EMERGENCY SURGICAL CARE FOR PATIENTS WITH COVID-19 AND TUBERCULOSIS COINFECTION AT MULTISPECIALTY HOSPITAL

The double burden of the novel coronavirus infection and tuberculosis (TB) is a global challenge. The aspects of emergency surgical care for patients with COVID-19 and TB coinfection remain understudied. The aim of this study was to assess treatment outcomes in acute surgical patients with COVID-19 and preexisting TB coinfection. In 2020, our Center delivered surgical care to 465 patients with COVID-19 and preexisting TB; a total of 64 emergency surgeries were performed on 36 (5.6%) patients, of whom 16 had HIV. Thirteen patients (36.1%) were diagnosed with pulmonary TB; 23 patients (63.9%) had disseminated TB. Chest CT scans showed >25% lung involvement in 61.9% of the patients with COVID-19 pneumonia, 25–50% lung involvement in 30.6% of the patients, and 50–75% lung involvement in 5.6% of the patients. By performing abdominal CT, we were able to detect abdominal TB complications, including perforated tuberculous ulcers of the intestine, intestinal obstruction and tuberculous peritonitis, as well as tuberculous spondylitis complicated by psoas abscess. Of all surgical interventions, 28.2% were abdominal, 23.2% were thoracic, 15.6% were surgeries for soft tissue infection, and 32.8% were other types of surgery. Postoperative mortality was 22.2%. We conclude that COVID-19 did not contribute significantly to postoperative mortality among acute surgical patients with TB.

The COVID-19 pandemic put tremendous strain on public health services, overwhelmed hospital capacities and dramatically increased the workload for healthcare workers. Elective surgeries were postponed due to the implementation of containment measures and hospital repurposing for patients with COVID-19. At the same time, hospitals for infectious diseases were admitting COVID-19 patients with acute surgical conditions, a urologic or gynecologic emergency or trauma [1][2][3][4]. During the pandemic, protecting healthcare workers against occupational exposure to the virus became a top priority. SARS-CoV-2 actively replicates in the respiratory and gastrointestinal tracts; this increases the risk of infection through aerosol transmission during gastrointestinal or respiratory tract surgery [5,6]. The diagnosis and treatment of the novel coronavirus infection are complicated by misleading symptoms and difficulty implementing international guidelines on operating room practices during the COVID-19 pandemic [7][8][9][10].
The aim of this study was to assess treatment outcomes among acute surgical patients with the novel coronavirus infection and preexisting tuberculosis. Twenty-three (63.9%) patients were referred to the Center by specialized TB hospitals of Moscow City. Thirteen patients (36.1%) were transferred by emergency medical services (EMS) from other medical facilities. Of them, 7 came from non-TB facilities due to suspicion of TB, 6 were taken by EMS either from a computed tomography facility or from their homes. Nineteen (52.8%) patients were Moscow City residents, 9 (25.0%) resided outside Moscow region, 5 (13.9%) were non-Russian residents, and 3 (8.3%) were homeless.

METHODS
The study included 31 (86.1%) male and 5 (13.9%) female patients aged 26 to 83 years (mean age was 41.9 ± 15.3 years). All of them had TB; 16 (44.4%) had HIV coinfection. On admission, all patients underwent a computed tomography (CT) scan of the chest; in addition, patients with acute abdominal pathology underwent an abdominal CT and/or abdominal ultrasound examination. The clinical diagnosis of COVID-19 was based on PCR tests for SARS-CoV-2 detection in nasopharyngeal swabs and chest CT findings. TB diagnosis was based on radiography findings, urine, sputum and stool tests, and biopsies. The collected specimens were studied for the pathomorphological features of TB and/or the presence of tuberculosis mycobacteria (MTB) using luminescence microscopy, liquid/solid culture-based methods and moleculargenetic analysis.
The significance of differences between quantitative variables was assessed using Student's t-test. For quantitative variables, the chi-square test was used.

