Duodenal gunshot injuries are a permanent challenge, with high morbidity and mortality, and complications following initial repair are about 64%¹.

SARS-CoV-2 infection can be considered a decisive factor as severe cases can cause serious instability in patients and limit the therapeutic options.

Postoperative pulmonary complications occur in 51% of patients with SARS-CoV-2 infection and are associated with high mortality (38%)². Non-surgical management must be considered instead of major surgeries with high mortality rate³ as many scientific societies have recommended^4-6.

A 54-year-old-male patient with a penetrating abdominal trauma in the epigastric region due to gunshot wound received advanced trauma life support and initial surgery in other institution. The abdomen was approached via midline laparotomy. An AAST (American Association for the Surgery of Trauma) grade 3 liver injury on segment V and a grade 3 injury of the second part of the duodenum were observed.

The liver was sutured, the pylorus was excluded and a gastroentorostomy was performed. The patient was transferred to our institution 10 hours after surgery.

On admission, the patient was awake (Glasgow score 15/15) and complaint of abdominal pain without rebound tenderness; the appearance of the surgical wound was adequate. Hematic fluid (500 mL) was draining from three abdominal drains.

A computed tomography (CT) scan was performed (Fig. 1a) which showed a projectile on the posterior aspect of the cardia in close contact with the abdominal aorta. The CT-scan objectified the path of the bullet, liver laceration, injury of the duodenum and both psoas muscles and fracture of the vertebral body L2. There was blood content in the renal pelvis of the left kidney, a perihepatic fluid collection and pneumoperitoneum.

Figure 1 A: Axial section of contrast-enhanced CT scan of the abdomen on admission showing a projectile on the posterior aspect of the cardia in close contact with the anterior border of the aorta. B: Axial section of CT scan of the thorax five days later showing bilateral patchy consolidation with air bronchogram suggestive of inflammation/infection of the lung parenchyma. C: Contrast-enhanced CT scan of the abdomen showing open abdomen with temporary closure, three drain catheters in the right parietocolic gutter and two in the duodenal fistula. 

The patient was admitted to the intensive care unit and a watchful waiting approach was decided. Five days later, he presented leukocytosis, atrial fibrillation and bile stained fluid was noted in the drains. A CT scan of the thorax and abdomen showed bilateral pulmonary infiltrates (Fig. 1b) and fluid collections in the subphrenic space and hypogastrium. Reverse transcription polymerase chain reaction (RT-PCR) was positive for SARS-CoV-2. The patient evolved with respiratory failure requiring mechanical ventilation and presented persistent fever > 38 °C, hemodynamic instability and multiple organ failure requiring vasoactive drugs and dialysis. He was treated with oseltamivir, hydroxychloroquine, lopanivir and ritonavir.

In view of the hemodynamic instability and SARS-CoV-2 infection, we considered re-exploration of the abdomen vs. minimally invasive surgery of the abdominal collections. Two 16 Fr multipurpose percutaneous drainage catheters were initially inserted in the left subphrenic space and hypogastrium under CT guidance, evacuating 100 mL of purulent fluid from each one. A CT scan performed 48 h later showed new abdominal collections which required the insertion of four 16 Fr multipurpose drainage catheters in the left iliac region, left lumbar region, left hypochondrium and right lumbar region, obtaining purulent fluid (Fig. 1c). The patient evolved favorably, the multiple organ failure reverted, and a definite treatment was feasible.

The drain output was progressively decreasing except for the one in the duodenal region and anastomosis which persisted with a biliary fluid output of 200 mL/day despite the administration of octreotride. The patient required percutaneous tracheostomy due to prolonged intubation and evacuation of a right pleural effusion.

On day 19 he underwent surgical exploration due to persistent fever, leuokocytosis and bile stained fluid in the drains.

Relaparotomy was performed. An anastomotic leak of the gastroenterostomy and a biliary leak in the second part of the duodenum (the site of the initial perforation), which was blocked by epiploic adhesions, were observed. Antrectomy with a new gastroenterostomy with mechanical stapler was performed and an external duodenal drainage was left in place. The abdomen was left open with temporary closure using Vicryl mesh, and vaccum-assisted closure (V.A.C) for control of infections and to prevent abdominal compartment syndrome. (Fig. 2a). One month after hospitalization, a second PCR test was negative, and isolation ceased.

Figure 2 A: Week one after placement of the V.A.C system. B: Abdominal wall status on discharge. Arrow: enteroatmospheric fistula. 

The V.A.C. dressing was changed once every 48 hours. Several gems were isolated: ESBL producing E. coli, Enterococcus faecalis, Candida Krusei and Achromobacter, and many antibiotics were administered, ending with tigecycline and anidulafungin.

The patient developed an enteroatmospheric fistula in the center of the abdomen with high output (500 mL/day) that was diagnosed after changing the V.A.C dressing. The duodenal drainage behaved like a duodenal fistula with moderate output and no intermediate cavity, and progressively stopped draining after four weeks. The patient was transferred to a tertiary care center two months and 20 days after admission, breathing room air, with critical illness myopathy. He had a V.A.C system in the abdominal wall and an enterocutaneous fistula draining 200 mL/day (until the presentation of this manuscript) (Fig. 2b). As the fracture of the vertebral body L2 did not consolidate 65 days after the injury, the patient underwent posterior lumbar interbody fusion from T12 to L4. Therapeutic plan: Definite closure of the abdomen and treatment of the enterocutaneous fistula in 6 months of greater.

Severe SARS-CoV-2 infection in the context of a surgical complication after a shotgun injury in the abdomen was one of the determining factors for decision-making in treatment options.

Although abdominal re-exploration is mandatory in the presence of intestinal fluid and other sources of contamination in the abdominal cavity, once failure of the initial surgery or unnoticed injuries had been ruled out this decision should have been postponed due to the hemodynamic instability of the patient secondary to the association of the abdominal condition with severe SARS-CoV-2 infection at the beginning of the pandemic in our country.

Percutaneous treatment was the strategy used until the hemodynamic condition improved. At that moment, gastrointestinal continuity was restored and an external drainage tube was placed.

We believe that an early aggressive intervention at the time of peak systemic impact of COVID-19 would have further increased morbidity and mortality and obscure the chance of recovery.

In COVID-19 patients, failure of gastrointestinal anastomosis has been reported and is corroborated in this presentation. The severity of the respiratory compromise limits the use of diagnostic testing, a factor to consider in the care of such complex patients.