key: cord-0939249-o8d9axsp
authors: Mustafa, Nada F; Jafri, Nadim S; Holtorf, Heidi L; Shah, Shinil K
title: Acute oesophageal necrosis in a patient with recent SARS-CoV-2
date: 2021-08-16
journal: BMJ Case Rep
DOI: 10.1136/bcr-2021-244164
sha: ab2798acdab8bf766239c1383e0ec1cc5bb8d407
doc_id: 939249
cord_uid: o8d9axsp

A 57-year-old Hispanic man with diabetes presented with dyspnoea. He had a positive SARS-CoV-2 PCR. He was intubated for severe hypoxia and treated with intermittent pressors, methylprednisolone and supportive care. He was extubated on hospital day (HD) 9 and discharged to a skilled nursing facility (SNF) on HD 18. Approximately 1 month later, he presented with melena. Endoscopy revealed two large 1.5–2 cm wide-based distal oesophageal ulcers without active bleeding. Histology showed ulcerated squamous mucosa with extensive necrosis extending to the muscularis propria and coccoid bacterial colonies with rare fungal forms suggestive of Candida. He was treated with fluconazole and pantoprazole and was discharged to a SNF. Approximately 3 weeks later, he was readmitted for complications. Repeat endoscopy demonstrated improvement and histology revealed chronic inflammation with reactive epithelial changes. Incidentally, SARS-CoV-2 PCR was positive during this visit without any respiratory symptoms.

A 57-year-old Hispanic man with diabetes presented with dyspnoea. He had a positive SARS-CoV-2 PCR. He was intubated for severe hypoxia and treated with intermittent pressors, methylprednisolone and supportive care. He was extubated on hospital day (HD) 9 and discharged to a skilled nursing facility (SNF) on HD 18. Approximately 1 month later, he presented with melena. Endoscopy revealed two large 1.5-2 cm wide-based distal oesophageal ulcers without active bleeding. Histology showed ulcerated squamous mucosa with extensive necrosis extending to the muscularis propria and coccoid bacterial colonies with rare fungal forms suggestive of Candida. He was treated with fluconazole and pantoprazole and was discharged to a SNF. Approximately 3 weeks later, he was readmitted for complications. Repeat endoscopy demonstrated improvement and histology revealed chronic inflammation with reactive epithelial changes. Incidentally, SARS-CoV-2 PCR was positive during this visit without any respiratory symptoms.

Acute oesophageal necrosis (AON) is a rare condition that classically appears as a circumferential black area in the oesophagus with the underlying aetiology being tissue hypoxia. SARS-CoV-2 disease (COVID-19) through dysregulation of inflammatory and coagulation mechanisms also causes tissue hypoxia. We report a case of AON presenting with gastrointestinal bleeding following a prolonged hospital stay for COVID-19. Many patients with COVID-19 receive anticoagulation due to their prothrombotic state, placing them at higher risk for gastrointestinal bleeding. AON should be considered as a potential cause of gastrointestinal bleeding in these patients.

