Surgical management of acute appendicitis during the European COVID-19 second wave: safe and effective

Introduction The COVID-19 (SARS-CoV-2) pandemic drove acute care surgeons to pivot from long established practice patterns. Early safety concerns regarding increased postoperative complication risk in those with active COVID infection promoted antibiotic-driven non-operative therapy for select conditions ahead of an evidence-base. Our study assesses whether active or recent SARS-CoV-2 positivity increases hospital length of stay (LOS) or postoperative complications following appendectomy. Methods Data were derived from the prospective multi-institutional observational SnapAppy cohort study. This preplanned data analysis assessed consecutive patients aged ≥ 15 years who underwent appendectomy for appendicitis (November 2020–May 2021). Patients were categorized based on SARS-CoV-2 seropositivity: no infection, active infection, and prior infection. Appendectomy method, LOS, and complications were abstracted. The association between SARS-CoV-2 seropositivity and complications was determined using Poisson regression, while the association with LOS was calculated using a quantile regression model. Results Appendectomy for acute appendicitis was performed in 4047 patients during the second and third European COVID waves. The majority were SARS-CoV-2 uninfected (3861, 95.4%), while 70 (1.7%) were acutely SARS-CoV-2 positive, and 116 (2.8%) reported prior SARS-CoV-2 infection. After confounder adjustment, there was no statistically significant association between SARS-CoV-2 seropositivity and LOS, any complication, or severe complications. Conclusion During sequential SARS-CoV-2 infection waves, neither active nor prior SARS-CoV-2 infection was associated with prolonged hospital LOS or postoperative complication. Despite early concerns regarding postoperative safety and outcome during active SARS-CoV-2 infection, no such association was noted for those with appendicitis who underwent operative management.


Introduction
Acute appendicitis is one of the most frequent surgical emergencies, and its management is one of the most commonly-performed emergency general surgery procedures [1][2][3][4]. While predominantly affecting younger patients [5], appendicitis may occur at any age and exhibits substantial variation in symptoms and severity [1,6]. Similarly, the etiologies of appendicitis span lymphoid proliferation to appendicolith-associated obstruction and suppuration, to malignant obstruction. Accordingly, a wide range of clinical management approaches are utilized and reflect, in part, clinical equipoise regarding a single optimal management strategy [7].
The initial wave of the SARS-CoV-2 (COVID-19) pandemic raised concerns regarding transmission of infection to operating team members during aerosol generating procedures such as airway control, tracheostomy, endoscopy, or laparoscopy. Concomitantly, safety concerns also surfaced during the COVIDSurg study during the initial phase of the pandemic regarding the advisability of undertaking operative management for patients acutely infected with SARS-CoV-2 [8]. Therefore, acute care surgeons pivoted from established practice patterns to pursue either delayed operative management or, for certain conditions such as appendicitis, non-operative management [10][11][12]. Importantly, the initial recommendations to pursue non-operative and antibiotic-driven care for those with appendicitis were articulated ahead of an evidencebase documenting enhanced safety. Furthermore, such recommendations may have been predicated upon the anticipation of a short time course for SARS-CoV-2 infection-an assumption that has been well disproved by virus variant evolution and multiple subsequent waves of infection. At the same time, knowledge and experience in caring for patients with SARS-CoV-2 infections improved and global vaccination programs were enacted. Therefore, it is worthwhile examining whether appendicitis patients who underwent appendectomy during later pandemic waves and who were actively infected with SARS-CoV-2, or had been previously infected with SARS-CoV-2, demonstrated prolonged hospitalization related to complications, compared to patients without SARS-CoV-2 infection.

Methods
This multi-center cohort study adhered to the standardized methodology for snapshot audits [9]. All centers received exemption from informed consent approval from the relevant institutional review board or equivalent as an observational study. All data were anonymized for entry into a secure user-encrypted server running on the Smart-Trial ® web application [9]. This study was also conducted in accordance with the Declaration of Helsinki as well as the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [10].
Any site treating emergency general surgery patients was eligible to participate in data collection. No minimum case volume, or center-specific limitations were applied. Data collection was conducted according to a predefined protocol registered with ClinicalTrials.gov (Trial # NCT04365491). The protocol and an invitation to participate was shared by email with registered members of the European Society of Trauma and Emergency Surgery and through national surgical societies. The study enrolled all consecutive patients aged 15 years or older admitted with acute appendicitis during a 3-month window (November 1, 2020-May 28, 2021); enrolled patients were followed for 90 days postoperatively. Data collected included age, sex, American Society of Anesthesiologists (ASA) classification, a history of previous abdominal surgery, ischemic heart disease, insulin-dependent diabetes, congestive heart failure, chronic kidney disease, current smoking status, immunosuppression, the American Association for the Surgery of Trauma (AAST) appendicitis grade, time to surgery from admission, laparoscopic surgery, conversion to open surgery, open surgery, admission white blood cell count and neutrophil percent, admission C-reactive protein concentration, as well as the country where the surgery was performed (Table 1). SARS-CoV-2 seropositivity was determined with a screening PCR or antigen test on admission, based on institutional protocols. The database was closed for analysis on August 31, 2021. Patients were grouped according to SARS-CoV-2 seropositivity into three sets: uninfected, actively infected, previously infected and recovered; patients whose seropositivity status were unknown or unreported were excluded. Only patients admitted with appendicitis who also underwent appendectomy (laparoscopic or open)as opposed to non-operative management-formed the dataset explored in this study. Patient demographics, type of operation, hospital length of stay (LOS), as well as complication occurrence, were abstracted from the database for analysis parsed by SARS-CoV-2 seropositivity.

