First Description of Sprue-Like Enteropathy due to Azilsartan: A Case Report Open Access

Sartan-induced enteropathy is an uncommon side effect of treatment with angiotensin II receptor antagonists (ARBs) [1]. This adverse drug reaction has predominantly been described for olmesartan. For olmesartan, this condition was reported for the first time in 2012 [2] after being approved for antihypertensive therapy in 2002 [3]. The main manifestations include diarrhea (95%) and weight loss (89%). Other common symptoms are vomiting, fatigue, abdominal bloating, and abdominal pain. [4]. Histopathological results in affected patients show intraepithelial lymphocytosis and intestinal villous atrophy, similar to the findings in celiac disease. However, celiac disease-specific antibodies (anti-transglutaminase 2 and anti-endomysial IgA antibodies) remain negative, and a gluten-free diet does not show any response [2, 5].

As similar symptoms and histological findings have been described for other ARBs over the years [5‒7], the question arises as to whether sartan-induced sprue-like enteropathy can be considered a class effect of ARBs. Here, we present a case report of a 74-year-old male with sprue-like enteropathy associated with azilsartan treatment confirmed by positive de-challenge and re-challenge of the drug. As far as our intensive literature research reveals, this is the first report of sprue-like enteropathy associated with the ARB azilsartan, and this should raise awareness for this emerging differential diagnosis in seronegative villous atrophy or refractory celiac disease.

This report presents the case of a 74-year-old male who was admitted to the hospital twice within 5 months because of voluminous and frequent non-bloody diarrhea with weight loss. Upon first presentation to our emergency department, the chief complaints were non-bloody, watery diarrhea up to 20 times per day, and a weight loss of 5 kilograms within 10 days. In addition, the patient complained about mild nausea and occasional emesis. He denied any fever episodes, abdominal pain, or food dependence of the symptoms. Travel and environmental history were negative. Seven days prior to the hospital admission, his general practitioner had already started loperamide for acute gastroenteritis, as well as a 3-day course of antibiotic treatment with azithromycin because bacterial etiology was suspected. Medical history revealed arterial hypertension and valvular heart disease. The patient’s current medications included azilsartan (80 mg qd), bisoprolol (5 mg qd), and amlodipine (10 mg qd).

Upon admission, the patient appeared debilitated but afebrile and normotensive. Clinical examination of the abdomen was unremarkable. The laboratory workup, including tests for intestinal pathogens and stool tests, was normal except for an elevated serum CRP level (15 mg/L, normal value <8 mg/L) and a slightly increased stool calprotectin level (96 µg/g, normal value <50 µg/g; Tables 1, 2). Abdominal ultrasound showed increased peristalsis. CT showed fluid-filled intestinal loops without any signs of mechanic obstruction.

Given these findings, viral enteritis was suspected and supportive therapy was continued with fluid administration, probiotics (Streptococcaceae SF68), and loperamide (2 mg, up to 6 times daily). As the symptoms persisted, other differential diagnoses had to be considered, and a colonoscopy and upper endoscopy were performed. In the stomach (particularly in the antrum), a mild-to-moderate Helicobacter pylori-positive gastritis was found. A duodenal quick hypolactasia test revealed hypolactasia. Furthermore, duodenal and ileocecal biopsies revealed subtotal villous atrophy, intraepithelial lymphocytosis, and crypt hyperplasia accompanied by chronic inflammation (Fig. 1). Because these findings correlate with a typical histology of sprue-like enteropathy (Marsh type IIIb) similar to celiac disease, antibodies against transglutaminase IgA and endomysium IgA were determined. However, the serology for celiac disease was negative, whereas total IgA was normal (2.5 g/L, normal value 0.7–4.1 g/L). At this time point, the patient was not on a gluten-free diet.

H. pylori-associated gastritis was eradicated with amoxicillin (1,000 mg bid), metronidazole (500 mg tid), and pantozole (40 mg bid) for a total of 14 days. Furthermore, a gluten-free and lactose-free diet was implemented. Severe hypolactasia was interpreted as secondary due to the sprue-like enteropathy.