RESULTS
The average duration of TB from its onset to the time of hospital admission was 6.7 months. Sputum tests came back negative for most patients (23 patients, or 63.9%). Thirteen patients (36.1%) tested positive for MTB; of them 4 (30.8%) were infected with MDR strains and 1 (7.7%) had XDR-TB.
In all cases, chest CT scans were suggestive of viral pneumonia. SARS-CoV-2 RNA was detected in the nasopharyngeal swabs of 28 (77.8%) patients; 8 (22.2%) tested negative for SARS-CoV-2, so the diagnosis of COVID-19 in these patients was based on the clinical and radiographic findings (Table 1).
Abdominal CT was performed on 11 (30.6%) patients. Two patients had free gas in the peritoneal cavity due to a perforated tuberculous ulcer of the ileum. In another patient, the accumulation of free gas was localized to the right ileal region and was due to the sealed perforation of the cecum. One patient had retroperitoneal abscess due to the retroperitoneal perforation of a tuberculous ulcer in the cecum. One patient had dilated bowel loops resulting from acute intestinal obstruction. Abdominal lymphadenopathy and encysted tuberculous peritonitis were observed in 2 patients. One patient had tense ascites and another one had tuberculous spondylitis complicated by psoas abscess. Two patients with strangulated hernia had no abdominal pathology.
A total of 64 emergency surgeries were performed on 36 patients; 12 patients required repeat surgery ( Table 2).
Abdominal surgery was the most common surgical intervention: a total of 18 (28.2%) abdominal surgeries were performed on 8 patients. Five patients underwent surgery for abdominal TB complications: 4 of them had a perforated tuberculous intestinal ulcer due to intestinal TB, one had acute intestinal obstruction due to a stenosing ulcer. It should be noted that abdominal surgical pathology is often characterized by a protracted clinical course and vague symptoms, and the patient starts seeking medical advice after the condition has progressed to advanced stages and requires a series of surgical interventions. Four patients with a perforated tuberculous ulcer of the intestine had to undergo repeat surgery. The patient with acute intestinal obstruction was treated with only one surgery.
Two patients with pulmonary TB and COVID-19 coinfection were admitted to the Center for suspected acute appendicitis. Diagnostic laparoscopy revealed that one of them had Thoracic surgery was performed on 10 patients with COVID-19 and preexisting TB; of them 5 had to undergo repeat surgery. All thoracic patients received pleural drains (6 patients required drains due to pneumothorax, the rest 4 had pleuritis or pleural empyema; see Figure).
Bronchial blocker placement during rigid bronchoscopy and mechanical ventilation was performed on 4 patients. Of them, 2 had fibrocavernous TB and developed pulmonary hemorrhage. Two other patients had a bronchopleural fistula, a collapsed lung and were unresponsive to pleural drainage. Tracheostomy was performed on 7 patients in intensive care requiring long-term mechanical ventilation.
There were 8 (22.2%) postoperative deaths. Of them, 4 patients with COVID-19/TB and respiratory failure were put on a ventilator and so underwent only one surgical intervention (tracheostomy). In their case, the immediate cause of death was COVID-19 and TB coinfection. Of those 4 patients, 2 had extensive pulmonary TB and COVID-19 coinfection (lung involvement score: 3); the other 2 patients had disseminated TB and COVID-19 coinfection (lung involvement score: 1 and 3, respectively).
Two patients died after multiple surgical interventions for abdominal TB complications, perforated tuberculous ulcers of the intestine accompanied by peritonitis. One patient, who underwent diagnostic laparoscopy for TB peritonitis, died of disseminated TB. Another death was due to decompensated liver cirrhosis and disseminated TB; this patient had tight ascites and had to undergo abdominal paracentesis. Four (50%) of the deceased patients had triple infection: COVID-19/HIV/TB; of them 2 also had hepatitis C.

