THU0503 Major Cardiovascular Events in Patients with Recurrent Gout, Cardiovascular and Chronic Kidney Disease, Initiating on Allopurinol or Febuxostat

BackgroundChronic management of hyperuricemia with xanthine oxidase inhibitors has been shown to reduce long term incidence of gout flares, but their resultant impact in highly comorbid populations is unclear.ObjectivesTo compare major cardiovascular event (MCE) rates in patients with gout, moderate to severe chronic kidney disease (CKD), and cardiovascular disease (CVD) after initiation of uric-lowering therapy (ULT), with allopurinol (ALO) or febuxostat (FBX).MethodsGout patients (ICD-9-CM 274.xx), aged ≥18, with a baseline diagnosis of stage 3/4 CKD, CVD [coronary artery disease (CAD), cerebrovascular disease (CBV), peripheral vascular disease (PVD)] and/or heart failure (HF), and ≥2 baseline gout flares were selected from the MarketScan® database (January 2009-June 2013) after initiating ULT with either ALO or FBX. Patients were followed until disenrollment, discontinuation of the qualifying ULT or initiation of the alternate study agent. MCE included myocardial infarction, stroke, transient ischemic attack, non-traumatic lower extremity (LE) amputation, and coronary, cerebrovascular, or LE revascularization. Cox proportional hazard models assessed MCE risk as a function of demographics, CVD and HF history, including CVD hospitalization, gout severity, resource intensivity and history of arrhythmia/conduction disorder.ResultsA total of 352 patients (148 FBX, 204 ALO) met eligibility criteria (69% male; mean age 71).A majority of patients (74%) had stage 3 CKD and, by definition, all patients had CVD (68% CAD, 15% CBV, and 23% PVD) and/or HF (53%) with no significant differences between cohorts. Mean daily dose was 51 mg in FBX and 198 mg in ALO cohorts. Mean baseline flares per patient were significantly higher in the FBX cohort (2.5 vs. 2.3, p=0.005). A total of 28 MCE occurred in 5.4% and 9.8% of FBX and ALO patients respectively. A total of 15 patients (4.3%) had a CAD-specific MCE, 4 (1.1%) a CBV-specific MCE, and 10 (2.8%) a PVD-specific MCE. The MCE rate per 1000 PY (95% CI) was 75.8 (33–149) in the FBX cohort vs. 128.1 (78–198) in the ALO cohort. Cox model results suggest significantly lower MCE risk in patients initiating on FBX (hazard ratio (HR) 0.4; 95% confidence interval (CI) 0.1–0.9, p=0.035) and those with a HF history (HR 0.4, 95% CI 0.16–0.93, p=0.034). Model results suggest increased risk associated with PVD history (HR 5.3; 95% CI 2.3–12.0, p