Introduction

The social, preventive and mandatory isolation due to the COVID-19 pandemic produced significant changes in the care and treatment of surgical patients^1-3. Healthcare institutions and their workers had to adapt to this reality, which led to situations of extreme stress and exhaustion and profound alterations in the family, social and economic spheres^4-6.

After one year of the pandemic, several of these aspects remain unresolved, the available data (complications and mortality) are limited and of low evidence, collateral damage has not been quantified, the guidelines are based on expert opinion, and there is uncertainty about new outbreaks^7-15.

For these reasons, the aims of the present study were: 1) to quantify volume of care in terms of clinic appointments, hospitalizations, surgeries, endoscopies, and bed occupancy during the year of COVID-19 pandemic and compare it with the same period without pandemic; 2) to estimate the healthcare and institutional impact; and 3) to make univariate and multivariate comparisons between COVID-positive (COVID+) patients vs. COVID-negative (COVID-) patients to determine the postoperative complications and mortality, and associated risk factors.

Material and methods

We conducted an observational, descriptive and retrospective cohort study, comparing data obtained during the pandemic (from March 19, 2020, to March 18, 2021) with a similar period without the pandemic (from March 19, 2019, to March 18, 2020). To evaluate the volume of care, we counted the number of clinic appointments, hospitalizations, surgeries, endoscopies, and days of bed occupancy.

To analyze postoperative complications and mortality among COVID+ and COVID- patients, we analyzed data matched by a ratio of 1:2 evaluating the following variables: age ≤ versus ≥ 75 years, sex, planned or emergency admission, benign diseases or cancer, operated patients versus non-operated patients, comorbidities, postoperative complications, admission to the intensive care unit (ICU) and need for mechanical ventilation (MV).

In operated patients, we analyzed type of anesthesia (general of blockade), type of approach (laparotomy or laparoscopy) and ASA grade (≤ 2 vs. ≥ 3). COVID+ patients were analyzed by separately. The patients included were >18 years, with a diagnosis confirmed by quantitative PCR or chest tomography or clinical suspicion, evaluated by a specialist, hospitalized due to gastrointestinal diseases or abdominal wall defects requiring surgery. Patients with clinical findings that were not confirmed in subsequent tests were excluded. Data were incorporated into a Microsoft Excel 97^® database and the variables were dichotomized and analyzed using Stata 14.0^® statistical software package. The chi-square test and Fisher’s exact test were used to compare differences between the groups that did not meet the assumptions.

Logistic regression analysis was considered with the result “risk factors for mortality” in COVID+ patients vs. COVID- patients.

A p value < 0.05 with a 95% confidence interval was considered statistically significant.

Results

All the variables analyzed decreased during the pandemic period (Table 1).

The statistical analysis of surgical admissions, surgeries and endoscopies, according to whether they were performed on an emergency or elective basis, showed a statistically significant reduction in the variables planned admissions, scheduled surgeries and non-urgent endoscopies (p = 0.0001). The same variables performed on an emergency basis also decreased without reaching statistical significance (Table 1 and Table 2).

Of the 979 admissions, 41 corresponded to COVID+ patients (4.1%). The observed mortality was 29.2% in COVID+ patients (12/41) and 7.3% in COVID- patients (6/82), and the difference was statistically significant (p = 0.021). The univariate analysis comparing COVID+ patients (41 patients) versus COVID- (82 patients) determined higher risk of mortality for COVID+, age ≥ 75 years, male sex, emergency admission, presence of pneumonia, requirement of ICU and MV. Surgeries were associated with a higher number of respiratory complications (pneumonias) (p = 0.0091) and with a non-significant increase in mortality (Tables 3 and 4).

Those variables without collinearity underwent multivariate analysis which identified COVID+ and requirement of MV as significant factors (Table 5). When multivariate analysis was performed only on COVID+ patients, requirement of MV explained the higher mortality (Table 6).

Discussion

Several controversies arose regarding the different aspects of medical care during this pandemic. We intend to analyze and discuss those related with the essence and objectives of our presentation. For this purpose, we will base on data from the Argentine and international literature, and on our own experience and results.

So far, few papers have analyzed the impact of health policies on health care services. Pirracchio et al. published the experience, after the massive suspension of surgeries implemented in the United States. They observed a sharp decline starting in February, that was maximal in March and a gradual increase starting in May 2020. There was significant heterogeneity in the results according to the institution involved and the local situation and recommended that future cancellation of services should respond to the local epidemiological context of hospitals¹.