DISCUSSION
There are a few important aspects to diagnosing and treating acute surgical patients with COVID-19 and TB coinfection. Before the COVID-19 pandemic, an abdominal ultrasound examination was the preferred diagnostic modality for patients with acute abdomen [15]. Now, considering the spread of the novel coronavirus infection, it is essential to assess the extent of lung damage caused not only by TB but also by rapidly progressing COVID-19 pneumonia [16]. We think that performing both chest and abdomen CT in patients with coinfection helps to clarify the underlying cause of abdominal pathology and detect an abdominal catastrophe if presenting symptoms are unobvious. Our opinion is consistent with the opinion of other researchers [17,18].
However, some authors report that abdominal CT examinations are being decreasingly performed on patients with acute abdomen because CT facilities are overwhelmed during the pandemic; such reports promote laparoscopy as a  [19,20]. However, patients mentioned in these publications do not have TB. It is well known that abdominal CT can clarify the need for laparoscopy as it is capable of detecting free gas in the peritoneum and fluid accumulation in bowel loops. In other words, abdominal CT can noninvasively detect complications of abdominal TB, including perforated tuberculous ulcers and acute intestinal obstruction. Therefore, we think it reasonable to perform an abdominal CT scan on patients with suspected acute surgical pathology and COVID-19/TB coinfection. In our study, only 2 patients, whose CT scans were inconclusive, were subjected to diagnostic laparoscopy. But the overarching surgical treatment strategy for our patients with abdominal TB complications was unchanged [21,22]. Pulmonary TB is a common cause of spontaneous pneumothorax. The rate of this complication varies between 0.6% and 1.4% [23]. In patients with TB, spontaneous pneumothorax develops due to ruptured pleural blebs or bullae compromising visceral pleura integrity. However, spontaneous pneumothorax is not always associated with TB in TB patients. For example, it can be linked to bullous emphysema which developed independently of TB. Recent publications report spontaneous pneumothorax in patients with COVID-19 who did not have TB and/or any other preexisting lung disease. Almost all of them were on mechanical ventilation, i.e. they might have sustained injury to the trachea during intubation. Their mortality rate was very high (72.7%) [24][25][26]. In our study, 6 of 7 patients with pneumothorax underwent pleural drainage. All of them had active pulmonary TB. One of them died: he had a past history of multiple surgeries for abdominal TB complications and pneumothorax.
The analysis of surgical interventions performed in April through December 2019 vs. the same period in 2020 reveals a rise in the total share of tracheostomies from 5.1% (12 of 234 emergency surgeries) in 2019 to 10.9% in 2020, which can be explained by the need for long-term mechanical ventilation in patients with COVID-19.
The analysis of postoperative mortality shows that abdominal TB complications (perforated tuberculous ulcer leading to peritonitis in patients with disseminated TB) was the primary cause of death in 50% of patients in 2019 and 2020. A multicenter cohort study reports an increase in postoperative mortality to 15.9% among COVID-19 patients requiring emergency surgical care, in comparison with patients without the novel coronavirus infection [27].
In our study, there was no reliable increase in postoperative mortality: 16.5% in 2019 vs. 22.2% in 2020 (χ 2 = 0.46; p = 0.49). Therefore, we conclude that COVID-19 did contribute significantly to postoperative mortality in acute surgical patients with TB. According to the literature, hospital mortality from COVID-19 varies from 6% to 25% [28]. Previously, we reported that overall hospital mortality among patients with COVID-19 and TB coinfection was 10.2%, whereas postoperative mortality in this cohort of patients was more than twice as high (22.2%) [14].
One of the limitations of this study is its small sample size due to the specific character of our patient cohort (triple COVID-19/HIV/TB infection). Another limitation is the short follow-up period, which did not allow us to analyze the longterm outcomes of the performed surgical interventions.

CONCLUSION
At our specialized TB hospital providing a wide range of health services to TB patients, emergency surgery was most commonly performed on patients with disseminated TB and COVID-19 coinfection (63.9%); patients with pulmonary TB required emergency surgical care less often (36.1%). Almost half of our The highest postoperative mortality (50%) was observed in the small group of 8 patients operated for abdominal conditions. Of all the patients who died in the postoperative period, 2 (50%) had COVID-19 and TB, and another 2 (50%) had quadruple infection with COVID-19, HIV, TB and hepatitis C.