Associations of the baseline level and change in glycosylated hemoglobin A1c with incident hypertension in non-diabetic individuals: a 3-year cohort study

• Published in: Diabetology and metabolic syndrome Vol. 14; no. 1; p. 54
• Main Authors: Liu, Lijuan, Zhen, Donghu, Fu, Songbo, Sun, Weiming, Li, Hongli, Zhao, Nan, Hou, Lijie, Tang, Xulei
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 18.04.2022; BioMed Central; BMC
• Subjects: Alcohol; Analysis; Antihypertensives; Blood pressure; Body mass index; Cardiovascular disease; Cholesterol; Cohort; Cohort analysis; Consent; Diabetes; Diabetes mellitus; Ethics; Glucose; Glycosylated hemoglobin; Glycosylated hemoglobin A1c; Hemoglobin; Hospitals; Hypertension; Middle age; Questionnaires; Regression analysis; Risk factors; Surveys; Type 2 diabetes; Variables; Womens health; Hypertension; Blood pressure; Cohort; Glycosylated hemoglobin A1c
• ISSN: 1758-5996; 1758-5996
• DOI: 10.1186/s13098-022-00827-8

Diabetes mellitus increases the risk of developing hypertension. The relationship between glycosylated hemoglobin A1c (HbA1c) level and incident hypertension remains controversial. This study examined the associations of the baseline level and change in the HbA1c level over 3 years with incident hypertension in non-diabetic individuals. This community-based cohort study was conducted with 2591 individuals aged 40-75 years without hypertension or diabetes at baseline, who participated in a longitudinal (REACTION) study program. Questionnaires were administered during interviews, and anthropometric and laboratory measurements were performed at baseline (2011) and follow-up (2014-2015). Multivariate logistic regression models were applied to estimate odds ratios (ORs) and 95% confidence intervals (CIs) of incident hypertension. Over a median follow-up period of 3.08 years (interquartile range 3.00, 3.25), 384 (14.82%) subjects developed hypertension. In the fully adjusted linear regression models, change in HbA1c remained significantly associated with changes in systolic blood pressure and diastolic blood pressure [β-coefficient (95% CI), 4.421 (2.811-6.032), 1.681 (0.695-2.667)]. Logistic regression analyses showed that baseline HbA1c level was positively associated with incident hypertension in the unadjusted model; however, the association was no longer significant after further adjustment. Change in HbA1c was positively associated with the development of hypertension, both as a categorical variable stratified by tertiles [adjusted OR (95% CI) in the highest tertile was 1.690 (1.240-2.303) versus the lowest tertile)] and as a continuous variable [adjusted OR (95% CI), 1.242 (1.106-1.394)], independent of age, sex, body mass index, systolic blood pressure, fasting plasma glucose level, lipid profile, the HbA1c level at baseline and 3-year change in body mass index. A higher baseline HbA1c level was not an independent risk factor for incident hypertension, whereas the change in HbA1c was independently associated with a greater longitudinal increase in blood pressure and an increased risk of incident hypertension in non-diabetic individuals.