Clinical-epidemiological changes in patients with non-traumatic acute abdomen during the COVID-19 pandemic: a retrospective study

ABSTRACT Objective: we intend to demonstrate the clinical alterations and the postoperative evolution in patients with acute abdomen non-traumatic in conservative or surgical treatment during the pandemic compared to a similar period in the last year. Method: a single-center retrospective study, including patients who received clinical-surgical treatment at Hospital do Trabalhador diagnosed with acute abdomen between March and August 2020 and a similar period in 2019.Variables studied ranged from demographic data to indices of social isolation. Results: 515 patients were included, 291 received treatment in a pre-pandemic period and 224 during. There was not statistical difference in relation to comorbidities (p=0.0685), time to diagnosis and seeking medical help. No statistical differences were observed in terms of days of hospitalization (p = 0.4738) and ICU need (p=0.2320). Regarding in-hospital deaths, there was statistical relevance in the age above 60 years (p=0.002) and there were more deaths during the pandemic period (p=0.032). However, when we analyze the factors associated with the number of days until diagnosis by a physician, there was no statistical difference. Conclusion: the analyzed data showed that the pandemic period and age over 60 years were the variables that increased the odds ratio for the in-hospital death outcome. However, the length of stay, days in intensive care unit and postoperative surgical complications showed no significant difference.


INTRODUCTION
I n late 2019, a novel coronavirus (SARS-CoV-2) was identified as a cause of pneumonia and severe acute respiratory syndrome.After rapid spread, on March 11,   2020, the World Health Organization (WHO) classified the situation as a global pandemic 1 .Since the first case of COVID-19, all health systems have adapted to this new dynamic, and new recommendations and protocols were developed 2,3 .Surgical departments were affected, recovery beds were transformed into ICU (intensive care unit) ones, elective procedures were postponed, and members of the surgical teams were allocated to reinforce the ICU 4 , certainly contributing to the decrease in the number of surgeries.
In this scenario, acute abdominal cases, conditions with potential complications and death in a short period of time, continue to appear 5 .Complications can present rates between 8.2 and 31.4%, and mortality is quite variable, being directly related to the underlying cause 6,7 .The burden on the health system caused by the COVID-19 pandemic results in difficult access and delay in the diagnosis and treatment of cases of acute abdomen, which can contribute to the increase in morbidity and mortality of diseases that require urgent surgical treatment.Indices of social isolation were also collected in the state where the hospital in question is located (Paraná), available on the public data platform "inloco"10.
These data were obtained for the months of March to August 2020 (during the pandemic), with average indices calculated for each study month.
The collected data were then analyzed using the statistical software STATA v14.2 11 .For descriptive analysis, we expressed measures of central tendency and dispersion as mean and standard deviation (mean ± SD) for continuous variables with normal distribution and as medians, minimum and maximum values (median, minimummaximum) for those with non-normal distribution.Categorical variables were expressed as absolute and relative frequencies.For inferential statistical analysis, we compared groups using the unpaired Student's t test for continuous dependent variables and the chisquare test for binary or categorical dependent variables.Finally, for the unpaired analysis of independent categorical variables, we used the KruskallWallis's test.

Multivariate logistic regression and linear regression
were used to identify factors related to hospital death (categorical dependent variable) and days between onset of symptoms and seeking medical care (continuous dependent variable), respectively.The regression models and their parameters were developed based on a model with biological and epidemiological plausibility, as well as using the Akaike Information Criterion (AIC) 12 .A significance level of 5% was considered for this study.

