Does COVID-19 infection change the need for future surgical interventions? An exploratory analysis.

Introduction

The coronavirus disease 2019 (COVID-19) global pandemic has led to fundamental shifts in everyday life around the world. Immense amounts of clinical research have examined the potential mechanisms and sequela of COVID-19 complications.¹ This includes a significant volume of research on acute, and long-term complications of COVID-19. As a previously unknown disease, an important part of this research process is exploratory analyses to determine which health conditions may be more prominent after a COVID-19 infection; this led to important studies which have identified the risk of long-term sequela such as neurologic, metabolic, cardiovascular and gastrointestinal disorders that are now recognized as being part of long-COVID-19.^2,3

For surgeons, most of the early study of COVID-19 centered around how surgical practices changed in response to the pandemic,⁴ safe operating room practices,⁵ or the abnormally high surgical mortality rate associated with operating on a COVID-19 patient.⁶ However, little research has examined if COVID-19 puts patients at increased risk of requiring specific surgical procedures after the period of acute infection. Given the wide-ranging impact of COVID-19 across several different organs, and the fact there is still much to be understood about long COVID-19, it is possible that patients with a previous COVID-19 infection are at an elevated risk of diseases that are managed operatively. Knowledge of this would be important for patients, physicians treating COVID-19 patients, and the surgical community.

Our objective was to use routinely collected administrative data to conduct an exploratory analysis of surgical procedures performed after a documented COVID-19 infection.

Methods

This is a retrospective cohort study. We made use of routinely collected data from the province of Ontario, Canada. All the datasets were linked using an encoded version of a unique patient identifier. All data was analysed at ICES Western. Ontario (Canada’s largest province with a population of 14.7 million people in 2020) has a single universal healthcare system, meaning this data is population based. The use of the data in this study was authorized under Ontario’s Personal Health Information Protection Act (Section 45), which does not require approval by a Research Ethics Board, or patient consent.⁷ A completed STROBE checklist, data collection plan and example analytic code for this study can be found as Extended data.²⁸

Results

We identified 30,284 people who had a positive COVID-19 test during the four-month study accrual period; of these 19,143 remained after our exclusion criteria. There were 528,581 people who had a negative COVID-19 test during the four-month study accrual period; 436,396 remained after applying the exclusion criteria, and 38,286 were matched to our COVID-19 positive patients (details in Figure 1).

Figure 1. Cohort creation flowchart.

The COVID-19 and control cohorts had similar age ranges and sex composition due to the matching. Other baseline variables were generally similar, other than a higher rate of chronic obstructive lung disease and heart disease among the control cohort (Table 1). The top ten surgical procedures that were identified were hand/wrist fracture fixation, Cesarean-section, ureteral stent placement, cholecystectomy, treatment of an upper tract urinary stone, hysterectomy, femur fracture repair, hip replacement, transurethral prostatectomy, and appendectomy. After adjusting for COPD cancer, and heart disease, there was a significantly reduced hazard of requiring upper tract renal stone surgery or ureteral stent placement, or undergoing a cholecystectomy among people who had a history of a positive COVID-19 test (Table 2).

Two post-hoc analyses were carried out after seeing our initial exploratory results. The first was measuring the use of nephrostomy tube insertions (OHIP code J046) during the same at-risk period as our primary outcomes (30 days after the index date until the end of follow-up). The COVID-19 cohort was significantly less likely to undergo this procedure (adjusted HR 0.251 (95% CI 0.075-0.840). The second was the frequency of requiring upper tract renal stone surgery or ureteral stent placement in the immediate 30 days after the index date; the adjusted HR for these outcomes during this earlier period were also significantly lower for the COVID-19 cohort (Table 2).

Discussion

In this study, we determined the relative risk of common operative procedures beyond the first 30 days of COVID-19 infection. For most of the surgical procedures, there was no significant increased risk, and many of these hazard ratios were close to one. We did however identify three surgical procedures that were significantly less likely to be carried out in patients who had a history of COVID-19 infection. To our knowledge, this has not been previously described.

A reduced risk of upper tract renal stone surgery or ureteral stent placement (HRs of approximately 0.50 for each) was observed. After seeing this result, we explored whether this could be due to an increased reliance on nephrostomy tube insertions (which generally only require sedation rather than a general anesthetic), however this procedure was rare and also significantly lower among COVID-19 patients. Similarly, the results were not explained by increased stone disease in the 30 days immediately after COVID-19 infection. Therefore, our results could be due to a reduced presentation of renal stones, a reluctance from urologists to intervene on stones after COVID-19 infection, or a lower propensity to form stones or have symptomatic renal stone disease after COVID-19. In general, the COVID-19 pandemic reduced the frequency of emergency room presentations of renal colic, and many hospitals setup triage procedures that precluded treatment of any non-urgent stone disease.¹⁹ This however should have impacted both the COVID-19 and control cohort. During this time period of the pandemic, people in general were less likely to go to the emergency room with renal colic and were more likely to avoid operative intervention.²⁰ This may have been amplified in post COVID-19 patients. It is likely that non-acute renal stone interventions were avoided among patients with a recent history of COVID-19 due to uncertainties about patient and operating room personnel safety, and due to the provincial operating room procedures.²¹ It is also possible that the specific renal effects of COVID-19 (mediated through the ACE receptors in the kidney) led to a lower propensity to develop kidney stones; the resulting kidney injury may have led to an increased focus on hydration after COVID-19 infection, patients may have experienced favorable anti-lithogenic electrolyte shifts during the acute infection, or care processes around post-COVID-19 infection follow-up may have decreased the risk of renal stone development and subsequent intervention.^22,23 Magnetic resonance imaging (MRI) abnormalities have also been detect in the kidney after COVID-19 infection, and these changes may impact renal stone risk.²⁴

The reduced risk of cholecystectomy was an interesting finding. Abdominal imaging in COVID-19 patients is more likely to detect biliary sludge and gallstones,²⁵ which would lead to a hypothetically increased risk of future complications and need for cholecystectomy (contrary to our findings). Similar to renal stones, it is possible that conservative management was the goal for the initial period after COVID-19, and surgical interventions were deferred for the same reasons previously discussed.²⁶

Some limitations of our study are important to acknowledge. The pandemic started in Canada in February of 2020, therefore our period of follow-up after COVID-19 infection is limited in our study to at most 10 months. Many hospitals in Ontario instituted surgical triage protocols which affected the delivery of elective care; however, our surgical procedure list does contain a mix of procedures which are traditionally elective and others that are more urgent. The time period of the study was characterized primarily by COVID-19 wildtype infections, and it is possible that the subsequent variants have different effects. People can have asymptomatic COVID-19, and early in 2020 testing for COVID-19 was limited to those who met specific criteria; therefore, some people with a prior COVID-19 infection may have been misclassified and included in the control group. The risk of this should be low, as a study of healthcare workers (a group that did have early access to testing and more stringent screening for symptoms) found that 3% had COVID-19 antibodies in the absence of a documented COVID-19 positive test.²⁷ Finally, all observational studies may have residual confounding.

Conclusions

After a COVID-19 infection there is not an increased risk of needing common surgical procedures. There appears to be a reduced risk of renal stone disease treatment and ureteral stent placement, and a reduced risk of cholecystectomy, however understanding the reasons for this will require more study.

Data availability