Association of home hypertension subtypes with glucose metabolism.

Abstract

Objective: This study was performed to investigate the association of home and office hypertension with glucose metabolism among the Japanese general population. Methods: The study involved 646 residents without treatment for hypertension or a history of diabetes from Ohasama, a rural Japanese community, who underwent an oral glucose tolerance test measuring blood glucose (BG) and insulin levels at fasting and 30, 60 and 120 min. Hypertension subtypes [normotension (NT), isolated systolic hypertension (ISH), isolated diastolic hypertension (IDH), and systolic and diastolic hypertension (SDH)] were defined based on home and office systolic and diastolic blood pressure (HBP and OBP, respectively). The estimated means of blood glucose (BG)-related indices among the groups were compared by analysis of covariance adjusted for possible confounding factors. In addition, the association between continuous HBP and OBP (as independent variables) and BG-related indices (as dependent variables) were assessed by multiple linear regression analysis. Results: Mean age was 62.4 years. Of participants, 29% were men. No estimated mean BG-related indices were different among the morning HBP-defined hypertension subtypes, although morning systolic BP and BG at 120 min were significantly positively associated. Participants with evening HBP-defined ISH had a significantly higher estimated mean BG at 120 min (146 vs 127 mg/dl), higher homeostasis model assessment-insulin resistance (HOMA-IR) (1.29 vs 1.02), and lower Matsuda–DeFronzo index (6.55 vs 8.06) than participants with NT. In addition, evening systolic BP levels with BG at 60 and 120 min and HOMA-IR and diastolic BP levels with fasting BG were significantly positively associated. Participants with OBP-defined SDH had significantly higher estimated means of fasting BG (101 vs 94 mg/dl), BG at 30 min (177 vs 158 mg/dl), 60 min (181 vs 157 mg/dl), and 120 min (145 vs 125 mg/dl); and HOMA-IR (1.30 vs 0.98) and a lower Matsuda–DeFronzo index (6.26 vs 8.30) than participants with NT. Both office systolic and diastolic BP levels were associated with BG levels at all time points, HOMA-IR and Matsuda–DeFronzo index. Conclusion: Evening HBP and OBP were clearly associated with BG-related indices. Participants with evening HBP-defined ISH and OBP-defined SDH had higher BG levels and insulin resistance than participants with correspondingly defined NT. However, morning HBP were not obviously associated, compared with evening HBP and OBP. These findings suggest the importance of measuring evening HBP and OBP for early detection of coexisting hypertension and diabetes.