- Email: [email protected]
Risk of New-Onset Diabetes Mellitus Versus Reduction in Cardiovascular Events With Statin Therapy
Risk of New-Onset Diabetes Mellitus Versus Reduction in Cardiovascular Events With Statin Therapy Kang-Ling Wang, MDa,b,c,e, Chia-Jen Liu, MDc,e,f, Tze-Fan Chao, MDc,e, Su-Jung Chen, MDc,e,f, Cheng-Hsueh Wu, MDc,e, Chi-Ming Huang, MDe, Chun-Chin Chang, MDc, Ko-Fan Wang, MDc, Tzeng-Ji Chen, MD, PhDd, Shing-Jong Lin, MD, PhDb,c,e, and Chern-En Chiang, MD, PhDa,b,c,e,* The Food and Drug Administration recently updated the safety warning concerning the association between statin therapy and new-onset diabetes mellitus (NODM). For prediabetes, little information is available for statins on cardiovascular outcome reduction and diabetogenic consequences. This study aimed to examine the risk of NODM and the reduction of cardiovascular events and death (MACE) after statin therapy in the prediabetic subjects. The medical and pharmacy claims of the prediabetic beneficiaries were retrieved from Taiwan National Health Insurance research database. The occurrence of NODM, MACE, and morbidity indexed by hospitalizations and emergency visits was ascertained by ambulatory and inpatient database. A propensity scoreematched model was constructed for statin users and nonusers. During follow-up (4.1 – 2.5 years), NODM and MACE occurred in 23.5% and 16.7%, respectively, of nonusers and 28.5% and 12.0%, respectively, of users. Statin therapy was associated with a greater risk of NODM (hazard ratio 1.20, 95% confidence interval 1.08 to 1.32) and less risk of MACE (hazard ratio 0.70, 95% confidence interval 0.61 to 0.80), both in dose-dependent fashions. The earlier and more persistent use correlated with the greater increase in risk of NODM offset by the proportionally larger reduction in MACE. Furthermore, the early persistent users had the lowest rate of hospitalizations and emergency visits. In conclusion, our findings suggested that the relation between NODM and therapeutic advantages of statins was parallel in the prediabetic population. Treatment benefits outweighed diabetic consequences in subjects receiving the earlier and more persistent treatment. Ó 2013 Elsevier Inc. All rights reserved. (Am J Cardiol 2013;-:-e-) Statins effectively reduce atherosclerotic complications, irrespective of the level of glucose metabolism disturbance.1,2 However, statin therapy is associated with an increased risk of new-onset diabetes mellitus (NODM).3 Previous analyses indicated that the risk of NODM was significantly outweighed by the reduction in vascular events or deaths among subjects with cardiovascular disease (CVD).4,5 Post hoc analyses of clinical trials identified that subjects with diabetes mellitus (DM) risk factors were at greater risk of NODM after statin therapy,6,7 particularly those who had impaired fasting glucose.8 Because clinical trials enrolled subjects who can be randomized into therapeutic allocations among prespecified
a
General Clinical Research Center, bDepartment of Medical Research and Education, cDepartment of Medicine, and dDepartment of Family Medicine, Taipei Veterans General Hospital, Taipei, Taiwan; and eSchool of Medicine and fInstitute of Public Health, National Yang-Ming University, Taipei, Taiwan. Manuscript received August 7, 2013; revised manuscript received and accepted October 29, 2013. Drs. Kang-Ling Wang and Chia-Jen Liu contributed equally to this study. This work was supported in part by grant DOH99-TD-B-111-008 from the Department of Health (Taipei, Taiwan), NSC99-2314-B-075-038-MY3 from the National Science Council (Taipei, Taiwan), and grants V102B028, V102C-002, and V100D-002-3 from Taipei Veterans General Hospital (Taipei, Taiwan). See page 6 for disclosure information. *Corresponding author: Tel: (þ886) 2-28757602; fax: (þ886) 2-28745422. E-mail address: [email protected] (C.-E. Chiang). 0002-9149/13/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2013.10.043
