Introduction: Recent data suggest Allopurinol use is associated with a decreased risk of stroke but the mechanism is unclear. Atrial fibrillation (AF) is a known risk factor for stroke and Allopurinol has numerous cardiac effects including left ventricular hypertrophy regression and improved myocardial energetics. We hypothesized that Allopurinol use is associated with a reduced risk of AF and tested this using data from the UK Clinical Practice Research Datalink (UKCPRD).
Methods: The UKCPRD contains verified prescription data. We included older adults (>65 years) with hypertension to miminise income-based confounding (free prescriptions in UK), ensure adequate follow up (subject to primary care surveillance in UK) and because these are risk factors for AF. Multivariable Cox proportional hazards models, assessing the effect of Allopurinol exposure over a 10-year period were used to estimate hazard ratios for AF. Allopurinol exposure was a time-dependent variable. A propensity-matched design was also used to reduce potential for confounding along with independent analyses for high (≥300mg) and low (<300mg) dose Allopurinol.
Results: A total of 2368 Allopurinol exposed and 41663 unexposed patients were included in our UKCPRD extract. After propensity matching (n=2074 per group) there was no relationship between Allopurinol exposure and risk of AF (HR=0.881, 95% CI (0.730, 1.062). The same was seen with high dose Allopurinol (n=1081 per group) (HR=0.881, 95% CI (0.699, 1.125). Low dose Allopurinol (n=993 per group) was associated with reduced risk of AF (HR=0.731, 95% CI (0.555, 0.961). This association persists when comparing low vs. high dose Allopurinol (n=1789) (HR=0.698, 95% CI (0.534, 0.912).
Conclusions: Use of low dose Allopurinol was associated with a decreased risk of AF in a large cohort of hypertensive patients. However, this was not evident in the high dose Allopurinol patients. Reasons for this warrant further examination.