Impact of Covid-19 Disease and National Lockdown on Outcomes of Emergency Surgery. Results From A Retrospective Comparative Cohort Study In A Tertiary Referral Teaching Hospital

BACKGROUND: The aim of this retrospective comparative study was to assess the impact of the COVID-19 related disease on emergency surgery, comparing clinical main outcomes in the period March –May 2019 (Group 1) with the same period in Covid-19 Italian lockdown (March-May 2020, Group 2). METHODS: A comparison (Group 1 versus 2) was performed between the demographic, anamnestic, surgical, clinical and management features. RESULTS: 246 patients were included, 137 in Group 1 and 109 in Group 2 (p=0.03). No signicant differences in peri-operative characteristics were registered. A declared delay in access to hospital and SARS-CoV-2 preoperatively infection rates were 15.5% and 5.8% respectively in Group 2. The overall morbidity (OR=2.22, 95%CI=1.08-4.55, p=0.03) and 30-day mortality (OR=1.34, 95%CI=0.33-5.50, =0.68) increased signicantly in Group 2. Delayed access cohort demonstrated a close correlation with increased morbidity (OR=3.19, 95%CI=0.89-11.44, p=0.07), blood transfusion (OR=5.13, 95%CI=1.05-25.15, p=0.04) and a 30-day mortality risk (OR=8.00, 95%CI=1.01-63.23, p=0.05). Positive SARS-CoV-2 patients had higher risk of blood transfusion (20% vs 7.8%, p= 0.37), ICU admissions (20% vs 2.6%, p= 0.17) and median LOS (9 days vs 4 days, p= 0.11). CONCLUSIONS: This article provides enhanced understanding of the effects of the COVID-19 pandemic on patients access to emergency surgical care. Our ndings suggest that COVID-19 changed the quality of surgical care with poorer prognosis and higher morbidity rates. A delayed Emergency Department access and a “lter effect” induced by a COVID-fear in the population resulted in only the most severe cases reaching the


Introduction
The SARS-CoV-2 virus infections and the related clinical signs of disease started in Wuhan, Hubei, China in the last days of 2019 and soon spread rapidly all around the world (1). On March 11, 2020, the World Health Organization (WHO) declared the infection to be a pandemic disease (2). COVID-19 became a public health crisis that has profoundly modi ed medical and surgical patient management. The Italian Government imposed a lockdown with signi cant restrictions on human contact, travel and business operations from March 10, 2020 onwards. During the period March-May 2020, Italy was the third most affected country after USA and Spain with 204,576 con rmed disease cases and 26,049 deaths (3). The Italian Health Care System`s response to COVID-19 and the related challenges have caused important changes in hospital activities. The vast number of COVID-19 patients requiring hospitalization and critical care exceeded the capacity of Italian hospitals. The most signi cant consequence was the stop of elective non-oncological surgical procedures and the switch from non-COVID-19 pathways to COVID-19 pathways according to international guidelines for all patients admitted to hospitals (4). Emergency Department (ED) and urgent surgical protocols were modi ed in all health regions with a subsequent complete redistribution and reorganization of activities and services. Our hospital is a regional hub for emergency surgery. During the rst COVID-19 phase all peripheral hospitals were changed into Covid hospitals within short and a much higher number of emergency surgical procedures had to be performed in the regional hub with a potential unfavourable and unpredictable patients'outcome.
The aim of this retrospective comparative study was to assess the clinical impact of the COVID-19 related disease and restrictive healthcare measures applied in Italy on emergency surgery in our tertiary referral teaching hospital, by comparing clinical and management outcomes between the Covid-19 Italian lockdown period (March-May 2020) and the same period (March-May) of the previous year 2019.

