.

Tertiary: To compare outcomes between symptomatic and asymptomatic COVID-19 patients undergoing a surgery with those who have healed from it.

Exposure variables:

For the primary objective, the exposure of interest will be the presence of symptoms at time of surgery. For the secondary objective, the exposure will be the length of time between first COVID-19 positive test and the date of surgery. For the tertiary objective, the status of the disease will be our exposure of interest.

Variable definitions:

All COVID-19 variables will be based on the first day of diagnosis (either clinical or by RT-PCR from either an oronasopharyngeal swab or an endotracheal specimen). The investigators will also capture asymptomatic SARS-CoV-2 carrier. Recovery at surgery will be defined as having two negative SARS-CoV-2 PCR tests between the last positive one and the surgery, or 1 negative test and a clinically significant period without symptoms between the last positive test and the negative one or no negative test but considered as recovered by the clinical team because of complete symptoms resolution.Criteria for "healed" status changed in November 2020 and were less restrictive (as detailed in table 1 of the protocol). No minimal or maximal delay will be applied between SARS-COV-2 infection and surgery for inclusion.

Surgical disease will be defined as the one reported by the surgeon. Surgeries will be further categorized into the following categories: neurosurgical (head, neck and spine), cardiac, thoracic, major vascular (intrathoracic and/or intra-abdominal), non-vascular abdominal (laparotomy or laparoscopy), urogenital (urology and/or gynecology), non-spine orthopedics, peripheral vascular, ENT and other. Any neurosurgical, cardiac, thoracic, major vascular or non-vascular abdominal surgery will be defined as a major surgery. The investigators will capture the surgical approach (minimally invasive or invasive) and the urgent nature of the procedure. Minimally invasive surgery will be defined as a one that does not need to open an anatomical cavity (laparoscopy, thoracoscopy, endoluminal). Urgency of surgery will be defined as one that needs to be done within 24 hours (emergent or urgent), or can wait for more than 24 hours (non-urgent).

The investigators will measure hospital survival. The investigators will also capture survival up to 30 days and 6 months using regular hospital follow-up. The investigators will use existing definitions for the postoperative pulmonary complications outcome. Non-pulmonary infectious complications will be defined as any infection that needed antibiotics for more than 72 hours. Acute kidney injury will be defined by the creatinine difference of the KDIGO-AKI criteria. Myocardial infraction and stroke will be defined as reported by treating physicians. Mechanical ventilation will be defined as any mechanical support (non-invasive or invasive). All 30-day outcomes will include any day with the outcome and the 30-day organ dysfunction free days will use existing definitions.

Sample size:

The investigators plan a convenient sample size depending on the number of patients and center included. Inclusion will start on the 13th of March 2020.

Analyses:

Primary objective The investigaors will first conduct descriptive analyses of baseline characteristics and outcomes of COVID-19 confirmed patients based on the presence of symptoms at the time of surgery (symptomatic versus asymptomatic). The investigators will report descriptively the following outcomes in both patients' population: proportion of patients with pulmonary complications, thromboembolic complications, non-pulmonary infectious complications, need for a new postoperative ICU admission, mean hospital length of stay, 30-day mechanical ventilation free-days and 30-day organ dysfunction free days. . To evaluate if presence of symptoms is associated with hospital survival in SARS-COV2 infected patients for the primary analysis, investigators will fit a multivariable Cox model that will include presence of symptoms, preoperative hospitalization, preoperative respiratory support need (invasive mechanical ventilation, oxygen therapy, no oxygen), urgency of surgery, importance of surgery (major versus minor) as independent variables and a frailty factor as a center random effect. If too few events our observed, investigators will limit covariables in their model.

Secondary objective The investigators will report descriptively patients who have recovered by categorizing them based on the time since first positive test (categorized as 0-4 weeks, 4-8 weeks and > 8 weeks). The investigators will report the same outcomes as previously mentioned. To analyze the effect of recovery time on postoperative survival, the investigators will fit a similar multivariable survival model . The investigators will include time since first positive COVID-19 test, preoperative hospitalization, preoperative respiratory support need (invasive mechanical ventilation, oxygen therapy, no oxygen), urgency of surgery, importance of surgery (major versus minor) as independent variables and a frailty factor as a center random effect. Investigators will explore non-linear relationship between time and the outcome.

Tertiary objective The investigators will report descriptively patients who have recovered and asymptomatic COVID-19 patients. To compare if those two groups of patients are different regarding hospital survival, the investigators will fit a multivariable survival model that will include status (recovered versus asymptomatic carrier), preoperative hospitalization, preoperative respiratory support need (invasive mechanical ventilation, oxygen therapy, no oxygen), urgency of surgery, importance of surgery (major versus minor) as independent variables and a frailty factor as a center random effect.