gms | German Medical Science

Background: Clinical guideline recommendations for statin treatment of patients with diabetes for cardiovascular disease (CVD) prevention have varied over time. Most current guidelines agree on recommending statin use for patients with diabetes aged 40 years or older, in particular those with a high CVD risk. Despite clear benefits of statin use, adverse effects on glucose metabolism are of concern due to an increased risk of diabetes found in clinical trials. Population-based studies with consideration of statin use duration and diabetes status are sparse.

Study questions:

1.

Is statin use associated with an increase of glycated hemoglobin A1C (HbA1c) among persons without diagnosed diabetes

2.

If yes, is the increase associated with duration of statin use?

3.

Is a similar relationship between HbA1c and statin use seen among persons with diagnosed diabetes who do not currently use antidiabetic medication?

Method: A cross-sectional analysis was carried out based on data from a large population-based sample of adults in Germany 18-79 years. Data collection comprised:

1.

sociodemographic factors and health-related behaviors using self-administered questionnaires;

2.

personal history of physician-diagnosed chronic diseases including diabetes, myocardial infarction or other coronary heart disease, stroke and a detailed review of all medications (including drug name, use duration) used in the past 7 days based on computer-assisted personal interviews;

3.

highly standardized anthropometric, blood pressure and laboratory measures including HbA1c, serum high-density lipoprotein (HDL) cholesterol. Diabetes was defined based on a history of physician-diagnosed diabetes or current use of antidiabetic drugs. The population-based sample of adults comprised 7,115 persons. After excluding persons younger than 40 years of age (n=1,912), those with incomplete information on diabetes status (n=30), statin use duration (n=21) and HbA1c values (n=44), women with gestational diabetes (n=22), and persons with diagnosed diabetes on current antidiabetic treatment (n=387), the present analysis included 4,699 persons (4,570 persons without diabetes and 129 persons with diagnosed diabetes but without antidiabetic medication). Duration of statin use was categorized as short term (< 1 year), intermediate term (1-3 years), and long term (≥ 3 years). Linear regression was used to examine the association of HbA1c with statin use, overall and in relation to years of statin use, adjusting for potential confounders (age, sex, region of residence, community size, education, body mass index, smoking, sports activity, alcohol consumption, aspirin use, HDL-cholesterol, diastolic blood pressure, history of myocardial infarction or other coronary heart disease, and stroke). Results were weighted to German population statistics 2010.

Results: A total of 550 or 10.4% of persons without diagnosed diabetes were current statin users as compared to 38 or 25.0% of those with diagnosed diabetes but not currently using antidiabetic treatment. The majority of statin users among both persons without and with diagnosed diabetes were long-term users (54.6% vs. 74.3%). After adjusting for confounders, statin use was associated with an increased HbA1c level both among persons without diagnosed diabetes (HbA1c +0.10, 95% confidence interval, 0.04-0.16, p < .001) and persons with diagnosed diabetes (+0.54, 0.17-0.91, p=.004). Among persons without diagnosed diabetes, the increase of HbA1c was most evident for short-term use (+0.15, 0.02-0.28, p=.026), and intermediate-term use (+0.12, 0.02-0.23, p=.017), but not found for long-term use (+0.07, -0.007-0.15, p=.075). Among persons with diagnosed diabetes, the increase of HbA1c was found for long-term use (+0.55, 0.18-0.93, p=.004), but not found at a significant level for intermediate or short-term use (+0.52, -0.01–1.04, p=.055).

Discussion: Statin use was associated with an increase of HbA1c among persons without antidiabetic medications. The increase of HbA1c seems to be attenuated along with the use duration among persons without diabetes but not among those with diabetes who were not taking antidiabetic medications. A major strength of the present study is that we were able to compare persons with and without diagnosed diabetes from the same well-defined population. A limitation of the population-based design is that persons 80 years of age and older were not included and that ill persons are likely to be missed. The group of patients with diabetes not currently using antidiabetic medication was small and may be subject to selection bias.Practical implications: Clinicians need to monitor the glycemic status of all patients under statin therapy. It should be further studied if and how the increase of HbA1c following statin use contributes to the development of diabetes, and if statin use exerts an effect on glucose metabolism among persons with diagnosed diabetes who are taking antidiabetic medication.