|
KMID : 9000220160010070047
|
|
NECA Research Report
2016 Volume.1 No. 2 p.1 ~ p.47
|
|
|
|
|
Kim Yun-Jung
Cho Song-Hee
|
|
|
Abstract
|
|
|
? Background and purpose
Statins are primary and secondary preventive drugs for cardiovascular diseases. It is the most widely used agent along with aspirin and its use is expected to increase due to recent updates of guideline on the treatment of hyperlipidemia. Despite the fact that long term safety evaluation of Statin is crucial, only few studies are available with large patient cohorts in the clinical setting. Moreover, studies investigating the association between long term usage of Statins and the onset of type II diabetes as well as subsequent secondary cardiovascular complications are very limited. Therefore, this study was conducted to understand the relationship between the Statin use and the occurrence of diabetes and its complications as well as to compare the incident rate of diabetes and its complications in terms of duration of use, types, and doses of Statin based on the National Health Insurance claims data and the National Periodic Health Examinations data.
? Methods
A retrospective cohort study was performed using claims data, eligibility data, and the National Periodic Health Examinations data from the National Health Insurance Service, and cause of death data from the Statistics Korea to evaluate the risk of diabetes and cardiovascular complications related to Statin use in the Statin eligible group.
Subjects with more than 240mg/dl in total cholesterol and no history of medical use related to cardiovascular diseases were eligible among persons aged over 40 who received a health examination between 1/1/2005 and 12/31/2012.
Subjects who had cancer, diabetes, or cardiovascular diseases, or who were prescribed for lipid lowering agent in last 3 years before enrolling the study were excluded. Additionally, subjects who died within a year, who took Statin one time, or who have diabetes or cardiovascular diseases within 6 months after enrolling the study were excluded.
Statin group was defined as who were prescribed for Statin more than two times within 6 months during the study period. Index date was set to be the initial date of the first treatment in prescriptions with more than two treatments. Non-Statin group was defined as who were never prescribed for Statin during the course of study and the enrollment date was determined by when the total cholesterol level was reported as more than 240mg/dl for the first time.
Primary outcome was the incidence of type II diabetes (cases with disease code and prescribed with oral hypoglycemic agents or insulin) and secondary outcome was the incidence of major cardiovascular diseases among diabetes complications (death due to cardiovascular diseases, or hospitalization due to myocardial infarction or stroke).
Risks of diabetes and cardiovascular diseases based on Statin use was presented using the Cox's proportional hazards model after matching to the propensity score taking into account sex, age, fasting blood sugar, body mass index, smoking, exercise, hypertension, Charlson Comorbidity Index (CCI), and concomitant cardiovascular medications. Also, subgroup analyses were conducted based on the defined daily dose (DDD) of Statin, (DDD is defined as the average dose that an adult with 70kg needs to take for one day for the major indication with each principal component of the agent), duration of Statin use, and types of Statin and the metabolic risk in order to understand the degree of risk compared to the non-Statin group.
? Results
Among the 2,162,119 of enrolled patients, the number of subjects in Statin group and non-Statin group was 638,625 (29.5%) and 1,523,494 (70.5%), respectively. The number of subjects included in the final analysis after propensity score matching was 518,491 each in Statin and non-Statin group. Before the matching, the mean age of Statin group was 56 year, higher than the non-Statin group (52 year), and the Statin group had higher rate of taking concomitant medications for hypertension and cardiovascular diseases. The differences in general characteristics were adjusted by the propensity score matching.
The incident rates of diabetes in Statin group and non-Statin group were
13.36 and 6.85 out of 1,000 person-year, respectively, indicating the Statin group had higher risk of diabetes. Analysis with the Cox's proportional hazards model after propensity score matching showed that the degree of incident risk for diabetes in Statin group compared to non-Statin group was 1.88 (95% CI: 1.85-1.93). Risk for diabetes increased in accordance with higher dose (HR: 2.52, 95% CI: 2.47-2.57) and the highest risk was observed in a group with Statin use longer than 2 years (HR: 2.62, 95% CI: 2.56-2.67). When stratified with diabetes risk factors (defects in fasting blood sugar, BMI, exercise, and number of diabetes risk factors), the Statin group had consistently higher incidence of diabetes regardless of the factors.
On the other hand, the Statin group appeared to have lower incident risk of cardiovascular diseases compared to the non-Statin group (HR: 0.67, 95% CI: 0.61-0.74). Subjects with high dose (HR: 0.77, 95% CI: 0.68-0.87) and subjects
with longer than 2 years of administration (HR: 0.72, 95% CI: 0.62-0.82) presented significantly low degree of incident risk. Based on the subject characteristics regarding diabetes risk, the incidence of cardiovascular diseases was lower in the Statin group than the non-Statin group.
? Conclusions and Recommendations
Most studies so far on Statins and its role in diabetes onset have been conducted as meta-analysis of randomized clinical trials or observational studies with limited subject number. Currently, Statins are widely used as major primary and secondary preventive agents in cardiovascular diseases. Nonetheless, with our best knowledge, no study has been undertaken thoughtfully on Statins and its relationship with diabetes onset based on a large size of patient cohorts with various characteristics in daily clinical settings. In this study, by utilizing public resources, a large population was analyzed on the association between Statin use and incident risk of diabetes as well as cardiovascular diseases. In the patients recommended to use Statins, use of Statins, higher DDD, and longer period of administration were related to be significantly higher risk for type II diabetes compared to the non-Statin group, whereas risk for cardiovascular diseases was lower in the Statin group compared to the non-Statin group.
According to the recently published guideline on Statins in the United States (2013 ACC/AHA guideline), the number of patients with the indications of Statin use has been increased by more than 1.5 fold compared to the past guideline published over 10 years ago (ATP-III guideline), and the frequency of Stain use is expected to increase further. Moreover, primary prevention of cardiovascular diseases becomes crucial in Korea as the risk factors for those diseases increase due to changes in dietary life and rapid increase of obese people. Therefore, in today’s medical environment where Statin use is recommended for primary prevention of cardiovascular diseases, this study will provide fundamental information to reevaluate the benefit-risk ratio of Statin use from a large scale cohort with Statin use. In addition, this study was conducted to investigate the risk of diabetes onset in Korean population in contrast to previous clinical studies targeting Western population, suggesting this study can be highly informative when applying health policies or
establishing a Korean guideline for Statin use.
As this study is a retrospective cohort study using public data with limited parameter, careful interpretation of the results with sufficient understanding of limitations is required, and further confirmation of the results is needed in the future through conducting a large scale prospective study.
|
|
|
KEYWORD
|
|
|
Statin, diabetes, diabetes complications, cardiovascular diseases
|
|
|
FullTexts / Linksout information
|
|
|
|
|
Listed journal information
|
|
|
|
|