Patients setting and study schedule
This study was planned as a retrospective cohort study comparing two separate groups of consecutive patients undergoing emergency surgical procedures in a tertiary referral teaching regional hospital before (March-May 2019, Group 1) and during the COVID-19 pandemic (March-May 2020, Group 2). Patients > 18 years of age admitted to the ED of the regional hub hospital for emergency operations in general surgery, gynecologic or urologic department were enrolled retrospectively. Exclusion criteria were age < 18 years, incomplete follow-up data, re-operations after elective surgery, outpatient clinical visits in the recent 7 days, patients on waiting list and day hospital admissions. Emergency surgery was de ned as any emergency or urgent procedures done by surgeons in an operative room under general anesthesia. Any emergency or urgent indication were eligible for inclusion which was done according to the STROBE statement (5).

Variables and outcome
Demographic, anamnestic, surgical, clinical and management characteristics were registered. The following scales, scores and parameters were registered at the time of admission to the ED: Numeric Rating Scale (NRS) for pain (6), National Early Warning Score (NEWS) (7), Charlson Comorbidity index score (CCI) (8), declared delayed access > 48 hours, patients` diagnosis, presence/absence of sepsis criteria at admission (9), American Society of Anesthesiologists (ASA) score, SARS-CoV-2 swab test (only for Group 2). Surgical approach was classi ed as open, laparoscopic and endoscopic procedures (operative cystoscopy, hysteroscopy and vaginal approach). The main outcomes were intra-operative complications, Intensive Care Unit (ICU) admission, Diagnosis Related Groups (DRG) score, length of hospital stay (LOS), discharge pathway, post-operative complications according to Clavien-Dindo morbidity classi cation (10), intra/post-operative blood transfusion rate, in-hospital mortality and 30-day mortality. In-hospital mortality was de ned as death during the same hospital stay and 30-day mortality was de ned as death within 30 days after surgery. DRG score is a system for remuneration of hospitals based on the classi cation of the Diagnosis Related Groups (11). It is a system for measuring and enhancing hospital admissions for acute illnesses, representative and speci c of the Italian reality. The system also provides tools for its continuous maintenance, with the aim of contributing to the improvement of hospital governance, allocation of resources, appropriateness and quality of care. Clinical data were collected by reviewing all electronic medical charts and, if needed, by reviewing medical ward charts and/or by calling the patients.
Data collection was performed by authors FAC, LM, AC, TC and statistical analysis was performed by MR.
The study was approved by the local institutional Ethical Committee and data acquisition and storage were compliant with guidelines of the institutional review board. The study was conducted in line with the Good Clinical Practice guidelines and the ethical principles according to the Declaration of Helsinki. All anamnestic, clinical or laboratory data containing sensitive information about patients were de-identi ed in order to ensure analysis of anonymous data only.

Statistical analyses
The descriptive variables were expressed by mean ± standard deviation (SD) in case of normal distribution or by median and rst and third quartiles (q1, q3, respectively) in case of non-normal distribution. The normality of the variables was tested with the Shapiro-Wilk test. The dichotomous variables or scores were expressed as frequencies and occurrence percentages. Characteristics of patient populations were compared with the most appropriate test according to the nature and normality of the data (Chi-squared test, Fisher exact test, Student t-test or U-Mann Whitney test when appropriate).
We performed univariate and multivariate logistic regression models to evaluate the real impact of COVID 19 on patients' outcomes. When appropriate, multivariate models were performed adjusting for ASA score, age, CCI at admission, and NEWS at admission. When logistic regression was not feasible, we reported numbers, percentages, p-value from Fisher exact test or median (q1-q3) and p-value from the U-Mann Whitney test.
Two subgroup analyses were performed for Group 2 (2020 patients). The rst subgroup analysis evaluated the outcomes according to the patients' declared delayed admission, the second one, aimed to assess the impact of SARS-Cov-2 positivity on patients' outcomes versus to SARS-Cov-2 negative patients.

