Eosinophilic colitis (EC) is a rare inflammatory condition involving eosinophilic infiltration of the mucosa in the absence of a secondary cause or peripheral eosinophilia.1 Clinically, EC presents with nonspecific symptoms and is often categorized by chronic watery diarrhea, abdominal pain, nausea, and vomiting.1 We present a case of a young patient with malignant EC that caused ischemic colitis.
A 30-year-old man with a medical history of EC on azathioprine, pulmonary embolism, superior mesenteric vein thrombus on rivaroxaban, and hereditary alpha tryptasemia had multiple admissions at an outside facility for abdominal pain and hematochezia in August 2021. The patient's symptoms date back to 2012, around the time he took a trip to Qatar. His symptoms included nausea, vomiting, poor oral tolerance, abdominal pain, and diarrhea. He previously had an extensive evaluation for these symptoms, including negative evaluation for Strongyloides, stool ova, cysts, and parasites. He underwent esophagogastroduodenoscopy and colonoscopy with biopsies in August 2019, which were both endoscopically normal. Biopsies from the esophagus showed rare eosinophils; gastric biopsies showed a marked increase in eosinophils (up to 150 eosinophils/high-power field), but no Helicobacter pylori and were negative for dysplasia; and small bowel biopsies showed up to 20 eosinophils/high-power field with intact villi, excluding celiac disease. The terminal ileum biopsies showed up to 20 eosinophils/high-power field, and random colon biopsies from every segment of the colon showed marked eosinophilia ranging between 30 and 65 eosinophils/high-power field, without typical histological findings for inflammatory bowel disease. Evaluation for other causes of eosinophilia was similarly unremarkable despite peripheral eosinophilia being as high as 26.7% (0%–5%). He had no history of atopic conditions and normal immunoglobulin E, and flow cytometry of peripheral blood showed no abnormal population of lymphoid cells.
The patient also had a long and complicated history with thrombosis. He had recurrent thrombosis in the past to include portal, splenic, and superior mesenteric vein thrombosis. He was seen by multiple specialists to include hematology, rheumatology, allergy and immunology, as well as gastroenterology. Workup for inherited thrombophilia including anticardiolipin, beta 2 microglobin, factor V Leiden, prothrombin mutation, paroxysmal nocturnal hemoglobinuria, and homocysteine was all negative. He had prior negative workup for mast cell activation syndrome and systemic mastocytosis. He was evaluated by rheumatology, which reported that his presentation was not typical for vasculitis because it usually does not lead to thrombosis. He had no prior oral or genital ulcers, renal involvement, or pulmonary symptoms to suggest Behcet disease.
He was seen by gastroenterology and was tried on different treatments in the past to include mesalamine, azathioprine, cromolyn, and steroids. He was on cromolyn 4 times daily and pantoprazole twice daily at the time of initial presentation.
During the initial admission, the patient was found to have Clostridioides difficile and was discharged with a course of fidaxomicin. After discharge, the presenting symptoms continued to worsen, and the patient was readmitted because of refractory symptoms with a negative C. difficile polymerase chain reaction. Flexible sigmoidoscopy demonstrated a diffuse area of severely friable, erythematous, and edematous mucosa with contact bleeding 25 cm proximal to the anus along with multiple nonbleeding aphthous ulcers. The colonic biopsies were consistent with active colitis without evidence of chronicity and showed 70 eosinophils per high powered field. Surgery was consulted, and they opted to monitor with serial abdominal examinations. Abdominal computed tomography showed a near-occlusive mesenteric and portal venous thrombus with perfusion and congestive changes in the colon to include ischemia in the sigmoid colon. The patient was started on a heparin drip, antibiotics, and intravenous steroids. Interventional radiology performed a percutaneous transhepatic superior mesenteric and portal vein thrombectomy that was unsuccessful and complicated by a large right hemothorax requiring chest tube placement. The patient was then transferred to our hospital for higher level care because of hemorrhagic shock.