populations, the conducts in trials do not always reflect usual clinical practice. Meanwhile, the population studies demonstrated the ethnic differences in the risk of NODM and in the distributions of metabolic risk factors attributed to CVD.9,10 The purpose of this study was to evaluate the disadvantages and advantages of statin therapy with regards to NODM and the reduction of major adverse cardiovascular events (MACEs) in the prediabetic population and to explore whether the earlier and more persistent statin therapy attenuates subsequent adverse outcomes in prediabetic subjects. Methods The Taiwan National Health Insurance research database (NHIRD) is the representative healthcare database of the Taiwan population. The longitudinal data from 2001 to 2010 were obtained for this retrospective analysis. Prediabetic subjects as the presence of International Classification of Diseases, Ninth Revision codes for abnormal glucose in the medical claims, who had not been prescribed antidiabetic medications at the consecutive visits (30 days apart), were identified. Among them, subjects aged 45 years, who continuously received statins 7 days before having a DM diagnosis, or those naive to statins before a DM diagnosis were enrolled. Those who had endocrine disorders, who ever received systemic steroid or hypolipidemic agents other than statins, or who had followup <30 days were excluded. Co-morbidities were confirmed www.ajconline.org
2
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics before and after propensity score matching Variable
Overall Cohort
Age (yrs) Women Hypertension Coronary heart disease Coronary revascularization Myocardial infarction Ischemic stroke Heart failure Peripheral arterial disease Charlson index Ambulatory care visit, per yr Plasma glucose test, per yr
Propensity ScoreeMatched Cohort
Nonuser, n ¼ 5,767 (%)
Statin User, n ¼ 3,288 (%)
p Value
Nonuser, n ¼ 3,069 (%)
Statin User, n ¼ 3,069 (%)
p Value
67 9 2,910 (51) 4,389 (76) 2,623 (46) 30 (1) 83 (1) 271 (5) 546 (10) 143 (3) 22 14 14 1.6 4.9
66 9 1,793 (55) 2,570 (78) 1,358 (41) 93 (3) 123 (4) 195 (6) 291 (9) 88 (3) 22 18 22 1.9 5.1
<0.001 <0.001 0.026 <0.001 <0.001 <0.001 0.011 0.329 0.568 0.851 <0.001 0.013
66 9 1,689 (55) 2,400 (78) 1,256 (41) 6 (<1) 59 (2) 178 (6) 250 (8) 80 (3) 22 16 15 1.7 5.1
66 9 1,692 (55) 2,378 (78) 1,250 (41) 3 (<1) 56 (2) 165 (5) 257 (8) 84 (3) 22 16 14 1.7 3.8
0.927 0.939 0.499 0.876 0.317 0.778 0.470 0.746 0.752 0.221 0.462 0.834
Figure 1. Kaplan-Meier plots of survival free from NODM (A) and MACE (B) in the propensity scoreematched cohort.
by International Classification of Diseases, Ninth Revision codes in the medical claims. The length of statin exposure was counted cumulatively from the date of prescription. The cumulative dose was calculated and converted to the dose equivalent to the strength of simvastatin 20 mg according to the Food and Drug Administration recommendation (http://www.fda.gov/ Drugs/DrugSafety/ucm256581.htm). According to the prescription patterns from the pharmacy claims, statin users were further divided into following 3 groups: statin users after prediabetes (the initial statin prescription date on or after the first prediabetes diagnosis date), early nonpersistent users (the initial statin prescription date before the first prediabetes diagnosis date and the adherence <80%), and early persistent users (the initial statin prescription date before the first prediabetes diagnosis date and the adherence 80%). The follow-up started from the date of a confirmed prediabetes diagnosis or the start of the continuous use of statins
