Long-term variation in HbA1c, SBP levels associated with CVD risk in type 2 diabetes
Diabetes patients who developed cardiovascular disease were found to be older, to have had diabetes longer, and to have significantly higher and lower levels for total cholesterol and HDL cholesterol, respectively, as well as having high coefficients of variation for HbA1c and systolic blood pressure.
Patients with type 2 diabetes who had visit-to-visit variation in HbA1c and systolic blood pressure (SBP) over the long term may be at higher risk for cardiovascular disease (CVD), a recent study said.
Researchers in Japan performed a retrospective study of patients with type 2 diabetes and no CVD history to determine whether SBP and HbA1c values that varied at different clinic visits were associated with CVD risk. Differences in the effects of each variable on CVD incidence according to mean values were also examined. Patients were included in the study if they had attended at least 4 clinic visits (at least 1 clinic visit per year) and had been followed for at least 1 year. Patients with impaired glucose tolerance or a history of CVD at the first visit or during the first year afterward were excluded. Follow-up for CVD incidence continued until June 2012. The study's primary end point was first CVD event (fatal or nonfatal acute myocardial infarction, coronary artery procedure, or ischemic or hemorrhagic stroke) requiring hospitalization. Results were published online Nov. 13 by BMJ Open Diabetes Research & Care.
Six hundred thirty-two patients were enrolled in the study, most of whom (81.5%) were men. Two hundred ninety-three (46.4%) completed follow-up; 26 (4.1%) died. The 313 remaining patients (49.5%) were mailed a questionnaire in June 2012, and 136 (21.5%) responded. The 177 patients (28.0%) who did not respond were considered to be censored at the final visit. The study's overall follow-up rate was 72% (455 of 632 patients). Median follow-up was 15.4 years. Overall, 81 patients (65 men and 16 women) developed CVD during the study period. After adjustment for age, sex, and duration of diabetes, those who had CVD were found to be older, to have had diabetes longer, and to have significantly higher and lower levels for total cholesterol and HDL cholesterol, respectively. Patients with CVD were also more likely than those without to be taking an antihypertensive agent and a lipid-lowering agent.
The coefficient of variation (CV) and the variation independent of mean (VIM) for both HbA1c and SBP were found to be significant CVD predictors, independent of mean HbA1c and mean SBP. The researchers classified patients into 4 groups according to median CV for HbA1c and SBP and median VIM for HbA1c and SBP. Hazard ratios for CVD were highest in the groups with high CV for both variables and high VIM for both variables. Patients who had a mean SBP of 135 mm Hg or higher had “drastically elevated” hazard ratios associated with CV and VIM for HbA1c versus patients with a mean SBP less than 130 mm Hg (P<0.05 for the interaction).
The authors noted that their study was retrospective and observational, that SBP data were based on a single measurement at each visit, the CVD incidence was partially self-reported, and that no data were available on nonadherence to antidiabetic and antihypertensive medications, among other limitations. However, they concluded that long-term variability by visit in HbA1c and SBP was associated with combined, additive risk for CVD incidence in patients with type 2 diabetes.
“In addition, it is suggested that a synergistic effect exists between HbA1c variability and mean SBP levels for the incidence of CVD. Even if mean SBP is maintained within the normal range, SBP variability can be a risk factor for a CVD event,” they wrote. “Our findings indicate the possibility that stabilization of variability in HbA1c and SBP as well as lowering of their mean levels can be an efficient strategy for preventing the incidence of CVD.”