Patients' characteristics, diagnoses and clinical management
No signi cant peri-operative characteristics differences were registered in the two groups in terms of age, gender, NRS, NEWS, CCI, ASA score, presence/absence of sepsis criteria, surgical approach, ICU admission. On the other hand, the transfer from outside to hospital showed an increasing in ambulance use (16.8% vs 51.4% in Group 1 and 2 respectively, p = 0.03) with a consequently reduction of personal vehicle (Group 1 81.0% vs Group2 46.8%, p = 0.03). Moreover, a declared delayed access to hospital was registered in 0% in Group 1 and in 16 subjects in Group 2 (15.5%) (p < 0.01). SARS-CoV-2 infection (tested only in Group 2) was diagnosed preoperatively in 5 cases of 86 (5.8%) patients with nasopharyngeal swab. A statistically signi cant difference in overall population size was discovered (Group 1, n = 137; Group 2, n = 109; p = 0.03) (Tables 1 and 2).  In Group 2 in respect to Group 1, the total number of procedures categorized by a single clinical unit showed differences with decreased percentages for General Surgery 1 and Urology, and with increased percentages for General Surgery 2 and Gynecology (p < 0.01). (Table 3).    Table 5). The adjusted analyses lowered the magnitude of the differences and canceled the statistical signi cance (Table 5). Positive SARS-CoV-2 patients had higher risk of blood transfusion (20% vs 7.8%, p = 0.37), ICU admissions (20% vs 2.6%, p = 0.17) and median LOS (9 days vs 4 days, p = 0.11), although differences did not reach statistical signi cance. Tables 6 and 7 show the impact of SARS-CoV-2 positivity on two sub-Group 2 (SARS-Cov-2 negative versus SARS-Cov-2 positive).  Legend: CI = con dence interval; ICU = Intensive Care Unit; LOS = length of stay; OR = Odds Ratio; q1 = rst quartile; q3 = third quartile.