(for whom received statins after prediabetes). The occurrence of NODM was ascertained by the continuous prescription of antidiabetic medications 30 days. MACE was the composite of myocardial infarction, ischemic stroke, coronary revascularization, and death. The morbidity was assessed by overall hospitalizations and emergency visits. Because the use of statins was not randomly assigned to study subjects, the effect of treatment selection bias and potential confounding in this observational study were mitigated by the propensity scoreematching approach. The propensity for statin therapy was determined by using the logistic regression model regardless of outcomes, and a propensity score for the use of statins was calculated for each subject. Baseline characteristics among groups are expressed as mean SDs or counts with percentages, as applicable. Continuous variables were compared by Student t tests or one-way analysis of variance, and categorical variables were compared by chi-square tests. Kaplan-Meier plots were
Preventive Cardiology/Statins in Prediabetes
3
Table 2 Risk of new-onset diabetes mellitus (NODM) and major adverse cardiovascular event (MACE) according to the use of statins in the propensity scoreematched cohort NODM HR (95% CI) Statin nonuser (n ¼ 3,069) Atorvastatin (n ¼ 1,136) Fluvastatin (n ¼ 653) Lovastatin (n ¼ 1,352) Pravastatin (n ¼ 662) Rosuvastatin (n ¼ 167) Simvastatin (n ¼ 1,098) Single statin (n ¼ 1,764) Multiple statins (n ¼ 1,305) Maximal daily equivalent dose* <20 mg (n ¼ 1,569) 20 mg (n ¼ 1,500) Mean daily equivalent dose* <20 mg (n ¼ 2,224) 20 mg (n ¼ 845) Persistence Statin <1 yr (n ¼ 2,101) Statin 1 yr (n ¼ 968)
Reference 1.28 (1.12e1.47) 1.13 (0.96e1.33) 1.14 (1.01e1.30) 1.15 (0.98e1.35) 1.19 (0.79e1.78) 1.22 (1.07e1.39) 1.22 (1.09e1.36) 1.17 (1.03e1.33) 1.17 (1.04e1.31) 1.23 (1.09e1.39) 1.16 (1.05e1.29) 1.30 (1.12e1.51) 1.12 (1.00e1.25) 1.40 (1.22e1.61)
MACE p <0.001 0.157 0.035 0.085 0.412 0.004 0.001 0.019 <0.001† 0.009 0.001 <0.001† 0.005 0.001 <0.001† 0.048 <0.001
HR (95% CI) Reference 0.78 (0.64e0.95) 0.66 (0.52e0.85) 0.63 (0.53e0.76) 0.70 (0.56e0.88) 0.58 (0.27e1.22) 0.77 (0.64e0.92) 0.72 (0.62e0.84) 0.67 (0.55e0.80) 0.69 (0.58e0.81) 0.71 (0.60e0.85) 0.71 (0.61e0.82) 0.66 (0.53e0.84) 0.71 (0.62e0.83) 0.66 (0.53e0.82)
p 0.012 0.001 <0.001 0.002 0.149 0.004 <0.001 <0.001 <0.001† <0.001 <0.001 <0.001† <0.001 0.001 <0.001† <0.001 <0.001
* Dose equivalent to the strength of simvastatin 20 mg according to the Food and Drug Administration recommendation. † Log-rank p for trend.
Table 3 Characteristics among statin users according to the pattern of statin therapy Variable Age, yr Women Hypertension Coronary heart disease Coronary revascularization Myocardial infarction Ischemic stroke Heart failure Peripheral arterial disease Charlson index Ambulatory care visit, per yr Plasma glucose test, per yr Daily equivalent dosage, %* Statin exposure, day
Statin User After Prediabetes, n ¼ 366 (%)
Early Nonpersistent User, n ¼ 2,157 (%)
Early Persistent User, n ¼ 546 (%)
p Value
68 8 219 (60) 308 (84) 167 (46) 0 (0) 4 (1) 26 (7) 33 (9) 10 (3) 22 15 14 2.0 5.0 62 44 59 72
66 9 1,171 (54) 1,620 (75) 828 (38) 2 (<1) 34 (2) 117 (5) 173 (8) 67 (3) 22 16 14 1.6 3.6 76 44 343 386
66 9 302 (55) 450 (82) 255 (47) 1 (<1) 18 (3) 22 (4) 51 (9) 7 (1) 22 15 15 1.8 3.8 77 50 815 806
<0.001 0.142 <0.001 <0.001 0.680 0.015 0.128 0.545 0.066 0.004 0.601 0.086 <0.001 <0.001
* Percentage of dose equivalent to the strength of simvastatin 20 mg according to the Food and Drug Administration recommendation.
constructed with comparisons between curves by log-rank tests. Cox proportional hazards models were used to estimate the hazard ratio (HR) and 95% confidence interval (CI). To assess the impact of different patterns of statin therapy, multivariate Cox models adjusted for baseline characteristics were further applied. The Mantel-Haenszel test evaluated whether the linear relation between statin therapy and the composite of hospitalizations and emergency visits existed. The data were linked and processed by Microsoft SQL Server 2008, and the statistical analyses were performed with SPSS 16 (SPSS Inc., Chicago, Illinois). A 2-tailed p value <0.05 was considered statistically significant for all analyses.
