Cardiovascular disease (CVD) mortality is the leading cause of death in most developed countries. Coronary heart disease (CHD), one of the four diagnostic categories of CVD, accounts for approximately one third to one half of CVD cases. It is well established that diabetes mellitus (DM) and chronic kidney disease (CKD) are considered as two most important CHD risk equivalents [7, 12, 14, 18]. Therefore, when this study is to test whether gout is an emerging risk factor for CVD, the confounding factors from DM and CKD should be tackled first using a different study design.
To our knowledge, this study is the first of its kind to examine the association of gout and CVD mortality using a nationwide population-based cohort study with an ad hoc design to deliberately exclude patients with DM and then stratify cohorts into groups with or without CKD at the beginning of the study. The advantage of excluding all patients with type 2 diabetes before forming study cohorts precludes the necessity in the future resorting to statistical adjustments which would be suboptimal in terms of evidence-based approach. This study also excludes patients with pre-existing major cardiovascular and/or cerebrovascular co-morbidities in the preceding one year of the study. The strength of this study includes the complete follow-up with no drop-outs due to nationwide registration of the subjects’ use of any medical resource in the country.
Our results show that the distribution of cases of gout declines gradually from 33.19% in the age group of 50–59 years down to 8.77% in the age group of ≥ 80 years (Table 1). This difference of the prevalence of gout in different age group can be explained by the fact that subjects who live longer are less likely afflicted with gout problems or its related complications. The strengths of this study is using smoking-related diagnosis, alcoholism-related diagnosis, Charlson-Deyo comorbidity index score to avoid the occurrence of recall bias from interview-based or questionnaire-based study design [16, 17].
Our results show that gout is an independent risk factor for CVD mortality. The relative risk for CVD mortality as the cause of death in the next five years is 1.71 with a narrow 95% confidence interval of 1.66 to 1.75. The crude hazard ratio for subsequent death from CVD in subjects with gout is 1.69 (95% CI, 1.65-1.74). After adjustment made for gender, age, smoking-related diagnosis, alcoholism-related diagnosis, hypertension, hyperlipidemia, atrial fibrillation and Charlson’s comorbidity index score, the multivariable Cox regression analysis still shows a statistically significant hazard ratio of 1.10 (95% CI, 1.07-1.13). This confirms that gout alone modestly increases the risk for subsequent CVD mortality in the next five years.
Review of the literature on the association between gout and CHD or CVD mortality discloses four relevant studies of heterogeneous study design and different follow-up time [3, 5, 19, 20]. The Health Professionals Follow Up Study and the Multiple Risk Factor Intervention Trial included Caucasian male subjects only[5, 19]. The HR for cardiovascular mortality in the latter trial was not statistically significant exhibiting an adjusted HR of 1.21 (95% CI, 0.99-1.49) . A single institution observational study in Taiwan using health screening program subjects shows that the adjusted HR for CVD death is 1.97 (95% CI, 1.08-3.59) in participants with gout . This study had short follow up time (mean, 56 months) and did not make adequate adjustment for significant confounders such as medical comorbidities. A recently published study from Singapore which included only Chinese aged 45 to 74 years with gout has demonstrated an HR for CHD mortality of 1.38 (95% CI, 1.10 - 1.73) . Nevertheless, this study necessitated the participants to recall a history of previously diagnosed gout as their standard definition of gout. Furthermore, the Singaporean study was not able to adequately adjust for the participants’ coexisting medical co-morbidities. Our present study was able to use physician-diagnosed gout as the basis of disease confirmation and adding Charlson comorbidity index score for adjustment in the multivariable Cox regression.
This study shows that gout coexists with CKD in 8.18% in subjects with gout whereas only 1.44% of subjects in the control group have had CKD. No causal relationship can be extrapolated from the current study design. Our study demonstrated that in subjects with no gout, the presence of CKD exhibited a RR of 3.05 (95% CI, 2.94 – 3.15) of CVD mortality. It is surprising to note that gout attenuates the association between CKD and the risk for CVD mortality in patients with CKD coexisting with gout demonstrating a RR of 1.84 (95% CI, 1.71 - 1.98). In a Cox proportional hazard model with multivariable adjustment, when compared to control without CKD (subjects with neither gout nor CKD), the group of subjects with no gout but with CKD has an adjusted HR of 1.76 (95% CI, 1.70 – 1.82), however, the coexistence of gout plus CKD pulls down the adjusted HR to 1.38 (95% CI, 1.29 - 1.48) (Table 3).
Our study is the first to show that the presence of gout may have protective effects against cardiovascular mortality in patients with chronic kidney disease. The possible explanations for this observation may not be the gout per se but on the medications used for managing gout such as urate-lowering drug, allopurinol [21–25]. Allopurinol, a xanthine oxidase inhibitor, is the most commonly used medication for urate-lowering purpose in gout and hyperuricemia. Preclinical data have shown that allopurinol has protective effect on both myocardial and renal ischemia-reperfusion injury [22, 24]. In an observation study, allopurinol was associated with a lower risk of all-cause mortality (HR 0.78; 95% CI 0.67, 0.91) in subjects with hyperuricemia . A small-scale prospective open-label randomized trial carried out in patients with chronic kidney disease with patients randomly assigned to treatment with allopurinol 100 mg/day (n = 57) or to continue the usual therapy (n = 56), allopurinol decreases C-reactive protein and slows down the progression of renal disease in patients with CKD . In addition, allopurinol reduces cardiovascular and hospitalization risk in these subjects. Allopurinol treatment reduces risk of cardiovascular events in 71% compared with standard therapy with an adjusted HR for new cardiovascular events such as congestive heart failure, cerebrovascular accidents, ischemic coronary events, and peripheral arteriopathy, at 0.29 (95% CI, 0.09 – 0.86) . Therefore, our present study with complete follow up demonstrating attenuated risks for subsequent cardiovascular mortality by the coexistence of gout and CKD may be explained by allopurinol prescribed in patients with gout among other explanatory variables. Further studies should examine the association between allopurinol use and the cardiovascular mortality.
One of the potential limitations of this study is whether the diagnosis of physician-diagnosed gout possibly contains pseudogout and other crystalline arthritides. Because each gout case was identified as having made three separate visits (same ICD code given on at least three clinical encounters) and this is a strict criterion for physician-diagnosed gout adopted by almost all research on administrative database, the potential misclassification bias is considered very modest due to relative ease of making a gout diagnosis by physicians in Taiwan . We do not have access to the relevant biochemical datasets and lifestyle personal histories such as abdominal obesity, psychosocial factors, consumption of fruits and vegetables, and regular physical activity. This is because individual identities are not available due to the de-identification of the individuals within the NHI databases. Another potential limitation is that data for medication and treatment for gout and CKD and the adherence to treatment were too complex to incorporate in the covariate analyses. Furthermore, the impact of undiagnosed diabetes mellitus cannot be excluded because diabetes and metabolic syndrome is related to both gout and cardiovascular mortality [27–29].
In summary, the results clearly show that, among individuals without diabetes who have no major cardio-cerebrovascular disease, gout presents a higher risk of death from cardiovascular causes.