Some private centers in the metropolitan area of Buenos Aires published their results after comparing the number of visits, hospitalizations and interventional procedures performed between April 1 and 30, 2019, with the same period in 2020. The authors reported an overall decrease of 75% in consultations, 80% in endoscopies, 48% in hospitalizations and 16% in the use of chemotherapy².

Huespe et al. observed a significant decline in the number of consultations and procedures performed in a percutaneous surgery unit by comparing the period between March 20, 2020 and May 10, 2020, with the same period in 2019. There was a significant decrease in scheduled interventions, but the impact was lower in oncology cases³.

In conclusion, most reports reveal a decrease in services from 40 to 80%^1-3. Our overall results are consistent with the mentioned reports and are described in the results of the study. The collateral damage to the population and institutions has not been well determined yet. Several studies from India and Italy show higher mortality from causes related with the interruption of activities and home isolation⁴. Wang et al. concluded that the main risk for patients with cancer was the limitation of medical services, especially in areas affected by the coronavirus⁴. In Italy, New York and California, out-ofhospital deaths in COVID- patients increased compared with other periods⁵. Surgical consultations decreased by 33.3% in Germany⁵. In Scotland, deaths from cancer, dementia and circulatory diseases increased⁴.

Pirracchio et al. concluded that more flexible measures of isolation and medical care in the United States would help to avoid the cancellation of surgical cases that affect the prognosis of patients¹.

For Bozovich et al., cancellations or delays in consultations and treatments would bring undesirable consequences in patients with pre-existing diseases, or in those susceptible to develop them or with conditions requiring urgent treatment, especially in cardiovascular and neurological diseases and neoplasms. They consider continuity of care is essential to avoid collateral damage from COVID-19².

Dreifuss et al. evaluated the consequences of delays in consultations for acute appendicitis and observed a significant increase in the number of complicated cases and severe peritonitis. They concluded that isolation and the population’s fear of the hospital environment might have contributed to the progression of the disease⁶. Gondolesi et al. reported that the number of liver, biliary and pancreatic surgeries, resection of hepatocellular carcinoma and liver transplantation were reduced by 47%, 49%, 31% and 36%, respectively. They considered that this reduction mostly affected centers with high bed occupancy due to COVID-19 and that the ultimate long-term impact should be evaluated⁷. At the institutional level, Pirracchio et al. emphasized that adopting extreme measures as not providing health care or canceling surgeries jeopardized hospitals’ financial security¹. Other institutions and associations also provided results and recommendations^8-14. Asociación Argentina de Cirugía was not absent from this problem and, in an editorial, Enrique Ortiz commented that our association was aware of the impact that this pandemic had on surgeons and their families, and of the risks to their health and the economic situation¹⁵. In our experience, the institutional impact of the data collected revealed a 59.3% reduction in the number of consultations, which, translated to the activity of the outpatient clinic, represents 5.9 months without providing care. The 49% decrease in surgeries is equivalent to closing the operating room for 5.4 months. A 79% reduction in the number of endoscopies represents 6.3 months of inactivity. Finally, bed occupancy was 40% lower. Based on the overall analysis of these presentations and on our experience, we can infer that the limitation or lack of medical care increased the severity, complications, morbidity and mortality of conditions requiring emergency surgery and of cancer patients. The need for quantifying postoperative complications, mortality and associated risk factors in COVID+ vs. COVID- patients was another aspect to discuss. Before the pandemic, several observational studies established postoperative pulmonary complication rates of up to 10%, with a mortality of 4%. These values changed dramatically during the COVID-19 pandemic. The COVIDSurg Collaborative group reported pulmonary complications (severe pneumonias) in 51% of those who underwent surgery¹⁶.

Knisely et al. reported 56% of perioperative complications in COVID+ patients. Pneumonia occurred in 50% of COVID+ patients vs. 2.8% in those COVID- (p = 0.0001)¹⁷. For Dogiletto et al., pneumonia was the most common and significant complication¹⁸. Lei et al. analyzed 34 COVID+ patients operated on. All of them developed pneumonia, 15 required admission to ICU and 7 died¹⁹. In our series, pneumonia was the most common complication, occurring in 85% of COVID+ patients vs. 11% in COVID- cases (p = 0.0001). Perioperative infection due to COVID-19, advanced age and surgery were risk factors for this complication. Admission to ICU and the need for MV were associated factors in COVID+ patients with pneumonia (p = 0.0089 and 0.0091). The usual postoperative complications (infections, abscesses, fistulas) remained at the usual values, with no differences between the groups. The reported mortality in COVID+ patients undergoing surgery ranges from 4.4% to 27.8%. Doglietto et al. reported mortality of 19.5% in COVID+ versus 2.4% in COVID- (p = 0.001)¹⁸.