Results There were 9,055 subjects enrolled in this study. Of them, 3,288 subjects had received statins. Table 1 reports baseline characteristics before and after propensity score matching. The cardiovascular and noncardiovascular background of 3,069 pairs of subjects after matching was similar as opposed to the entire cohort. During the mean follow-up of 4.1 2.5 years, NODM and MACE occurred in 23.5% and 16.7%, respectively, of nonusers and 28.5% and 12.0%, respectively, of users. Statin therapy was associated with the increased occurrence of NODM (HR 1.20, 95% CI 1.08 to 1.32, p <0.001) and the reduction in MACE (HR 0.70, 95%
4
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Kaplan-Meier plots of survival free from NODM (A) and MACE (B) stratified by the patterns of the statin therapy.
CI 0.61 to 0.80, p <0.001) compared with nonusers (Figure 1). The most commonly prescribed statins were lovastatin, followed by atorvastatin and simvastatin. The average dose was 10 mg/day for atorvastatin, 46 mg/day for fluvastatin, 21 mg/day for lovastatin, 11 mg/day for pravastatin, 17 mg/day for simvastatin, and 8 mg/day for rosuvastatin. Table 2
summaries the relation of the characteristic use of stains with NODM and MACE. The commonly prescribed statins were more pronouncedly associated with NODM development. All statins, except rosuvastatin, were consistently associated with a lower rate of MACE. Statin therapy constantly correlated with the increased occurrence of NODM and the reduction in MACE across dosages and
Preventive Cardiology/Statins in Prediabetes
5
depicts the distributions of the morbidity assessed by overall hospitalizations and emergency visits. Discussion
Figure 3. The morbidity indexed by overall hospitalizations and emergency department visits stratified by the patterns of the statin therapy. There were 30%, 12%, and 18% of statin nonusers who had 1 to 2, 3 to 4, and 5 hospitalizations or emergency visits, respectively. The proportions of subjects who had 1 to 2, 3 to 4, and 5 hospitalizations or emergency visits were 31%, 12%, and 14% in statin users after prediabetes, 28%, 11%, and 13% in early nonpersistent users, and 28%, 10%, and 12% in early persistent users, respectively.
treatment duration. Furthermore, there was a consistent trend of dose-proportional diabetogenic risk of statins (both maximal and mean daily doses and exposure duration). Table 3 lists baseline characteristics according to the patterns of statin therapy. Except for demographic differences, early persistent users had a larger proportion of taking high-potency statins, using greater daily dosages, and receiving longer exposure. Figure 2 shows the survival free from NODM and MACE according to the patterns of statin therapy. The highest rate of NODM occurred in the early persistent use (7.2% per year) followed by the early nonpersistent use (6.9% per year) and statin use after prediabetes (6.3% per year). However, the lowest rate of MACE occurred in the early persistent use (2.7% per year) followed by the early nonpersistent use (2.8% per year) and statin use after prediabetes (3.9% per year). After multivariate adjustment for baseline characteristics, the HR for NODM was 1.05 (95% CI 0.84 to 1.30), 1.19 (95% CI 1.07 to 1.33), and 1.36 (95% CI 1.16 to 1.59) for statin users after prediabetes, early nonpersistent users, and early persistent users, respectively; the HR for MACE was 0.85 (95% CI 0.65 to 1.12), 0.66 (95% CI 0.56 to 0.76), and 0.63 (95% CI 0.50 to 0.81) for statin users after prediabetes, early nonpersistent users, and early persistent users, respectively. There were 5,878 hospitalizations and 8,382 emergency visits during the follow-up. The number of hospitalizations was 1.1 2.2, 1.0 2.0, 0.8 1.7, and 0.8 1.5 for statin nonusers, statin users after prediabetes, early nonpersistent users, and early persistent users, respectively (p <0.001 for trend). The number of emergency visits was 1.6 3.8, 1.2 2.3, 1.2 3.0, and 1.0 2.3 for statin nonusers, statin users after prediabetes, early nonpersistent users, and early persistent users, respectively (p <0.001 for trend). Figure 3
