Cross-sectional study of associations between normal body weight with central obesity and hyperuricemia in Japan

Hyperuricemia is associated with increased risk of gout [1], hypertension, chronic kidney diseases, congestive heart failure, metabolic syndrome, type 2 diabetes mellitus and cardiovascular disease (CVD) [2]. Therefore, the prevention of hyperuricemia is an important public health issue.

Several studies have associated normal weight with central obesity (NWCO), defined by body mass index (BMI) and waist-to-height ratio (WHtR), and the CVD risk factors of hypertension, dyslipidemia and diabetes [3, 4]. In addition, NWCO is associated with a significantly higher risk for metabolic syndrome [5]. Thus, measuring degrees of central fat distribution appear important for the early detection of health risks, even among individuals with normal weight (NW) [6]. However, relationships between NWCO and hyperuricemia remain obscure.

The present study aimed to determine relationships between NWCO and hyperuricemia among Japanese middle-aged adults. We postulated that Japanese men and women are at higher risk for hyperuricemia when they have NWCO, than NW.

Table 1 shows the characteristics of study participants with mean ages of 50.4 and 50.6 years for men and women, respectively. Means serum uric acid values and the prevalence of hyperuricemia were higher in men than in women (6.1 vs. 4.6 mg/dL and 21.4% vs. 11.0%, respectively). The proportions of NWCO among men and women were 15.6 and 30.0%, respectively, and those of men and women who consumed alcohol were 73.2 and 46.3%, respectively.

Tables 2 and 3 respectively show the characteristics of men and women with hyperuricemia and normouricemia. Differences were statistically significant between the nature of the obesity and hyperuricemia in men and women. The prevalences of OBCO in hyperuricemia and normouricemia were 43.3 and 26.7% among men, and 44.8 and 19.1% among women. The prevalences of hypertension were 45.2 and 35.3% in men with hyperuricemia and normouricemia, respectively, and 43.8 and 28.4% in women with hyperuricemia and normouricemia, respectively. Those of dyslipidemia were 66.0 and 51.1% in men with hyperuricemia and normouricemia, and 60.7 and 41.0% in women with hyperuricemia and normouricemia. The prevalences of diabetes among men with hyperuricemia and normouricemia were 7.2 and 9.5%, while those among women with hyperuricemia and normouricemia were 8.0 and 3.6%. Moreover, the proportions of those who habitually consumed alcohol were higher in both men and women with, than without hyperuricemia.

Tables 4 and 5 show the crude and adjusted OR and 95%CI for hyperuricemia among men and women, respectively. The adjusted OR for hyperuricemia was significantly increased in men and women with OBCO compared with NW after adjustment for age, lifestyle factors, hypertension, dyslipidemia, and diabetes (OR, 2.12; 95%CI, 2.03–2.21 and OR, 3.54; 95%CI, 3.21–3.90, respectively). The OR for hyperuricemia were also significantly increased in men and women with NWCO compared with NW (OR, 1.44; 95%CI, 1.36–1.52 and OR, 1.41; 95%CI, 1.27–1.57, respectively). An analysis of subgroups stratified by alcohol intake found that the OR for hyperuricemia was significantly increased among men and women with OBCO who consumed alcohol (OR, 1.95; 95%CI, 1.86–2.05 and OR, 2.94; 95%CI, 2.59–3.35, respectively) and among those who did not (OR, 3.27; 95%CI, 2.97–3.60 and OR, 5.29; 95%CI, 4.50–6.22, respectively) compared with NW. The OR for hyperuricemia were statistically significant among men and women with NWCO who consumed alcohol (OR, 1.42; 95%CI, 1.34–1.51 and OR, 1.25; 95%CI, 1.09–1.42, respectively) compared with those who did not consume alcohol (OR, 1.56: 95%CI, 1.36–1.78 and OR, 1.85; 95%CI, 1.55–2.21, respectively). These results persisted even after adjustment for age, lifestyle factors, hypertension, dyslipidemia, diabetes, and the estimated glomerular filtration rate.

The present study investigated the relationships between NWCO and hyperuricemia in middle-aged Japanese adults. We found that hyperuricemia was significantly associated with NWCO and OBCO, compared with men and women of NW. The results were similar regardless of alcohol consumption.