As symptoms still did not resolve, drug-induced enteropathy was taken into consideration, and azilsartan treatment was stopped. Without further treatment, symptoms started to improve 5 days after stopping azilsartan, and the patient was transferred to outpatient care after 17 days of hospitalization.

The patient was regularly seen by a hospital-based nutritionist while compliantly following a gluten-free and lactose-free diet. Upon discharge, antihypertensive treatment was continued with perindopril (5 mg qd), bisoprolol (5 mg qd), and amlodipine (5 mg qd).

Within the next 3 months, the patient developed a persistent cough, which was considered to be induced by perindopril. Therefore, perindopril was stopped by his general practitioner and replaced by re-initiating azilsartan (80 mg qd). Eleven days after re-exposure to azilsartan, the initial symptoms (excessive diarrhea and 12 kilograms weight loss within 3 weeks) restarted, leading to a hospital admission 3 weeks after re-exposure to azilsartan.

Physical examination of the abdomen only revealed discrete pain on palpation in the lower left quadrant, otherwise unremarkable. Laboratory workup included intensified pathogen diagnostics and stool examination, showing an elevated serum CRP level (10 mg/L) and increased fecal calprotectin level (3,990 µg/g), as well as a positive result for PCR of Clostridioides difficile. In addition, acute renal failure stage I (creatinine 108 µmol/L, normal value <115 µmol/L, estimated glomerular filtration rate (GFR) about 54 mL/min/1.73 m²) was found (details listed in Tables 1, 2).

Clostridioides difficile infection was treated initially with metronidazole (500 mg TID for 2 days), with no impact on the symptoms. Since a fever occurred, antibiotic therapy was switched to vancomycin (125 mg po QID). Under this treatment, the patient was afebrile and had a slight improvement in diarrhea frequency. Acute renal failure was interpreted as result of the watery diarrhea because it rapidly improved under intensive fluid therapy. However, excessive diarrhea persisted despite adequate antibiotic treatment of the Clostridioides difficile infection.

Another upper endoscopy was performed, which showed histopathological findings identical to the first episode 5 months earlier. Repeated serological tests were negative for transglutaminase IgA and endomysium IgA, whereas total IgA was in the normal range (3.8 g/L).

Upon repeated detailed questioning, the patient admitted to the re-initiation of antihypertensive treatment with azilsartan (80 mg qd) 11 days before the symptoms relapsed. A causative association between re-occurring symptoms and re-exposure to azilsartan was suspected, and azilsartan was stopped immediately. Without further specific treatment, a decrease in the frequency of the diarrhea was observed after 1–2 weeks.

As gastrointestinal symptoms resolved after discontinuing azilsartan, gluten was re-implemented into the diet on a trial basis 7 days later and was well tolerated. Ten days later, lactose was also re-implemented and tolerated well. The patient was transferred to outpatient care with antihypertensive therapy consisting of amlodipine (10 mg qd), moxonidine (0.2 mg qd), and bisoprolol (5 mg qd). There was no restriction regarding daily gluten and lactose intake.

Taking into account the clinical presentation with diarrhea and weight loss combined with a positive de-challenge and re-challenge with azilsartan, as well as the ongoing remission despite a gluten-containing diet, the diagnosis of sartan-induced enteropathy due to azilsartan treatment could be established in the patient presented in this case report.

When the patient was seen 5 months later, he was free of clinical symptoms. Upper endoscopy and colonoscopy showed normally composed duodenal mucosa and colonic biopsies free of inflammation (Fig. 2). The patient is advised to omit treatment with all ARBs.

This case illustrates sprue-like enteropathy as a potential side effect of ARBs, in this case azilsartan [8]. The causality of our findings was confirmed by drug de-challenge and re-challenge resulting in typical histological changes found in sprue-like enteropathy. Clinical symptoms included excessive diarrhea and weight loss. Blood and stool diagnostics showed only nonspecific elevation of inflammation markers. Endoscopic biopsies revealed diffuse lymphoplasmacytic inflammation and proliferation of intraepithelial lymphocytes in the presence of subtotal or partial villous atrophy in the duodenum and jejunum, consistent with celiac disease. However, celiac disease-specific antibodies were negative, and adherence to a strictly gluten-free diet did not show any clinical response.

