Background Hyperuricemia is an established risk factor for gout, but there is less certainty regarding the effect of serum uric acid (SrUA) levels on cardiovascular (CV) disease.
Objectives To use electronic health records (EHR) data to evaluate the relationship between SrUA levels and risk of CV disease in patients with gout.
Methods This retrospective study used de-identified EHR data from the Humedica database from 2010 through 2013. Patients were adults (≥18 years) with ≥2 ICD-9-CM diagnoses for gout ≥30 days apart (the first diagnosis was the index date), with ≥1 SrUA assessments on or after the index date, and ≥6 months pre- and ≥12 months post-index activity. ICD-9 codes were used to identify CV diseases including myocardial infarction (MI), congestive heart failure (CHF), and coronary artery disease (CAD). Patients were stratified by four defined, clinically relevant SrUA ranges based on their SrUA assessment closest to the CV diagnosis (those with CV diagnoses), or their last assessment during the study period (those without a CV diagnosis): Group 1, 0.01-4.0mg/dL (n=5594); Group 2, 4.1-6.0mg/dL (n=9267); Group 3, 6.1-8.0mg/dL (n=12,548); and Group 4, ≥8.1mg/dL (n=9267). Survival curves for each group were generated using Kaplan-Meier (KM) estimates and compared using the log-rank test; patients with each of the CV diagnoses during the 6-month pre-index period were excluded from that particular analysis.
Results Patients were mostly male (71.6%), white (79.4%) with a mean ± SD age of 62.4±13.3 years and a mean ± SD Charlson Comorbidity Index of 0.81±1.31. KM curves for diagnosis of any CV event during the 12-month post-index period (Figure) showed significant differences across the groups (P<0.0001). In pairwise comparisons, Group 4 (≥8.1mg/dL) had a significantly higher incidence of any CV diagnosis, 17.4% (P<0.05 vs all other groups), and Group 1 (0.01-4.0mg/dL) was significantly greater relative to Group 2 (14.4% vs 10.5%; P<0.05) but similar to Group 3 (13.4%). In analyses of individual CV diagnoses, although the incidence of MI was low in all groups, significant differences were observed only for Group 4 (1.8%) relative to Group 2 (1.2%; P<0.0001) and Group 3 (1.5%; P<0.05). However, Group 4 also had significantly higher incidence of CHF (8.2%) and CAD (13.4%) diagnoses relative to all other groups (P<0.0001 for all pairwise comparisons). Group 1 (0.01-4.0mg/dL) had a significantly higher incidence of CHF (5.9%) relative to Group 2 (4.3%; P<0.0001) and Group 3 (4.8%; P<0.001). Similarly for CAD, the incidence in Group 1, 11.6%, was higher than for Groups 2 and 3, 10.2% and 11.1%, respectively, but the difference was only significant relative to the former (P<0.05). Median times to all CV diagnoses were consistently shorter in Group 4.
Conclusions High SrUA appears to be associated with increased risk of CV disease, which was primarily driven by CAD and CHF, and very low SrUA may also increase the CV risk. However, neither the effects of pharmacologic therapy for hyperuricemia on these outcomes nor the longitudinal changes in SrUA were determined. Further research is warranted.
Disclosure of Interest M. Essex Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, M. Hopps Employee of: Pfizer Inc, M. Udall Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, C. Fu Consultant for: Pfizer Inc, E. Bienen Consultant for: Pfizer Inc, J. Mardekian Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, G. Makinson Shareholder of: Pfizer Inc, Employee of: Pfizer Inc