On arrival, the patient continued to have severe abdominal pain, shortness of breath, and hematochezia. His laboratory test results were significant for aspartate aminotransferase 1,165 U/L, alanine aminotransferase 1,165 U/L, creatinine as high as 1.6 mg/dL (baseline 0.7 mg/dL), and venous lactate on blood gas measurement 1.86 mmo/L (normal range 0.90–1.70). After resuscitation, a flexible sigmoidoscopy showed friability, contact bleeding, surface ulceration, and sloughing of mucosa (Figure 1a). The colonic biopsies were consistent with active colitis without evidence of chronicity and showed 60 eosinophils per high powered field. Abdominal-pelvic computed tomography showed no evidence of bleeding, but noted a thrombus at the portosplenic confluence extending into the main portal vein with associated heterogeneous appearance of the liver (Figure 1b). Surgery was consulted, and the patient was taken for a subtotal abdominal colectomy with end ileostomy. The pathology of the colonic tissue removed from the surgery showed extensive vascular thrombosis, ulceration, and ischemic changes without significant eosinophils because of active steroid therapy. No transmural disease was present that would be consistent with Crohn’s disease. His clinical status improved significantly after colectomy with normalization of hepatic and renal function. After discussion with allergy/immunology, he was initiated on the interleukin-5 (IL-5) inhibitor benralizumab for his EC and was discharged with Lovenox and a prednisone taper for 3 months. Benralizumab was picked because it was the one IL-5 drug that was approved by the patient's insurance.
Outpatient follow-up was notable for continued rectal bleeding with minimal improvement since discharge. Repeat flexible sigmoidoscopy showed diffuse severe erythema, contact bleeding, and friable mucosa throughout the rectal pouch (Figure 1c). Vedolizumab was started in combination with benralizumab. However, on long-term outpatient follow-up, no notable clinical and/or endoscopic improvement was noted with combination therapy. A multidisciplinary decision in conjunction with the patient resulted in planned proctocolectomy with permanent end ileostomy.
EC lacks well-established diagnostic criteria, making the exact prevalence of the disease difficult to determine.1 Diagnosis of EC must include the presence of gastrointestinal symptoms, infiltration of eosinophils into the bowel wall, and exclusion of other potential causes. Symptom severity is multifactorial, but largely determined by the depth of mucosal eosinophilic invasion. Endoscopy with subsequent histologic evaluation remains the preferred modality for diagnosis.1
Based on deeper biopsies, there was lack of chronicity and/or significant histological features suggestive of inflammatory bowel disease (IBD). Early IBD could potentially still be in the differential, but based on the data we gathered, it has a low likelihood for an overlap with IBD.
There are currently no Food and Drug Administration-approved management options for EC. The available data for biologic therapies are mostly found in the treatment of eosinophilic esophagitis (EoE). Monoclonal antibodies targeting IL-5 have been used in EoE and eosinophilic asthma. IL-5 assists in the development and release of eosinophils from the bone marrow and their activation within the tissues. The IL-5 inhibitor benralizumab is Food and Drug Administration-approved for the treatment of eosinophilic asthma and is currently being studied as a therapy for EoE and eosinophilic gastritis.2 Vedolizumab is a monoclonal antibody that functions to inhibit the trafficking of T lymphocytes and eosinophils into the intestinal mucosa.3 In a small number of case reports, it has shown some clinical improvement and reduction in eosinophilia in patients with eosinophilic gastrointestinal disorders.2 Owing to the rarity of EC diagnosis, data for therapies in EC are severely lacking, although the use of these targeted therapies may be reasonably considered for use in this condition.
To the best of our knowledge, this is the first case of a patient with severe refractory EC that led to the development of ischemic colitis. Despite combination therapy for salvage, the patient ultimately underwent a total proctocolectomy with permanent end ileostomy. More studies need to be performed to treat patients with severe refractory cases of EC.
Author contributions: C. Awad took care of the patient, performed literature review, wrote the original case presentation, and is the article guarantor. B. Bailes assisted in the literature review and wrote the case report. C. Horn took care of the patient, performed some literature review, and edited and reviewed the case report. A. Patel took care of the patient, reviewed, and provided feedback on the case report.
Financial disclosure: None to report.
Informed consent was obtained for this case report.
1. Gonsalves N. Eosinophilic gastrointestinal disorders. Clinic Rev Allerg Immunol. 2019;57(2):272–85.
2. Peterson K, Safroneeva E, Schoepfer A. Emerging therapies for eosinophilic gastrointestinal diseases. J Allergy Clin Immunol Pract. 2021;9(9):3276–81.
3. Kim HP, Reed CC, Herfarth HH, Dellon ES. Vedolizumab treatment may reduce steroid burden and improve histology in patients with eosinophilic gastroenteritis. Clin Gastroenterol Hepatol. 2018;16(12):1992–4.