Discussion
Covid 19 is a global pandemic disease that have forced major changes in human healthcare systems all around the world. In Italy, the exponential increase in severe cases led to a rise in hospitalizations with a progressive overcrowding in ICUs. As a consequence, we witnessed a switch from permanent emergency room's respirator stations to beds equipped with mechanical ventilation. This reorganization resulted in a greater demand for dedicated ICU personnel and a rapid decrease in planned surgeries.
During the lockdown period, EDs experienced a large in ux of patients with respiratory failure due to COVID 19 related pneumonia and a rapid increase in cases requiring immediate treatment in parallel with the growing outbreak. A restructuration of all medical services was provided and in particular for patients in need of surgical care. Elective non-oncological surgery was temporarily allocated to a stand-by position mainly to relocate staff members, particularly anesthesiologists, to help with emergency cases and to switch the operative theatres into emergency rooms for the sickest COVID-19 patients. Non-operative management (NOM) of surgical patients had to be considered whenever possible. Only emergency cases and selected oncological procedures continued taking place.
In our region the public healthcare system closed ve peripheral EDs to concentrate emergency surgical activity in our hub hospital only. Dedicated COVID 19 protocols was established. All surgical patients we screened for Covid-19 before admission to EDs. To prevent contamination of holding areas no patients were moved between different hospital areas until their destination had been con rmed ready and a dedicated COVID-19 operating theatre was designated.
To our knowledge this is one of the rst studies analyzing the impact of COVID 19 disease and lockdown on emergency surgical procedures in a single hub hospital by comparing two cohorts of patients enrolled in the same time periods before and during the pandemic (2019 versus 2020). A unique feature of our study is the comparison of outcomes in the same clinical units having the same senior surgical staff before and during the pandemic. Our ndings demonstrate the consequences of reduced ED resources for ordinary non-COVID-19 patients in need of emergency surgery during the pandemic.
We observed a decrease in overall emergency surgical activities of 20.4% (p = 0.03) during the pandemic which is comparable to other international ndings (12,13). Nevertheless, bowel obstruction, acute appendicitis, extra-uterine pregnancy and pelvic in ammatory disease (PID) increased. Other authors have also found a change in clinical presentation of emergency cases during the COVID-19 pandemic with a reduction in less severe conditions like urinary tract pathology and an increase in some surgical conditions such as bowel obstruction, acute appendicitis, extra-uterine pregnancy and PID (14). These observations most likely have a multifactorial explanation. The Italian Government said that the most important action to save lives was for people to stay at home as much as possible. This restriction might explain the reduction of patients with hospital access due to "non-essential" surgical consults and trauma. Many people may have avoided visiting hospitals in fear of contracting COVID 19. General practitioners probably managed symptoms of abdominal pain, pelvic pain and urinary burning in patients` homes in a larger extent than before the pandemic. Lifestyle changes during lockdown may explain the lower incidence of some diseases like acute cholecystitis. The lower rate of cholecystectomy during the pandemic might be due to NOM using percutaneous cholecystostomy in more cases. Another aspect could be that surgeons adopted a more conservative behavior driven by fear of becoming patients themselves (15). Estimates show that 85% of healthcare workers get exposed to the virus and the International Council of Nurses estimated an infection rate of 9% in Italy during period March-April 2020 (16). Finally, as many elective procedures were postponed, fewer patients required emergency surgical revision due to complications. Interestingly, patients undergoing surgery were similar before and during the pandemic regarding gender, age, frailty and comorbidity, as assessed by NEWS and CCI (17). The Italian World Bank Staff estimates that Italy has the second largest proportion of older adults in the world (18). Therefore, Italian EDs are used to manage older subjects who are more susceptible to and more strongly affected by COVID-19 with at greater risk of developing emergency surgical conditions and related complications. In this study patients` median age was only slightly lower during the pandemic but we didn't observe any statistically signi cant differences regarding the age.
A declared delayed access to ED of 48-92 hours from the onset of clinical signs was observed in 15.5% of patients during the pandemic. Several factors could explain this nding such as changes in outpatient pathways (19) and patients' fear of going to hospitals and becoming infected with COVID-19 (20). Many surgical patients were initially treated without surgery. Acute appendicitis, acute cholecystitis, uncomplicated diverticulitis and urinary tract infections were given antibiotics only. Hence, many subjects were not referred to hospital until after NOM failure and in a worse clinical condition. Cano-Valderrama et al (21)   pandemic has to be analyzed based on data from the real-world scenario.
Moreover, for subgroup analyses the small sample size and small number of events did not allow us to draw solid conclusions. Some diagnoses and baseline characteristics were not homogeneous between groups, thus limiting the robustness of the results. However, a detailed description of the observed differences still provides data of clinical relevance. Indeed, the Covid-19 pandemic is a global healthcare emergency that generates a cascade of important health problems which deserves further analysis and comments.

Conclusions
Our tertiary referral teaching regional hospital experienced a decrease in surgical emergency admissions and operations during the COVID-19 pandemic and lockdown period in the spring of 2020. SARS-CoV-2 infection was diagnosed preoperatively in 5/86 of operated patients (5.8%). As compared with the same time period in 2019 there was an increase in the use of ambulance for pre-hospital transfer of patients. A declared delay in access to hospital was registered signi cantly more often during the pandemic (15.5%, p < 0.01). Our ndings show that COVID-19 affected surgical care with higher morbidity rates due to more low-grade complications. The overall morbidity and 30-day mortality increased signi cantly. Delayed access was closely related with increased morbidity, blood transfusion and the 30-day mortality risk. Positive SARS-CoV-2 patients had higher risk of blood transfusion, ICU admissions and median LOS.
We believe delayed ED access is due a " lter effect" induced by the COVID-fear in the population. As a result, only the most severe cases were referred to the emergency department. Further research is needed to corroborate our assumption that the COVID-19 pandemic changed the emergency surgery management and its results. Further analysis of pooled data could allow the international community to better identify patients that should be referred promptly to emergency departments and which patients that may bene t from non-operative management. Availability of data and material: The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Abbreviations
Con icts of interest/Competing interests : The authors declare that they have no competing interests. All authors certify that they have no a liations with or involvement in any organization or entity with any nancial interest or non-nancial interest in the subject matter or materials discussed in this manuscript.
Funding : The authors did not receive support from any organization for the submitted work.