Our study indicates that statin therapy was associated with an increased risk of NODM in prediabetic subjects. The clinical benefits assessed by MACE and morbidity outweighed such disadvantage. In addition, the earlier and more persistent statin therapy correlated with the greater reduction in MACE, hospitalizations, and emergency visits. Our finding is in line with the results of the meta-analysis targeting subjects with CVD that the benefits of statins were not offset by the potential risk of NODM.4 Furthermore, the assessment of morbidity indexed by hospitalizations and emergency visits empowered our investigation to capture the most potential events to ascertain the therapeutic advantage in this lower event rate cohort.11 The relation between diabetogenic consequences and statins has been reported from the analyses of randomized trials with the incremental risk of 9%.3 The risk was dose proportional and was particularly increased in those carrying unfavorable metabolic profiles.3,4,6e8 However, the heterogeneity among trials selected in the meta-analyses and statins per se has been argued.12 Our study investigating prediabetic subjects, who were more susceptible to DM conversion, indicated that the risk was further amplified by 20% compared with statin nonusers. The commonly prescribed statins were significantly associated with NODM, whereas the least-used statins had broader CIs, which suggested a statistically underpowered trend (HR 1.11, 95% CI 0.98 to 1.27, p ¼ 0.108, n ¼ 1,245 from the pooled analysis of fluvastatin, pravastatin, and rosuvastatin). The smaller statin dose used in our study was commonly observed in other investigations based on the practice survey.13 Despite the lower dose, the trend of dose-proportional diabetogenic risk of statins was consistent. The analysis from the Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) trial indicated that rosuvastatin was associated with the increased risk for DM by 28% in subjects carrying 1 DM risk factor.6 Over a median of 1.9 years, the DM incidences were 1.7% to 2.1% in the group carrying unfavorable metabolic profiles. The DM conversion rates were higher in our subjects (5.9% in the overall cohort and 6.4% in the propensity scoreematched cohort). The difference might be because the JUPITER trial enrolled subjects with median fasting glucose of 94 mg/dl, whereas subjects enrolled from the clinical practice database might have worse glucose homeostasis, which agreed with the previous analyses.14,15 Additionally, our results are similar to those of the analysis of trials involving atorvastatin, which showed the conversion rate generally exceeded 10% over nearly 5 years among subjects with fasting glucose >100 mg/dl.8 Prediabetes is a major cardiovascular risk factor beyond being serving as a pioneering signal of DM.14,16 The similar approach to prevent CVD in DM was advocated for prediabetic subjects.17e19 Results of long-term follow-up of statin trial participants indicated the benefit of early treatment extended beyond the ends of trials.20 Our analysis
The American Journal of Cardiology (www.ajconline.org)
6
showing the benefit of the earlier and more persistent statin therapies to reduce adverse outcomes echoed the legacy effect demonstrated by randomized trials. The legacy effect observed in our study was not compromised by the augmented risk of NODM. However, the weaknesses of our retrospective analysis should be acknowledged. First, the personal information including significant predictors for NODM (e.g., family history of DM, body weight, caloric intake, and exercise habits), smoking condition, and laboratory results were not available in the claim database. Second, the accuracy of the DM but not prediabetes diagnosis in NHIRD has been validated. There was possibility that subjects at the early stage of DM under lifestyle interventions miscoded with prediabetes were enrolled. Meanwhile, the variations in the prediabetic population (e.g., subjects complicated with obesity or metabolic syndrome) could skew the data independently of the statin effects. Furthermore, the use of statins was not randomly assigned, and the factors attributed to prescriptions could not be fully assessed and accounted. The propensity scoreematched analysis enabled a rigorous adjustment for factors actually measured. However, the selection bias and confounding inherent to observational studies can only be partially detected and controlled. It requests further investigations whether the risk-benefit balance of statin therapy differs in the greater risk population (e.g., metabolic syndrome). Finally, the average dose was relatively smaller than doses used in randomized trials. Nevertheless, the trend of dose-proportional diabetogenic risk was concordant with results of the published analysis.4
4.
5.
6.
7.
8.
9.
10.
11.
Acknowledgment: The study was based in part on data from the NHIRD provided by the Bureau of National Health Insurance, Department of Health and managed by the National Health Research Institutes. The interpretation and conclusions contained herein do not represent those of the aforementioned offices.
12.
13.
Disclosures The authors have no conflicts of interest to disclose. Supplementary Data
14.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2013.10.043.
15.
