Background Few longitudinal studies on the determinants of increase in seru
m uric acid (SUA) have been completed.
Methods In all, 1445 hyperuricaemia-free (<7.5 mg/dl SUA, no medication for
and no past history of hyperuricaemia) male office workers aged 30-54 year
s of T Corporation in Osaka, Japan were re-examined for six successive year
s. Subjects who were found to be hyperuricaemic or had started medication f
or hyperuricaemia during repeat surveys were defined as incident cases.
Results Among the subjects (n = 1365) not receiving medication for hyperten
sion, diabetes mellitus or renal disease, multivariate analysis using the C
ox proportional hazards model indicated that the incidence of hyperuricaemi
a had significant relationships with body mass index (adjusted hazard ratio
[HR] = 1.13 for a 2 kg/m(2) increase; 95% CI: 1.02-1.26), mean blood press
ure (HR = 1.07 for a 5 mmHg increase; 95% CI: 1.00-1.13), log triglyceride
level (HR = 2.21 for a 10 mg/dl increase; 95% CI: 1.12-4.37), alcohol intak
e (HR = 2.33 for drinking 46.0 g of ethanol per day or more relative to non
-drinking; 95% CI: 1.55-3.50) and smoking (HR = 0.65 for current-smoking re
lative to non-smoking; 95% CI: 0.46-0.92). Age (HR = 0.89 for a 5-year incr
ease; 95% CI: 0.78-1.00) and haemoglobin A(1c) (HbA(1c),) (HR = 0.89 for a
0.5% increase; 95% CI: 0.78-1.00) achieved marginal significance.
Conclusions Obesity, high blood pressure, high triglyceride level, and alco
hol intake are contributory factors for the development of hyperuricaemia a
mong middle-aged Japanese men. High HbA(1c), level and smoking may be negat
ive factors for the incidence of hyperuricaemia.