Celiac disease is typically diagnosed based on positive serology (IgA anti-transglutaminase 2 in the presence of a normal serum IgA level) and histological confirmation of crypt hyperplasia and villous atrophy in biopsies taken from the duodenum. Clinically, patients with celiac disease often have signs and symptoms of malabsorption. Seronegative celiac disease rarely occurs (<2% of all cases) [9], and a histopathological workup is needed in these cases. In most patients, celiac disease is associated with an HLA-DQ2 or HLA-DQ8 genotype and with clinical and histological improvement when adhering to a gluten-free diet [10]. Other causes of similar histopathological results can be found in so-called celiac disease-like enteropathies: sartan-induced enteropathy, infectious causes like giardiasis, Whipple’s disease, tuberculosis, HIV, autoimmune enteropathy, common-variable immunodeficiency, radiation enteritis, Zollinger-Ellison syndrome, eosinophilic gastroenteritis, intestinal lymphoma, or Crohn’s disease [11]. The patient in this case report did not experience remission on a gluten-free diet. In addition, he did not suffer from any extraintestinal manifestations or complications, such as dermatitis herpetiformis, persistent iron deficiency, peripheral neuropathy, chronic fatigue, or celiac hepatitis, which are occasionally observed in adults with celiac disease [10]. Furthermore, he had no history of any autoimmune disorders. HLA genotyping was not performed due to the lack of therapeutic consequences.

The pathogenesis of sartan-induced enteropathy is not yet completely understood; however, several hypotheses have been suggested. ARBs act selectively at angiotensin II receptor subtype 1 (AT₁ receptor), leading to vasodilatation, which increases glomerular filtration and decreases cellular hypertrophy [12]. Angiotensin II receptor subtype 2 (AT₂ receptor) seems to initiate pro-apoptotic and anti-proliferative cellular processes [13]. By inhibiting AT₁ receptor via ARB therapy, a proportional increase in angiotensin II activity at the AT₂ receptor could be postulated. In addition, in animal studies, blockade of the AT₁ receptor has demonstrated an increase in expression of AT₂ receptors, upregulating angiotensin II action at that receptor [13]. According to Sun et al. [14], apoptosis in intestinal epithelial cells can be induced by angiotensin II through the AT₂ receptor. A cell-mediated immune response to the prodrug olmesartan medoxomil has been suggested to cause the mucosal damage associated with ARBs because long delays were observed between initiation of ARB treatment and the onset of adverse symptoms [2]. Similar to olmesartan medoxomil, azilsartan is typically given as azilsartan medoxomil, which is a prodrug, which is rapidly hydrolyzed by esterases in the gastrointestinal tract into the active drug azilsartan. Therefore, azilsartan medoxomil may have a similar risk for sartan-induced enteropathy as olmesartan medoxomil.

Kagami et al. [15] suggested that angiotensin II stimulates the synthesis of extracellular matrix proteins through induced expression of transforming growth factor-beta, a cytokine crucial in gut homeostasis [16]. Increased CD8+ cell infiltrates and a significant increase in interleukin-15 have been found in duodenal biopsies. The latter causes CD8+ T cells to be less sensitive to suppression by regulatory T cells, resulting in increased perishing of epithelial cells. In addition, changes in ZO-1, a tight-junction protein, cause disruption in the epithelial cells [17]. More than half of all patients diagnosed with sartan-induced enteropathy are HLA-DQ2/8 positive or suffer from other autoimmune diseases, suggesting a genetic predisposition to this disease entity in certain individuals. This also supports the hypothesis of an immune-mediated pathogenesis of sartan-induced enteropathy [2, 18].