1. Kearney PM, Blackwell L, Collins R, Keech A, Simes J, Peto R, Armitage J, Baigent C. Efficacy of cholesterol-lowering therapy in 18, 686 people with diabetes in 14 randomised trials of statins: a metaanalysis. Lancet 2008;371:117e125. 2. Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, Peto R, Barnes EH, Keech A, Simes J, Collins R. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010;376: 1670e1681. 3. Sattar N, Preiss D, Murray HM, Welsh P, Buckley BM, de Craen AJ, Seshasai SR, McMurray JJ, Freeman DJ, Jukema JW, Macfarlane PW, Packard CJ, Stott DJ, Westendorp RG, Shepherd J, Davis BR, Pressel SL,
16. 17. 18. 19. 20.
Marchioli R, Marfisi RM, Maggioni AP, Tavazzi L, Tognoni G, Kjekshus J, Pedersen TR, Cook TJ, Gotto AM, Clearfield MB, Downs JR, Nakamura H, Ohashi Y, Mizuno K, Ray KK, Ford I. Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials. Lancet 2010;375:735e742. Preiss D, Seshasai SR, Welsh P, Murphy SA, Ho JE, Waters DD, DeMicco DA, Barter P, Cannon CP, Sabatine MS, Braunwald E, Kastelein JJ, de Lemos JA, Blazing MA, Pedersen TR, Tikkanen MJ, Sattar N, Ray KK. Risk of incident diabetes with intensive-dose compared with moderate-dose statin therapy: a meta-analysis. Jama 2011;305:2556e2564. Wang KL, Liu CJ, Chao TF, Huang CM, Wu CH, Chen SJ, Chen TJ, Lin SJ, Chiang CE. Statins, risk of diabetes, and implications on outcomes in the general population. J Am Coll Cardiol 2012;60: 1231e1238. Ridker PM, Pradhan A, MacFadyen JG, Libby P, Glynn RJ. Cardiovascular benefits and diabetes risks of statin therapy in primary prevention: an analysis from the JUPITER trial. Lancet 2012;380: 565e571. Waters DD, Ho JE, Boekholdt SM, Demicco DA, Kastelein JJ, Messig M, Breazna A, Pedersen TR. Cardiovascular event reduction versus new-onset diabetes during atorvastatin therapy: effect of baseline risk factors for diabetes. J Am Coll Cardiol 2013;61:148e152. Waters DD, Ho JE, DeMicco DA, Breazna A, Arsenault BJ, Wun CC, Kastelein JJ, Colhoun H, Barter P. Predictors of new-onset diabetes in patients treated with atorvastatin: results from 3 large randomized clinical trials. J Am Coll Cardiol 2011;57:1535e1545. Karter AJ, Schillinger D, Adams AS, Moffet HH, Liu J, Adler NE, Kanaya AM. Elevated rates of diabetes in Pacific Islanders and Asian subgroups: The Diabetes Study of Northern California (DISTANCE). Diabetes Care 2013;36:574e579. Tillin T, Hughes AD, Mayet J, Whincup P, Sattar N, Forouhi NG, McKeigue PM, Chaturvedi N. The relationship between metabolic risk factors and incident cardiovascular disease in Europeans, South Asians, and African Caribbeans: SABRE (Southall and Brent Revisited)—a prospective population-based study. J Am Coll Cardiol 2013;61: 1777e1786. Anker SD, McMurray JJ. Time to move on from ‘time-to-first’: should all events be included in the analysis of clinical trials? Eur Heart J 2012;33:2764e2765. Navarese EP, Buffon A, Andreotti F, Kozinski M, Welton N, Fabiszak T, Caputo S, Grzesk G, Kubica A, Swiatkiewicz I, Sukiennik A, Kelm M, De Servi S, Kubica J. Meta-analysis of impact of different types and doses of statins on new-onset diabetes mellitus. Am J Cardiol 2013;111:1123e1130. Kim HS, Wu Y, Lin SJ, Deerochanawong C, Zambahari R, Zhao L, Zhang Q, Yan P. Current status of cholesterol goal attainment after statin therapy among patients with hypercholesterolemia in Asian countries and region: the Return on Expenditure Achieved for Lipid Therapy in Asia (REALITY-Asia) study. Curr Med Res Opin 2008;24: 1951e1963. Grundy SM. Pre-diabetes, metabolic syndrome, and cardiovascular risk. J Am Coll Cardiol 2012;59:635e643. Nichols GA, Hillier TA, Brown JB. Progression from newly acquired impaired fasting glucose to type 2 diabetes. Diabetes Care 2007;30: 228e233. Ford ES, Zhao G, Li C. Pre-diabetes and the risk for cardiovascular disease: a systematic review of the evidence. J Am Coll Cardiol 2010;55:1310e1317. Garber AJ. Hypertension and lipid management in prediabetic states. J Clin Hypertens (Greenwich) 2011;13:270e274. Bitzur R. Diabetes and cardiovascular disease: when it comes to lipids, statins are all you need. Diabetes Care 2011;34(Suppl 2):S380eS382. DeFronzo RA, Abdul-Ghani M. Assessment and treatment of cardiovascular risk in prediabetes: impaired glucose tolerance and impaired fasting glucose. Am J Cardiol 2011;108:3Be24B. Sever PS, Chang CL, Gupta AK, Whitehouse A, Poulter NR. The Anglo-Scandinavian Cardiac Outcomes Trial: 11-year mortality follow-up of the lipid-lowering arm in the U.K. Eur Heart J 2011;32: 2525e2532.
