Background There is evidence that xanthine oxidase inhibitors (XOI) may reduce the risk of major adverse cardiovascular events (MACE) (1,2) and lower blood pressure (3). To date, this evidence is based mainly on observational studies (2).
Objectives To compare the incidence of MACE, mortality, and total and specific cardiovascular (CV) events in patients enrolled in randomized controlled trials (RCTs) comparing XOI with placebo or no treatment.
Methods A systematic review (CRD42015016073) searching for RCTs using PubMed, EMBASE, Cochrane Library, Web of Science, and Lilacs databases, and hand searching, was ended in Dec 2016. All RCTs comparing XOIs with placebo or no treatment lasting ≥4 weeks and including only adult individuals were eligible. The primary outcomes were the incidence of MACE (CV death, non-fatal myocardial infarction, unstable angina requiring urgent revascularization, or non-fatal stroke) and mortality; total CV events (TCE), specific CV outcomes, and serious adverse events (SAE) served as secondary outcomes. Associations were tested using the Peto odds ratio (OR) without zero-cell continuity correction.
Results In total, 81 studies including approx. 11,000 individuals reported extractable data on CV events. The use of XOI tended to be associated with lower incidence of MACE (OR=0.64, 95% CI 0.41–1.01, P=0.056), but not with mortality (0.95, 0.63–1.44). However, there was a significantly reduced incidence of TCE (0.66, 0.54–0.80, P<0.001), especially new/worsening hypertension (0.57, 0.37–0.87, P=0.009), and a trend for reduction in the incidence of new/worsening heart failure (0.74, 0.53–1.04, P=0.086). The incidence of SAE (0.86, 0.71–1.05) did not differ significantly. Subgroup analysis suggested a protective effect for MACE in studies with high prevalence (>50%) of cardiac diseases (0.52, 0.30–0.91, P=0.021). Sensitivity analysis excluding studies at high or unknown risk of bias produced no significant change in results, but reinforced the association with reduced incidence of hypertension (0.26, 0.11–0.60, P=0.001) and heart failure (0.55, 0.32–0.94, P=0.030). Separate analysis of data on purine-like XOI (allopurinol and oxypurinol) confirmed the results of the primary analysis. Exploratory metaregression analysis showed association of dose of allopurinol with higher incidence of TCE (P=0.023, random effects) and SAE (P<0.001, see Figure 1). Accordingly, in the subgroup with doses ≤300 mg/day of allopurinol, a reduction of incidence of MACE (0.36, 0.18–0.68, P=0.002), TCE (0.38, 0.26–0.54, P<0.001), and SAE (0.49, 0.34–0.71, P<0.001) was observed, while the SAE risk increased in doses >300 mg/day (1.39, 1.04–1.91, P=0.047). There was no association of dose of non-purine-like XOI with incidence of TCE and SAE. Significant statistical heterogeneity was not observed in any test reported here.
Conclusions Our data from a meta-analysis of RCTs suggest that XOI reduce the incidence of CV events, an effect possibly related (at least partly) to control of hypertension. However, higher doses of allopurinol (>300 mg/day) may possibly be associated with higher risk of serious adverse events and loss of cardiovascular protection.
Richette et al. Nat Rev Rheumatol 2014;10:654–61.
Singh et al. Arthritis Res Ther 2016;18:209.
Agarval et al. J Clin Hypertens (Greenwich) 2013;15:435–42.
Disclosure of Interest None declared