Statistical analysis
Patient data within each group were summarized as means and standard deviations (SDs) for continuous normally distributed variables, medians, and interquartile ranges (IQRs) for continuous non-normally distributed variables, as well as counts and percentages for categorical variables. An analysis of variance (ANOVA) or a Kruskal-Wallis test was used to assess for differences in continuous variables. A Chi-square The association between a patient's SARS-CoV-2 seropositivity and complications was analyzed using Poisson regression models with robust standard errors to account for heteroscedasticity [11]. Any complication or severe complications was the dependent variable, while the predictors were the patient's SARS-CoV-2 seropositivity along with potential confounding variables ( Table 2). Results are presented as incident rate ratios (IRRs) and 95% confidence intervals (CIs). The relationship between a patient's SARS-CoV-2 seropositivity and LOS was explored using a quantile regression model, using LOS as the dependent variable, while SARS-CoV-2 seropositivity and potential confounding variables were included as explanatory variables (Table 3). Results are presented as the change in median LOS along with 95% CIs.
A two-tailed p value < 0.05 was considered statistically significant in all analyses. Missing data were managed using multiple imputation by chained equations. Logistic regression models were used for binary variables, and Bayesian polytomous regression was used for nominal variables. Proportional odds models were used for ordinal variables.  Analyses were performed using statistical software R (R Foundation for Statistical Computing, Vienna, Austria) with the tidyverse, mice, lubridate, readxl, writexl, robustbase, and quantreg packages [12].

Results
Four thousand forty seven consecutive patients from 71 centers in 14 countries were included in the dataset; these countries included Bahrain, Estonia, Finland, Iran, Ireland, Israel, Italy, Portugal, Romania, Spain, Sweden, Switzerland, the UK, and the USA [13]. The majority were SARS-CoV-2 uninfected (3861, 95.4%), while 70 (1.7%) were acutely SARS-CoV-2 positive, and 116 (2.8%) reported prior SARS-CoV-2 infection. Patients with an active SARS-CoV-2 infection were younger compared to patients with and without prior infection [median (IQR) 30  years vs 37  years in those with a history of prior infection and 35  years in those never infected, p = 0.019]. There were no statistically significant betweencohort differences in body mass index, ASA classification, comorbidities, smoking history, or admission laboratory data. However, the proportion of patients with a respiratory rate above 20 was significantly higher among patients with active and previous SARS-CoV-2 infections compared to patients who had never had SARS-CoV-2 (5.7% and 7.8% vs 2.9%, respectively, p = 0.004) ( Table 1). The prevalence of perforation and the time to surgery from admission were similar across groups. Patients with active and prior SARS-CoV-2 infections were significantly more likely to undergo an open surgical procedure (24.3% and 26.7% vs 9.6%, respectively, p < 0.001). Relatedly, procedure duration and the crude rate of any or severe complications were also similar across groups. Patients with active SARS-CoV-2 infection demonstrated a longer crude LOS compared to patients without prior SARS-CoV-2 infection (2.7 days vs 1.9 days, p = 0.002) ( Table 1). No associations between SARS-CoV-2 seropositivity and LOS, any complication, or severe complication were identified after adjusting for confounders in the regression analyses (Tables 2 and 3).