The mean serum uric acid value was higher in men than in women. Several studies have investigated serum uric acid levels in Japanese adults. Reported serum uric acid levels for men and women are respectively, 6.2 and 4.4 mg/dL [14], 5.9 and 4.2 mg/dL [15], 6.5 and 4.7 mg/dL [16], and 6.0 and 4.5 mg/dL [22]. Therefore, our findings of higher serum acid levels in men were similar to these findings.

Several epidemiological studies reported that hyperuricemia was associated with diseases including diabetes mellitus, dyslipidemia, hypertension, CVD, and metabolic syndrome [2, 8, 14, 16]. Similarly, the present study showed that the prevalences of hypertension, dyslipidemia, and diabetes were higher with than without hyperuricemia in both men and women. Moreover, we also found that OBCO significantly increased the OR for hyperuricemia, compared with NW, regardless of sex. Others have positively associated hyperuricemia with obesity in both men and women [16, 23]. Nagahama et al. reported that the adjusted OR (95%CI) for hyperuricemia in men and women with BMI ≥25 kg/m² were 1.75 (1.56–1.97) and 2.02 (1.62–2.53) [16]. Yu et al. reported that hyperuricemia was related to abdominal obesity and general obesity in both sexes [23]. The present results are in line with these findings. Even after adjustment for age, lifestyle factors, hypertension, dyslipidemia, and diabetes, our results showed that OBCO had an independently increased risk of hyperuricemia and this was statistically significant between OBCO and hyperuricemia. As one of the reasons, obesity may increase hyperuricemia through increased urate production and decreased renal clearance, and that renal excretion of urate was also reduced in the presence of insulin resistance [24]. However, the underlying mechanism by which serum uric acid increased in obese individuals still remains to explore [19]. Thus, more studies need to be conducted in order to elucidate the mechanism of the association between serum uric acid and obesity.

In addition to OBCO, NWCO was also significantly associated with hyperuricemia compared with NW. A previous study in Japan found significantly increased OR (95%CI) for hyperuricemia in men with NWCO compared with NW, but not among women: 1.43 (1.15–1.76) and 2.20 (0.36–33.35), respectively [6]. We found that NWCO was associated with hyperuricemia risk compared with NW, regardless of sex. Moreover, WC and uric acid levels demonstrated a positive correlation in NW and NWCO group, regardless of sex (men, r = 0.19; P <  0.001, women, r = 0.23; P <  0.001). Higher and lower BMI are associated with an increased and decreased prevalence of hyperuricemia, respectively [8]. Therefore, because individuals with NWCO are considered to be of normal weight, they do not usually receive appropriate health education or prompt intervention to prevent hyperuricemia. Our findings suggest that men and women with NWCO need to be identified in addition to those with OBCO, even if their weight is normal.

Alcohol consumption is also an important risk factor for hyperuricemia [8, 25‐27]. We stratified men and women according to alcohol intake and found that the OR for hyperuricemia were significantly increased in those with OBCO, compared with NW, regardless of alcohol consumption. That is, hyperuricemia was positively associated among men and women with OBCO, regardless of alcohol intake. Moreover, the OR for hyperuricemia were significantly increased in men and women with NWCO, compared with NW, regardless of whether alcohol was consumed. Therefore, screening men and women for NWCO and active intervention to prevent hyperuricemia are important, regardless of alcohol intake.

To the best of our knowledge, this is the first investigation into the relationships between hyperuricemia and NWCO among middle-aged Japanese adults. The strengths of the present study are the large sample size (about 100,000 participants), which helped to decrease random error, and fact that trained technicians measured the anthropometric variables of height, weight and WC of the participants, that were used to define OB and CO. Serum uric acid was measured using a standardized method at a clinical testing laboratory. In contrast, the limitations of the present study include potential confounding factors that were not determined in this study that might have affected our findings. For instance, we did not collect details of dietary intake such as the ingestion of large amounts of purine sources (animal protein and beer) [2]. Another limitation is that the cross-sectional study design caused difficulties with assessing causal relationships. For the nature of the study design, we did not observe these weight variations, although weight was very important clinical indicator in order to determine hyperuricemia throughout research in an individual. Thus, further longitudinal studies will be needed to establish causality and to consider weight variation.

The present study showed that being of NW but having CO, and being obese in general were associated with hyperuricemia among middle-aged Japanese men and women. These findings suggest that middle-aged Japanese men and women with NWCO should be identified using a combination of BMI and WHtR, then educated about the prevention of hyperuricemia.