Association of microalbuminuria with diabetes is stronger in people with prehypertension compared to those with ideal blood pressure

Abstract

Aim: Microalbuminuria (MA) has been demonstrated as a biomarker for microvascular dysfunction. This study is aimed to evaluate the association of glycaemic status with MA in prehypertensive and ideal BP subjects and to evaluate the interaction between glycaemic and blood pressure status as risk factors for MA prevalence.

Methods: 1059 subjects aged 40-70 with non-hypertension who were recruited from six districts of Tianjin were divided into a prehypertensive group (622 cases) and an ideal blood BP group (437 cases). Subjects of the prehypertensive group and the ideal BP group were divided respectively into three subgroups: normoglycaemia subgroup, prediabetes subgroup and diabetes subgroup. The prevalence of MA in the above three subgroups of subjects with prehypertension and ideal BP were assessed. We performed a statistical analysis for interaction test between glycaemia and BP status on microalbuminuria in the overall study sample by a multivariate logistic regression model. The association of glycaemic status (defined as normoglycaemia, prediabetes, and diabetes) with MA was evaluated separately in prehypertensive and ideal BP subjects.

Results: Results showed that the prevalence of MA in both prehypertensive and ideal BP groups rose with the increasing of classification of glycaemic level of subgroups (32.6%, 18.3%, 14.8% vs. 23.1%, 16.2%, 13.4%), the differences in prehypertensive group were statistically significant (Pearson χ² = 15.24, P < 0.001). The ORs (95% CI) of MA were 1.25 (0.86-1.83) for prediabetes and 2.56 (1.62-4.03) for diabetes in the fully adjusted model. There was no interaction between prediabetes and BP status regarding MA (P = 0.237) but we found a significant interaction between diabetes and BP status (P < 0.001). In the prehypertensive group, multivariate logistic regression models showed that the diabetes subgroup had a significant association with MA, and the adjusted odds ratio of the diabetes subgroup to the normoglycaemia subgroup was 2.68 (95%CI 1.54-4.67) (P < 0.001). However, there was no significant association of glycaemic status with MA in the ideal BP group. Stratified analysis by a multivariate logistic regression model in the whole study population showed that people with both prehypertension and diabetes had the highest risk of MA (adjusted OR = 2.50, 95%CI 1.16-5.36; P = 0.019), compared with those with ideal BP and normoglycaemia (reference group).

Conclusions: Our findings suggest that there was a statistically significant association between diabetes and microalbuminuria only in prehypertensive subjects. In addition, our study highlights the interaction between prehypertension and diabetes as a risk factor for MA.

Keywords: Diabetes; Dysglycaemia; Logistic regression analysis; Microalbuminuria; Prediabetes; Prehypertension.