a
General Clinical Research Center, bDepartment of Medical Research and Education, cDepartment of Medicine, and dDepartment of Family Medicine, Taipei Veterans General Hospital, Taipei, Taiwan; and eSchool of Medicine and fInstitute of Public Health, National Yang-Ming University, Taipei, Taiwan. Manuscript received August 7, 2013; revised manuscript received and accepted October 29, 2013. Drs. Kang-Ling Wang and Chia-Jen Liu contributed equally to this study. This work was supported in part by grant DOH99-TD-B-111-008 from the Department of Health (Taipei, Taiwan), NSC99-2314-B-075-038-MY3 from the National Science Council (Taipei, Taiwan), and grants V102B028, V102C-002, and V100D-002-3 from Taipei Veterans General Hospital (Taipei, Taiwan). See page 6 for disclosure information. *Corresponding author: Tel: (þ886) 2-28757602; fax: (þ886) 2-28745422. E-mail address: [email protected] (C.-E. Chiang). 0002-9149/13/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2013.10.043
populations, the conducts in trials do not always reflect usual clinical practice. Meanwhile, the population studies demonstrated the ethnic differences in the risk of NODM and in the distributions of metabolic risk factors attributed to CVD.9,10 The purpose of this study was to evaluate the disadvantages and advantages of statin therapy with regards to NODM and the reduction of major adverse cardiovascular events (MACEs) in the prediabetic population and to explore whether the earlier and more persistent statin therapy attenuates subsequent adverse outcomes in prediabetic subjects. Methods The Taiwan National Health Insurance research database (NHIRD) is the representative healthcare database of the Taiwan population. The longitudinal data from 2001 to 2010 were obtained for this retrospective analysis. Prediabetic subjects as the presence of International Classification of Diseases, Ninth Revision codes for abnormal glucose in the medical claims, who had not been prescribed antidiabetic medications at the consecutive visits (30 days apart), were identified. Among them, subjects aged 45 years, who continuously received statins 7 days before having a DM diagnosis, or those naive to statins before a DM diagnosis were enrolled. Those who had endocrine disorders, who ever received systemic steroid or hypolipidemic agents other than statins, or who had followup <30 days were excluded. Co-morbidities were confirmed www.ajconline.org
2
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics before and after propensity score matching Variable
Overall Cohort
Age (yrs) Women Hypertension Coronary heart disease Coronary revascularization Myocardial infarction Ischemic stroke Heart failure Peripheral arterial disease Charlson index Ambulatory care visit, per yr Plasma glucose test, per yr
Propensity ScoreeMatched Cohort
Nonuser, n ¼ 5,767 (%)
Statin User, n ¼ 3,288 (%)
p Value
Nonuser, n ¼ 3,069 (%)
Statin User, n ¼ 3,069 (%)
p Value
67 9 2,910 (51) 4,389 (76) 2,623 (46) 30 (1) 83 (1) 271 (5) 546 (10) 143 (3) 22 14 14 1.6 4.9
66 9 1,793 (55) 2,570 (78) 1,358 (41) 93 (3) 123 (4) 195 (6) 291 (9) 88 (3) 22 18 22 1.9 5.1
<0.001 <0.001 0.026 <0.001 <0.001 <0.001 0.011 0.329 0.568 0.851 <0.001 0.013
66 9 1,689 (55) 2,400 (78) 1,256 (41) 6 (<1) 59 (2) 178 (6) 250 (8) 80 (3) 22 16 15 1.7 5.1
66 9 1,692 (55) 2,378 (78) 1,250 (41) 3 (<1) 56 (2) 165 (5) 257 (8) 84 (3) 22 16 14 1.7 3.8
0.927 0.939 0.499 0.876 0.317 0.778 0.470 0.746 0.752 0.221 0.462 0.834
Figure 1. Kaplan-Meier plots of survival free from NODM (A) and MACE (B) in the propensity scoreematched cohort.