A 57-year-old Hispanic man presented to the emergency room with dyspnoea for 1 week. He had no relevant medical history, although later during the course of his hospitalisation was found to have diabetes (Hemoglobin A1c 7.5%). He had no surgical or family history and denied medication use. He did not have any alcohol, tobacco or recreational drug use. He was intubated secondary to severe hypoxia. Initial laboratory data revealed positive SARS-CoV-2 PCR, hyponatraemia, anion gap respiratory acidosis, elevated creatinine indicating acute kidney injury, mildly elevated transaminases, elevated inflammatory markers and neutrophil predominant leucocytosis. Chest radiograph demonstrated bilateral patchy airspace opacities. The D-dimer was elevated but secondary to reduced glomerular filtration rate (GFR), CT of the chest with intravenous contrast could not be done to rule out pulmonary embolism. Bilateral lower extremity ultrasound studies were negative for deep vein thrombosis (Investigations table 1 ). The patient received fluid resuscitation, pressor support with norepinephrine, intravenous cefepime and vancomycin (dosed as per GFR), intravenous methylprednisolone (40 mg every 8 hours), prophylactic dose of subcutaneous enoxaparin (dosed per GFR) and famotidine for gastrointestinal prophylaxis. Initially, he required high FiO 2 to maintain oxygen saturations but gradually showed improvement. Pressors were weaned off on hospital day (HD) 4. His electrolyte abnormalities and creatinine gradually improved. He had adequate urine output. On HD 7, he received convalescent plasma therapy. His chest radiograph continued to show moderate diffuse ground-glass opacities and diffuse peribronchial thickening bilaterally. Respiratory and blood cultures remained negative. He completed a 10-day course of broad-spectrum antibiotics and was gradually weaned off of methylprednisolone (3 weeks taper) after being successfully extubated on HD 9. His COVID-19 PCR test was still positive on HD 14. He was requiring supplemental oxygen via nasal cannula, which was weaned down to 4 L by the day of discharge on HD 18. Due to significant deconditioning from his prolonged hospitalisation, he was discharged to a skilled nursing facility (SNF). All of his inflammatory markers were decreasing at the time of discharge. Creatinine, though improved, was still elevated at discharge, suggesting development of or underlying chronic kidney disease.

Approximately 1 month after discharge, the patient developed melena, for which he was readmitted to the hospital. He had normal vital signs and oxygen saturation was 100% on room air. Physical examination exhibited generalised deconditioning but was otherwise unremarkable. The differential diagnosis included peptic ulcer disease given the preceding critical illness. Gastritis and oesophagitis from prolonged steroid use were considered to represent other possible aetiologies. The probability of Mallory-Weiss syndrome was low as there was no reported history of persistent vomiting. Angiodysplasia and mass lesions were also considered.

The patient's haemoglobin on admission was 78 g/L and SARS-CoV-2 PCR was negative. Oesophagogastroduodenoscopy was performed, revealing Case report two large 1.5-2 cm wide-based ulcers in the distal oesophagus without active bleeding (figure 1A). His haemoglobin remained stable after endoscopy. Histology of the oesophageal ulcer (figure 1B) revealed ulcerated squamous mucosa with extensive necrosis extending to the muscularis propria. Coccoid bacterial colonies and rare fungal forms suggestive of Candida species were seen in the necrotic areas. Cytomegalovirus and herpes simplex virus immunoperoxidase stains were negative.

The patient was initially started on a pantoprazole intravenous drip and received 1 unit of packed red cells with an appropriate response in haemoglobin. After endoscopy, the pantoprazole intravenous drip was converted to pantoprazole 40 mg oral two times per day and fluconazole 200 mg oral daily for 14 days was added. He was tolerating a diet without any nausea, vomiting or abdominal pain. He developed urinary retention that was managed by Foley catheterisation and he was eventually discharged back to the SNF on HD 9.

Approximately 3 weeks later, the patient was readmitted to the hospital due to a urinary tract infection, Clostridium difficile diarrhoea and bilateral lower extremity deep vein thrombosis.

He was significantly deconditioned and weak with severe protein-calorie malnutrition as demonstrated by a drop in his body mass index from 32 kg/m 2 at the index hospitalisation to 20 kg/m 2 on this admission. His oral intake and nutritional status remained poor during his treatment in the hospital. His albumin was 1.2 g/dL. A decision was made to proceed with placement of a percutaneous endoscopic gastrostomy tube for enteral nutrition. He tolerated this well. Repeat endoscopy at this time showed improvement in the oesophageal ulceration ( figure 1C ). Histology demonstrated squamous mucosa with chronic inflammation and reactive epithelial changes. Incidentally, SARS-CoV-2 PCR was positive during this visit without any respiratory symptoms. After clinical improvement, the patient was discharged home.