The question arises as to whether sartan-induced sprue-like enteropathy should be regarded as a class effect of ARBs. Over the last few years, this adverse drug reaction has been described predominantly for olmesartan; however, there were similar symptoms and histological findings reported for the first time in a patient receiving telmisartan [6] in 2014, followed by several case reports of enteropathies associated with non-olmesartan ARB use [5‒7]. To date, the largest cohort was a database of 4,546,680 patients [19]. This study concluded that the risk of hospitalization due to intestinal malabsorption or celiac disease (diagnosis at discharge) was increased in olmesartan-treated patients compared to patients treated with ACE inhibitors or other ARBs [19]. However, these data were solely based on ICD-10 codes, not symptoms or histopathological findings that led to the diagnosis of intestinal malabsorption. Even though Basson et al. [19] suggested that the adverse drug reaction is drug-specific for olmesartan, emerging case reports of other ARBs being associated with sartan-induced enteropathy should not be neglected. Although olmesartan is the most common ARB associated with sartan-induced enteropathy, a current systematic review from Schiepatti et al. [20] concluded that there are well-described cases for nearly every available ARB. Furthermore, the similarity of currently available ARBs regarding selectivity at the AT₁ receptor, which plays a crucial role in the disease pathogenesis as outlined above, allows a class effect to be taken into consideration. Common molecular structures in all ARBs, which activate the same signaling pathways, lead to the desired treatment effect but may potentially cause similar side effects [21]. Azilsartan in particular shares a metabolization pathway with olmesartan, for which most cases of ARB-induced sprue-like enteropathy have been described. As mentioned above, both drugs are given as medoxomil-containing prodrugs and are metabolized into the bioactive component through esterases in the gut mucosa, portal blood, and liver [22, 23]. Considering that both drugs share drug-specific uptake and metabolization, as well as drug-induced signaling pathways, similar side effects may be expected. Reporting and observation bias may also be present as not all ARBs are prescribed equally as often. On the other hand, considering that sartan-induced enteropathy may develop even many months after the start of ARB treatment, adverse drug reactions of ARBs may be easily overlooked. Further large-scale studies have to be performed to significantly assess the association of sprue-like enteropathy with different ARBs and to gain detailed insight into the pathogenetic mechanisms of this adverse drug reaction. In conclusion, this case report demonstrates for the first time through azilsartan exposure, de-challenge, and re-challenge combined with the clinical and histological findings sprue-like enteropathy as a potential side effect of azilsartan, emphasizing to consider this differential diagnosis in ARB-treated patients with chronic diarrhea [8]. To comply with the quality criteria of a case report publication, the CARE checklist has been completed by the authors for this case report, attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000545217).

Written informed consent was obtained from the patient for this publication including consent for publication of the laboratory results and the accompanying histological and endoscopic images. Throughout the report, the patient’s identity has been kept anonymous. According to the local guidelines, this retrospective review of patient data did not require ethical approval.

The authors have no conflicts of interest to declare.

This research was performed as part of the employment of the authors; the employer is the Kantonsspital St. Gallen, Switzerland. There are no additional funding sources.

Luisa Stoeckli: writing – original draft and data curation. Markus Koster: writing – original draft. Tugce Tuerkmen-Uthayanan: collection of patient data and contribution to writing – original draft. Sergio Cogliatti: providing the histological images and review and editing of the manuscript. Stephan Brand: writing – original draft, review and editing, providing the endoscopic images, data curation, and validation, conceptualization, and resources.

Parts of this case report have been presented as poster presentations at the 6th Spring congress of the Swiss Society of General Internal Medicine (SGAI) on June 2, 2022 and at the Annual Meeting of the Swiss Society of Gastroenterology (SGG-SSG), the Swiss Society of Visceral Surgery (SGVC-SSCV), the Swiss Association for the Study of the Liver (SASL), and the Swiss Society of Endoscopy Nurses and Associates (SVEP-ASPE) on September 14, 2023. These poster presentations have been published in abstract form only [1, 8].

All data analyzed in this study are included in this article and its online supplementary materials. Additional inquiries should be directed to the corresponding author.