In the publication by Moreira et al. overall mortality was 23.8% and was 25.6% in emergency procedures²⁰, while Knisely et al. found a mortality rate of 16.7% in COVID+ vs. 1,4% in COVID-, and this difference was statistically significant¹⁷. Of 44 cases operated on, mortality was 20.5% in the study by Lei et al¹⁹. In COVID+ patients undergoing transplantation, Fernandez Ruiz et al. reported a mortality rate of 27.8%²¹. The mortality rate reported by the COVIDSurg Collaborative group was 23.8%¹⁶. The lowest mortality was published by Pinares Castillo et al.: 4.4% (most were appendectomies in young patients)²². In our series, mortality was 29.2% in COVID+ (12/41) and 7.3% in COVID- (6/82) and was statistically significant. Knisely et al. found that surgery, pneumonia, need for ICU and MV were significant predictors of mortality¹⁷. The COVIDSurg Collaborative group described age, male sex, ASA grade ≥ 3, neoplasms and emergency surgery as determinant factors¹⁶. In this experience, only ASA ≥ grade 3 was an independent predictor of mortality¹⁶. Pinares Carrillo et al. and other authors added endotracheal intubation during general anesthesia, operative time and comorbidities as factors to consider²². In our series, male sex and age ≥ 75 years were determinants of overall mortality (especially in severe cases and in patients who underwent surgery). The presence of comorbidities (cardiovascular diseases, respiratory diseases and diabetes) was associated with a non-significant increase in mortality. Emergency hospitalizations and the need for surgery resulted in higher mortality in the COVID+ group than in controls but was only significant for emergencies. In our patients operated on, mortality was higher in laparotomies with general anesthesia and ASA grade ≥ 3, without reaching statistical significance. When we performed logistic regression with these variables, only COVID+ and MV remained statistically significant. In COVID+ patients, pneumonia was the complication with the highest impact on mortality (p = 0.0001), followed by ICU stay and requirement of MV. In multivariate analysis, requirement of MV was an independent predictor of mortality. Several authors wondered whether mortality varied in COVID+ patients with and without pneumonia, and with the severity of the pulmonary infection. The published results in terms of morbidity and mortality differ depending on the presence or absence of pneumonia and if it is clinically mild or severe^23-27.

Melendi et al. made a prospective analysis of a cohort of COVID+ patients with and without pneumonia²⁸. None of the patients without pneumonia developed critical disease or died.

The patients with pneumonia developed severe disease (33.3%) that was critical in 9.5%, and 9.1% were transferred to the ICU. Mortality rate was 6.6%²⁸. The COVIDSurg Collaborative group reported pulmonary infections in 51% of those who underwent surgery; mortality rate in those with severe pneumonia was 38%¹⁶. In the publication by Moreira et al. more than half of the deaths were caused by severe pneumonia²⁰. Our group divided patients with pneumonia into mild cases (21) and severe cases (14). Of the 14 severe patients, 12 (87%) were transferred to the ICU compared with only 1 of the mild cases (5.7%) (p = 0.0012). Of those admitted to the ICU, 9 required MV. Ten patients with severe pneumonia and 1 with mild pneumonia died (p = 0.00001).

As a final reflection, we can state that we walked through a year of pandemic full of uncertainty and with few certainties, we learned from our own mistakes and from those of others when facing an unknown disease but passing of time also provided us with experience and allowed us to analyze results. Since then, we were able to change some initial strategies. We are now going through a new period in which our healthcare activity is almost back to normal, but we are threatened by other dangers (the possibility of new outbreaks or new viral strains) that force us to remain alert.

Conclusions

The following conclusions are the result of the experience after one year of analysis, in a high complexity institution in an area where the pandemic had a high impact (Región Sanitaria VI).

1. There was an overall decrease in services, but it only reached statistical significance for planned hospitalizations, elective surgeries and non-urgent endoscopies.

2. COVID+ patients had greater rate of postoperative complications and mortality than the control group (p = 0.021).

3. Pneumonia was the most common postoperative complication in COVID+ patients, especially in the elderly and in those who underwent emergency surgery. Severe cases were associated with greater need for ICU admission and MV and higher mortality.

4. On multivariate analysis COVID+ and requirement of MV resulted significant independent predictors of mortality.

5. Mortality analysis only in COVID+ patients indicates that severe pneumonia was the major complication (p = 0.0001) and the need for MV was the significant independent factor.