Risk of serious adverse gastrointestinal events when using potassium binders for the treatment of hyperkalemia in hospitalized patients

SESSION TITLE: Education, Research, and Quality Improvement Posters 1 SESSION TYPE: Original Investigation Posters PRESENTED ON: 10/10/2023 12:00 pm - 12:45 pm PURPOSE: Concerns about severe GI adverse effects with sodium polystyrene sulfonate (SPS) have led to the development of two new potassium binder cation exchange resins, patiromer and sodium zirconium cyclosilicate (SZC), for the treatment of hyperkalemia. We examined whether their use in hospitalized patients is associated with intestinal ischemia/thrombosis or other severe GI adverse events. METHODS: We included patients admitted to Veterans Affairs hospitals for 30 days or less between 1/1/2016 -9/30/2022 who were prescribed SPS, patiromer, SZC, or no potassium binder. Primary outcomes were 1) intestinal ischemia/thrombosis and 2) a composite outcome of major adverse GI events (intestinal ischemia/thrombosis, GI ulceration/perforation or bowel resection/ostomy) within 30 days. We used ICD-9 and 10 codes for ischemia/thrombosis and GI ulceration/perforation and CPT codes for bowel resection/ostomy. Patients were excluded if they received a potassium binder in the previous 180 days or if they had prior intestinal ischemia/thrombosis, GI ulceration/perforation or resection/ostomy. A generalized estimating equation (GEE) logistic regression was used to evaluate the association between receipt of the 3 potassium binders and 30-day risk of intestinal ischemia/thrombosis and the composite outcome, accounting for multiple admissions for the same patient nested within VHA facilities using an autoregressive correlation structure. RESULTS: Our cohort included 3,146,229 admissions from 1,462,218 unique Veterans. Of these, no potassium binders were identified for 3,105,650 admissions, while 30,040 (1.0%) had SPS, 3,750 (0.1%) had patiromer, and 5,520 (0.2%) had SZC. Over the defined 30-day risk period, Intestinal ischemia/thrombosis occurred in 0.2% of admissions without potassium binder, as compared to 0.4% with SPS, 0.3% with patiromer and 0.4% with SZC. The corresponding incidences of composite GI adverse events were 2.4%, 2.5%, 2.6%, and 2.6% respectively. After accounting for potential confounding (age, sex, race, ethnicity, Charlson Comorbidity Index, and admission years), the odds ratio (OR) (95% CI) for intestinal ischemia/thrombosis with SPS was 1.42 (1.18,1.72) p <0.001, patiromer=1.36 (0.79, 2.36) p = 0.3, and SZC=1.80 (1.22,2.65) p=0.003; The OR (95% CI) for composite GI adverse events with SPS was 0.99 (0.93-1.06) p=0.8, patiromer=1.07 (0.87-1.30) p=0.5, and SZC=1.07 (0.92, 1.26) p=0.4. CONCLUSIONS: After accounting for potential confounding, SPS and SZC use were associated with increased risk of intestinal ischemia/thrombosis. Associations with intestinal ischemia/thrombosis or composite GI adverse events do not differ statistically between the 2 newer potassium binders and SPS. Potassium binder use was not associated with a composite outcome of major GI adverse events. (p>0.05). CLINICAL IMPLICATIONS: While the incidence of intestinal ischemia/thrombosis is low albeit potentially life threatening, our data suggests that increased risk may be a class effect and not limited to a single drug. DISCLOSURES: No relevant relationships by Kathleen Akgun No relevant relationships by Lori Bastian No relevant relationships by Cynthia Brandt No relevant relationships by craig gunderson No relevant relationships by Ling Han No relevant relationships by Jurgen Holleck No relevant relationships by Melissa Perkal No relevant relationships by Melissa Skanderson