Akiko Yamada1, Kyoko K Sato1, Shigeki Kinuhata1, Shinichiro Uehara1, Ginji Endo1, Yonezo Hikita2, Wilfred Y Fujimoto3, Edward J Boyko4, Tomoshige Hayashi5. 1. Osaka City University Graduate School of Medicine, Osaka, Japan. 2. Ohtori Health Promotion Center, Sakai, Japan. 3. University of Washington, Seattle. 4. Epidemiologic Research and Information Center, Veterans Affairs Puget Sound Health Care System, and University of Washington, Seattle. 5. Osaka City University Graduate School of Medicine, Osaka, Japan, and University of Washington, Seattle.
Abstract
OBJECTIVE: To examine cross-sectionally whether intraabdominal fat area (IAFA), i.e., visceral fat, and liver fat assessed by computed tomography (CT) are independently associated with hyperuricemia. METHODS: Subjects were 801 Japanese men not taking antidiabetic, antihypertensive, or urate-lowering medications, without any history of renal disease, cardiovascular disease, or cancer, and with serum creatinine <1.5 mg/dl. Abdominal, thoracic, and thigh fat areas were measured by CT. Total fat area (TFA) was the sum of these fat areas. Total subcutaneous fat area (TSFA) was TFA minus IAFA. Liver fat was assessed by liver-to-spleen (L/S) ratio measured by CT. Hyperuricemia was defined as serum uric acid level >7.0 mg/dl. Its association with adiposity was tested using logistic regression. RESULTS: The prevalence of hyperuricemia was 19.6% (157 men). Both greater IAFA and lower L/S ratio were independently associated with hyperuricemia in models that simultaneously included IAFA and L/S ratio: multiple-adjusted odds ratios of hyperuricemia for quintiles 3, 4, and 5 of IAFA were 2.16 (95% confidence interval [95% CI] 1.02-4.59), 2.41 (95% CI 1.13-5.16), and 4.00 (95% CI 1.81-8.85), respectively, compared to quintile 1, and the L/S ratios for quintiles 3, 2, and 1 were 2.34 (95% CI 1.16-4.75), 2.15 (95% CI 1.06-4.34), and 2.79 (95% CI 1.35-5.76), respectively, compared to quintile 5. Both IAFA and L/S ratio remained significant even after adjusting for abdominal subcutaneous fat area, TFA, TSFA, body mass index, or waist circumference. Of all fat measurements, IAFA had the strongest association with hyperuricemia by Akaike's information criteria. CONCLUSION: Greater amounts of both visceral fat and liver fat were independently associated with hyperuricemia.
OBJECTIVE: To examine cross-sectionally whether intraabdominal fat area (IAFA), i.e., visceral fat, and liver fat assessed by computed tomography (CT) are independently associated with hyperuricemia. METHODS: Subjects were 801 Japanese men not taking antidiabetic, antihypertensive, or urate-lowering medications, without any history of renal disease, cardiovascular disease, or cancer, and with serum creatinine <1.5 mg/dl. Abdominal, thoracic, and thigh fat areas were measured by CT. Total fat area (TFA) was the sum of these fat areas. Total subcutaneous fat area (TSFA) was TFA minus IAFA. Liver fat was assessed by liver-to-spleen (L/S) ratio measured by CT. Hyperuricemia was defined as serum uric acid level >7.0 mg/dl. Its association with adiposity was tested using logistic regression. RESULTS: The prevalence of hyperuricemia was 19.6% (157 men). Both greater IAFA and lower L/S ratio were independently associated with hyperuricemia in models that simultaneously included IAFA and L/S ratio: multiple-adjusted odds ratios of hyperuricemia for quintiles 3, 4, and 5 of IAFA were 2.16 (95% confidence interval [95% CI] 1.02-4.59), 2.41 (95% CI 1.13-5.16), and 4.00 (95% CI 1.81-8.85), respectively, compared to quintile 1, and the L/S ratios for quintiles 3, 2, and 1 were 2.34 (95% CI 1.16-4.75), 2.15 (95% CI 1.06-4.34), and 2.79 (95% CI 1.35-5.76), respectively, compared to quintile 5. Both IAFA and L/S ratio remained significant even after adjusting for abdominal subcutaneous fat area, TFA, TSFA, body mass index, or waist circumference. Of all fat measurements, IAFA had the strongest association with hyperuricemia by Akaike's information criteria. CONCLUSION: Greater amounts of both visceral fat and liver fat were independently associated with hyperuricemia.