Discussion
The treatment of patients with acute surgical emergencies in the context of the COVID-19 pandemic has been challenging throughout different phases of the pandemicfrom early knowledge gaps and resource-exhaustion, through concern for excess postoperative morbidity and potential aerosolization of viral particles through laparoscopy, to cancellation of elective surgery and deviation from usual practice patterns, the introduction of population vaccination, and most recently, phased reintroduction of scheduled surgical services [14][15][16].
The 'evidence-to-practice gap' between guidelinebased recommendations and widespread adoption and implementation is well chronicled in the surgical literature and a burgeoning focus of implementation science [17]. Under usual conditions, there are several known barriers to recommendation implementation, including care inertia, lack of knowledge of new recommendations, resource limitation, as well as disagreement with guideline recommendations [17,18]. In stark comparison, the recent SARS-CoV-2 global pandemic engendered rapid creation, adoption and implementation of recommendations ahead of mature large-data evidence [19]. Concerns regarding staff safety as well as untoward patient outcomes following operative procedures performed during active SARS-CoV-2 infection shuttered elective and semi-elective procedures [20][21][22]. Multiple medical professional organizations, including the European Society for Trauma and Emergency Surgery (ESTES), rapidly crafted guidelines and statement. These supported delaying operative therapy, and prioritizing nonoperative approaches, during acute SARS-CoV-2 infection as well as recommending the avoidance of laparoscopic interventions due to the potential risk of the uncontrolled release of pressurized gas, which could result in the infection of surgical staff [23][24][25][26]. Accordingly, urgent procedures, including appendectomy for acute appendicitis, were diverted along a non-operative pathway. As a result, the recommendation for non-operative appendicitis management was readily embraced by surgeons who would previously have pursued routine operative management [27]. Only emergency operations such as those for injury, or life-saving organ transplantation, were undertaken during the early phase of the pandemic. There was a clear need for data to inform practice.
Multinational collaborations rapidly arose to assess outcomes of different therapeutics for those with acute SARS-CoV-2 infection. Some early therapies, such as glucocorticoid administration, remdesivir, monoclonal antibody rescue, and prone position therapy were found to be beneficial when large datasets were interrogated [28,29]. Other practices driven by early observations, such as routine therapeutic anticoagulation or early invasive mechanical ventilation, have been intensively investigated, determined to be lacking an evidence base, and abandoned as part of routine care [30][31][32]. Our data aligned with the latter studies in that we identified no untoward consequences for those with active SARS-CoV-2 infection who underwent appendectomy with active infection compared to those without infection, as well as those who recovered from a prior infection [33]. Since those who undergo appendectomy regularly demonstrate excellent outcomes, any deleterious impact of active SARS-CoV-2 infection is anticipated to be readily recognizable, even in a small cohort. It was also apparent that the recommendations regarding operative technique had an effect, given that the proportion of patients who underwent an open surgery, among those that had an active or prior SARS-CoV-2 infection, was nearly triple the proportion observed in uninfected patients.
The outcomes from this study span 90 days, a sufficient time frame to capture delayed events including hospital readmission for operative domain or pulmonary system failures; these adverse events were not observed. Importantly, our data are different from observations made early in the pandemic. One reason for such differences may be viral evolution, a process that has been repeatedly observed with different variants with some demonstrating enhanced infectivity but less virulence, and vice versa [34,35]. Other explanations include changes in the vulnerable patient population and enhanced acute care paradigms-a key aspect as our data were collected during the European second and sometimes third wave (delta variant dominant periods). It is also possible that patient selection informed which patients were deemed suitable for operation, potentially selecting a less physiologically encumbered group. While that is possible, we did not observe differences between groups with regard to ASA score or overall comorbidity burden. It is also possible that intra-operative pulmonary management shifted to routinely utilizing PEEP in the wake of additional understanding of acute SARS-CoV-2 infection. Such an approach may have reduced atelectasis and supported pulmonary flow particularly during abdominal insufflation and may have contributed to the observed lung-related outcomes. This is a supposition that is plausible, and one that is not investigable from the current database-a shortcoming of using a database with pre-specified fields as is typical for snapshot audits and other prospective analyses [9].
Our prospective time-bound multi-center observational cohort study allowed for a comprehensive defined dataset to be gathered in line with pre-publication, open-access protocols filed with clinical trial repositories. Based on the pre-specified data fields, we did not capture all comorbidities, nor how patients were selected for nonoperative versus operative management. Moreover, our data is limited to those with acute appendicitis and may not be applicable to those who required longer or more complex operative procedures, nor those who may require postoperative invasive mechanical ventilation. Similarly, we did not capture SARS-CoV-2 related therapy such as steroids, remdesivir, or the use of non-invasive ventilation or prone position therapy. Relatedly, we did not capture whether patients were ill from their SARS-CoV-2 infection, or asymptomatically infected as existing recommendations addressed such infection in a binary fashion. We were instead interested in assessing outcome across multiple centers for those who underwent operation regardless of how their acute SARS-CoV-2 infection was managed. Indeed, current management leverages some different than what was used during our study period. Additionally, our results present the median effects observed in the study population. We can therefore not eliminate the risk that a particular subgroup, for example frail or elderly patients, might exhibit an increased risk when undergoing surgical management for acute appendicitis with a concomitant SARS-CoV-2 infection. Nevertheless, owing to the age distribution of appendicitis patients, this cohort is relatively small with only 3.5% of patients in this sample being older than 75 and none of them having an active SARS-CoV-2 infection on admission [5]. Furthermore, the usual caveats applicable to observational studies also apply, such as the risk of residual confounding, selection bias, and limitations on causal inference.

Conclusions
The current study failed to detect any association between SARS-CoV-2 infection status and post-appendectomy complications or hospital length of stay. This provides evidence that the most common management approach for acute appendicitis-appendectomy-may be safely performed in patients who present with acute or recently recovered SARS-CoV-2 infection.
Author contributions All collaborators will be listed on PubMed as authors; see end of manuscript for list of Manuscript Writing Group, SnapAppy Steering Committee and Study Collaborators, and their affiliations.
Funding Open access funding provided by Örebro University. No funding was received for the execution of the current study.
Data availability All data and codes are available for retrieval on reasonable request.

Conflict of interest The authors have no conflict of interest to report.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.