RESULTS
A total of 515 patients were included, with 291 attended in the pre-pandemic period and 224 in the COVID-19 pandemic period.The number of male patients was higher in both groups (Table 1).The age distribution was statistically different between the groups (p=0.036),though both showed a preponderance of non-elderly adults (<60 years).
There was no significant difference between the groups regarding previous comorbidities measured by the CCI (p=0.0685), as well as the presence of previous abdominal surgery (p=0.2700).There was also no difference in the time between the onset of symptoms and diagnosis when comparing the pre-pandemic (3.89 ± 6.11 days) and pandemic (3.80 ± 12.56 days) groups (p=0.9108).
The frequency of each diagnosis of acute abdomen among the patients included in this study was also similar between the groups (p=0.2200).Acute appendicitis and acute cholecystitis were the two most prevalent diagnoses in both groups.Regarding hospital course (Table 2), the average of hospitalization days was similar between groups, 4.16 ± 5.30 days pre-pandemic and 4.50 ± 5.49 days during the pandemic.There was also no statistically significant difference between the groups regarding the use of antibiotic therapy, the need for a surgical approach, or the prevalence of different surgical techniques (open versus laparoscopic).
As for the need for intensive care, the percentage of patients who required ICU was slightly lower during the pandemic, while the average length of stay in the ICU in days was slightly longer.However, these differences were not statistically significant.
Finally, there was also no difference between the groups regarding complications during hospitalization according to the Clavien-Dindo classification (p=0.6960).In both groups, grade III B was the most prevalent, characterizing surgical, endoscopic, or radiological intervention under general anesthesia.
Six patients died in the hospital in the pre-pandemic period and nine during the pandemic, with no statistical difference.
We selected a logistic regression model using the independent variables sex, age, CCI, and group (pre-pandemic vs. pandemic).The dependent variable for this model was hospital death (Table 3).
In In addition, after adjusting for covariates, being seen during the pandemic proved to be a factor associated with in-hospital death (Odds Ratio 3.54, 95% CI 1.1112.27,p=0.032).
We then built a linear regression model using the independent variables sex, age, CCI, and social isolation index per month (during the pre-pandemic period).The dependent variable of this model was the number of days between the onset of symptoms and diagnosis (Table 4).None of the variables was associated with delay to diagnosis, either in the univariate analysis or after adjusting for covariates.

DISCUSSION
With the evolution of the COVID-19 pandemic, several aspects of care for acute abdominal disorders were adapted to optimize available resources.
International surgical societies have published new protocols with guidelines on surgical treatment versus conservative treatment in cases of acute abdomen during the pandemic 13-15.The pandemic has brought unprecedented challenges to health systems 16 , but even after more than 24 months, there are still few reports of its effect on the care of acute abdomen cases.Some authors have reported delays in medical care for the pediatric population 17,18 , which may be related to parents' fear of submitting their children to a hospital environment during the pandemic 19 .In addition, the profile of patients admitted to a hospital environment has changed 20 , reducing the number of surgical patients in several hospitals5.Perhaps due to more restrictive measures, there was a decrease in the number of patients hospitalized for non-traumatic causes in South Africa 20 .Other international studies show a reduction in emergency admissions, such as an Italian study 21 and a North American one 22 .A study from New Zealand, for example, found a reduction of 26% 23 .In our study, despite the decrease in the absolute number of patients, this difference was not statistically relevant.
During the pandemic, patients under 60 years of age without previous abdominal surgery predominated, with an average of 3.80 days from the onset of symptoms to diagnosis.These data are statistically similar to the pre-pandemic period.As for diagnoses, appendicitis and acute cholecystitis predominated in the two periods studied.Thus, we observed that patients continued to seek medical help, despite the restrictions imposed by social isolation.The reasons may be the maintenance of access to emergency rooms even with the increase in hospital occupancy and to the milder isolation restrictions applied in the institution's city and state, or even to the communication carried out by the media in general and through medical channels, guiding the population.
When evaluating data on length of stay, days in the intensive care unit, and number of deaths, we found no significant difference between the groups.Such findings are similar to data on mortality and length of stay found in the literature, in which there was also no statistical difference during the pandemic period 23 .
We used the Clavien-Dindo classification of surgical complications to verify the presence and severity of postoperative complications, and we did not obtain statistical difference between the groups, the III B classification being the most prevalent in both, which characterizes the need for surgical, endoscopic, or radiological intervention under general anesthesia.
When we turned to the literature to make a comparison between the period before and during the pandemic, British studies found similarities and good results in surgeries for acute appendicitis 24,25 , although some trials showed higher mortality in patients undergoing surgical procedures during the pandemic 26 .
However, when using logistic regression models, as shown in Tables 3 and 4, we obtained interesting findings.did not interfere with death, but being over 60 years old displayed an odds ratio of 13.96 times more in-hospital deaths, and being treated during the pandemic had a mortality risk 3.54 times greater than patients seen in the pre-pandemic period.This last finding brings an important reflection: if there was no statistical difference between the time until diagnosis and there were no differences in postoperative complications, as seen in the data in Tables 1 and 2, this difference could be related to in-hospital difficulties due to overcrowding and scarcity of resources.
Apart from the first two months of the pandemic, the team always worked with limited capacity, due to the demand from COVID, trauma, and non-traumatic acute abdomen, but we did not find detailed monthly data on scarcity of resources and intra-hospital difficulties.
In Table 4, we evaluated the variables age, sex, and social isolation index with the number of days from the onset of symptoms to hospital care and we found no association between the variables and the outcome.We could infer that social isolation did not delay the search for hospital care, as the elderly continued to seek the emergency room during the pandemic.
Our work has some limitations inherent to the study design, as it is a non-randomized study in only one institution and with a limited number of patients.
However, it is valid as a portrait of the possible effects of the pandemic on the care of acute abdominal disorders in Brazil and as a support for future work to be carried out with larger samples and with more participating institutions.