The WHO declared SARS-CoV-2 disease (COVID-19) a pandemic on 11 March 2020. The majority of cases are reported to be mild but 19% of patients present with severe symptoms. The overall estimated mortality is 2.3%. Similar to SARS-CoV and Middle East respiratory syndrome coronavirus (MERS-CoV), SARS-CoV-2 is a beta-coronavirus. Bats are thought to be the most likely natural hosts. The virus, which then undergoes mutation or recombination in animal hosts, can infect humans. Pangolins have been implicated as the intermediate animal hosts of SARS-CoV-2. 1 Although the most common presentation of COVID-19 is respiratory, gastrointestinal problems including diarrhoea and elevated transaminases have been frequently noted. 2 This can be easily explained by the wide distribution of the ACE2 receptor through which SARS-CoV-2 gains entry into target cells. ACE2 receptors are present in the oral mucosa, oesophagus, small intestine, colon, liver and spleen in addition to the respiratory system. Common gastrointestinal symptoms include diarrhoea, nausea, vomiting and lack of appetite. 3 Cases of ischaemic colitis, 4 bowel perforation, 5 6 paralytic ileus, 7 gastrointestinal bleeding 8 9 and Case report pneumatosis intestinalis 10 in patients with SARS-CoV-2 have been reported. Dysregulation of the renin angiotensin system, cytokine storm, oxidative stress and coagulation abnormalities with resulting tissue hypoxia seem to be the underlying mechanisms of the disease. 11 Additionally in the intestine, SARS-CoV-2 has been postulated to decrease tryptophan absorption via ACE2, causing alteration of the intestinal flora, a decrease in antimicrobial peptides and resultant mucosal inflammation. 12 Whether intestinal disease in SARS-CoV-2 is due to the primary infection or a secondary response from systemic inflammation is unclear. 3 Historically, AON has been described as a rare condition diagnosed classically by a characteristic endoscopic appearance of a circumferential black area. The most common primary cause seems to be tissue hypoxia through hypoperfusion from cardiac dysfunction or sepsis. The less vascular lower oesophagus is predictably the most commonly affected site. 13 Other direct causes include severe gastric reflux, viral or fungal infections, allergic reactions to antibiotics, hypothermia or caustic injuries. 14 15 Hyperglycaemic or alcohol intoxication (causing transient gastropathy and malnutrition) have also been implicated as possible mechanisms. 16 17 Seen most commonly in older men with cardiopulmonary comorbidities, the typical presentation is with upper gastrointestinal bleeding. 13 15 Some of the most common risk factors include diabetes mellitus, cardiovascular disease and chronic renal failure. 13 18 Endoscopy reveals diffuse circumferential black discolouration of the oesophagus with sharp demarcation at the Z-line. Histology generally demonstrates necrotic debris, mucosal and submucosal necrosis with local inflammation. 15 Initial conservative treatment, including hydration, parenteral nutrition, antibiotics, proton pump inhibitors and sucralfate, is typically recommended. In cases with perforation, early surgical or endoscopic interventions have been applied. Common complications include strictures, stenosis, abscesses, tracheoesophageal fistula and perforation. The mortality rate has been reported to be up to 30%, which may also be from underlying comorbidities and risk factors. 15 Mortality directly from the oesophageal disease is reported to be around 6%. 17 Interestingly, the gastrointestinal bleeding caused by AON has not been observed to be significant enough to cause haemodynamic instability. 13 The cause of the AON in our patient is likely the period of critical illness during his initial hospitalisation as well as the undiagnosed diabetes. It is unclear if there was a direct relationship to SARS-CoV-2, as we could not test the specimen for SARS-CoV-2 RNA. There has been a report of SARS-CoV-2 causing herpetic erosions and ulcers with lymphocytic infiltration in the oesophagus where viral RNA was detected. Similar to our patient, this patient was also an elderly man who presented with ventilatordependent respiratory insufficiency with resultant gastrointestinal bleeding. 19 The classic endoscopic appearance of a black circumferential area was not present in our case, although there was extensive necrosis extending into the muscularis propria on histology. This is possibly because of the timing of the endoscopy, which occurred approximately a month after discharge from the initial period of illness. A more classic appearance may plausibly have been seen, had he had the endoscopy during the period of initial critical illness.