CONCLUSIONS
During the pandemic, patients over 60 years old had a higher risk of complications and death than in the pre-pandemic period.However, we did not find changes regarding the time of clinical history and diagnosis, nor there were any changes regarding the length of stay, days in the intensive care unit, and postoperative surgical complications when comparing the two periods studied.

R E S U M O R E S U M O
Therefore, this study aims to assess possible changes in the clinical course and postoperative outcomes of patients with non-traumatic acute abdomen undergoing surgical or conservative treatment during the patients who met such criteria and who were seen between March to August 2019 and March to August 2020.We excluded trauma patients, pediatric patients, individuals undergoing emergency surgery related to previous elective surgeries, and those with chronic abdominal pain (without concomitant acute complications), in addition to patients treated outside the pre-established preoperative period.We set up two study groups according to the date of medical care.Patients seen between March 2020 and August 2020 constituted the group of patients seen during the COVID-19 pandemic.Another comparative group was formed by patients seen in a similar period, between March 2019 and August 2019, constituting the pre-COVID-19 pandemic group.Only patients who had respiratory symptoms or symptoms compatible with COVID-19 were tested for the identification of the new coronavirus.Recorded variables included patient demographics, clinical-surgical diagnosis, time between symptoms onset and diagnosis, comorbidities (calculation of the Charlson Comorbidity Index, CCI), length of stay, in-hospital complications (ClavienDindo score), and surgical treatment.The CCI is one of the most used comorbidity indexes to predict mortality, identifying the present comorbidities and applying weights to these diseases, that is, mild (1-2), moderate (3-4), and severe (≥5) 8 .The ClavienDindo score was a standardized classification proposed in 1992 and revised in 2004, with severity levels based on the therapeutic intervention applied to the management of surgical complications 9 .

Table 1 -
Distribution of 515 patients with acute abdomen according to clinical characteristics, divided into two groups, pre-pandemic and during the pandemic.
the univariate analysis, the factors age ≥60 years

Table 2 -
Complications and hospital course pre-and during the pandemic.

Table 3 -
Logistic regression, factors associated with in-hospital death.

Table 4 -
Linear regression, factors associated with the number of days until diagnosis.