Given that 19% of patients with SARS-CoV-2 present with severe illness 1 and consequently have some degree of tissue hypoperfusion, 11 AON in such patients may be more frequently seen, specifically as survival rates are expected to increase with newer therapeutic options. With the associated prothrombotic state and frequent anticoagulation, these patients are at an increased risk of gastrointestinal bleeding. 20 It is important for physicians to be aware of AON as an aetiology of gastrointestinal bleeding in these patients. A staging system for AON based on clinical and endoscopic findings 15 may assist in risk stratification and development of evidence-based management guidelines.

► Acute oesophageal necrosis (AON) is a rare condition diagnosed classically by a characteristic endoscopic appearance of a circumferential black area. The most common primary cause seems to be tissue hypoxia through hypoperfusion from cardiac dysfunction or sepsis. ► It is estimated that 19% of patients with SARS-CoV-2 are severely ill and may have resulting tissue hypoperfusion which increases their risk of developing AON, especially since survival rates are expected to increase with newer therapeutic options. ► With the associated prothrombotic state and frequent anticoagulation, these patients are at increased risk of gastrointestinal bleeding. It is important for physicians to be aware of AON as an aetiology of gastrointestinal bleeding in these patients. ► A staging system for AON based on clinical and endoscopic findings may assist in risk stratification and development of evidence-based management guidelines.

Virology, epidemiology, pathogenesis, and control of COVID-19

AGA Institute rapid review of the gastrointestinal and liver manifestations of COVID-19, meta-analysis of international data, and recommendations for the consultative management of patients with COVID-19

COVID-19 and the digestive system

Coronavirus disease 2019 (COVID-19) and ischemic colitis: an under-recognized complication

Bowel perforation in a Covid-19 patient: case report

Gastrointestinal perforation in a critically ill patient with COVID-19 pneumonia

Self-Limited gastrointestinal bleeding in COVID-19

Uncommon presentation of COVID-19: gastrointestinal bleeding

Pneumatosis intestinalis in COVID-19

Vascular Manifestations of COVID-19 -Thromboembolism and Microvascular Dysfunction

Alterations in gut microbiota of patients with COVID-19 during time of hospitalization

Acute necrotizing esophagitis: a large retrospective case series

Black esophagus: report of six cases and review of the literature

Acute esophageal necrosis: a systematic review and pooled analysis

Diagnosis and management of acute esophageal necrosis

Acute esophageal necrosis: a view in the dark

Black esophagus: new insights and multicenter international experience in 2014

Upper gastrointestinal bleeding caused by SARS-CoV-2 infection

Ppis and beyond: a framework for managing Anticoagulation-Related gastrointestinal bleeding in the era of COVID-19

Acknowledgements The authors acknowledge Dr M. Elaine Kling's (Pathologist, Memorial Hermann Sugar Land Hospital) contribution towards helping collect data, writing the description of the slides and proofreading the manuscript.

Contributors NFM: Conception/design, drafting/revision; acquisition/analysis/ interpretation of data; final approval; article guarantor. NSJ: Acquisition/analysis/ interpretation of data; revision; final approval. HLH: Acquisition/analysis/ interpretation of data; revision; final approval. SKS: Design; drafting/revision; acquisition/analysis/interpretation of data; final approval.

The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.Competing interests None declared.

Provenance and peer review Not commissioned; externally peer reviewed.This article is made freely available for use in accordance with BMJ's website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.

Nada F Mustafa http:// orcid. org/ 0000-0001-6515-4949