by International Classification of Diseases, Ninth Revision codes in the medical claims. The length of statin exposure was counted cumulatively from the date of prescription. The cumulative dose was calculated and converted to the dose equivalent to the strength of simvastatin 20 mg according to the Food and Drug Administration recommendation (http://www.fda.gov/ Drugs/DrugSafety/ucm256581.htm). According to the prescription patterns from the pharmacy claims, statin users were further divided into following 3 groups: statin users after prediabetes (the initial statin prescription date on or after the first prediabetes diagnosis date), early nonpersistent users (the initial statin prescription date before the first prediabetes diagnosis date and the adherence <80%), and early persistent users (the initial statin prescription date before the first prediabetes diagnosis date and the adherence 80%). The follow-up started from the date of a confirmed prediabetes diagnosis or the start of the continuous use of statins
(for whom received statins after prediabetes). The occurrence of NODM was ascertained by the continuous prescription of antidiabetic medications 30 days. MACE was the composite of myocardial infarction, ischemic stroke, coronary revascularization, and death. The morbidity was assessed by overall hospitalizations and emergency visits. Because the use of statins was not randomly assigned to study subjects, the effect of treatment selection bias and potential confounding in this observational study were mitigated by the propensity scoreematching approach. The propensity for statin therapy was determined by using the logistic regression model regardless of outcomes, and a propensity score for the use of statins was calculated for each subject. Baseline characteristics among groups are expressed as mean SDs or counts with percentages, as applicable. Continuous variables were compared by Student t tests or one-way analysis of variance, and categorical variables were compared by chi-square tests. Kaplan-Meier plots were
Preventive Cardiology/Statins in Prediabetes
3
Table 2 Risk of new-onset diabetes mellitus (NODM) and major adverse cardiovascular event (MACE) according to the use of statins in the propensity scoreematched cohort NODM HR (95% CI) Statin nonuser (n ¼ 3,069) Atorvastatin (n ¼ 1,136) Fluvastatin (n ¼ 653) Lovastatin (n ¼ 1,352) Pravastatin (n ¼ 662) Rosuvastatin (n ¼ 167) Simvastatin (n ¼ 1,098) Single statin (n ¼ 1,764) Multiple statins (n ¼ 1,305) Maximal daily equivalent dose* <20 mg (n ¼ 1,569) 20 mg (n ¼ 1,500) Mean daily equivalent dose* <20 mg (n ¼ 2,224) 20 mg (n ¼ 845) Persistence Statin <1 yr (n ¼ 2,101) Statin 1 yr (n ¼ 968)
Reference 1.28 (1.12e1.47) 1.13 (0.96e1.33) 1.14 (1.01e1.30) 1.15 (0.98e1.35) 1.19 (0.79e1.78) 1.22 (1.07e1.39) 1.22 (1.09e1.36) 1.17 (1.03e1.33) 1.17 (1.04e1.31) 1.23 (1.09e1.39) 1.16 (1.05e1.29) 1.30 (1.12e1.51) 1.12 (1.00e1.25) 1.40 (1.22e1.61)
MACE p <0.001 0.157 0.035 0.085 0.412 0.004 0.001 0.019 <0.001† 0.009 0.001 <0.001† 0.005 0.001 <0.001† 0.048 <0.001
HR (95% CI) Reference 0.78 (0.64e0.95) 0.66 (0.52e0.85) 0.63 (0.53e0.76) 0.70 (0.56e0.88) 0.58 (0.27e1.22) 0.77 (0.64e0.92) 0.72 (0.62e0.84) 0.67 (0.55e0.80) 0.69 (0.58e0.81) 0.71 (0.60e0.85) 0.71 (0.61e0.82) 0.66 (0.53e0.84) 0.71 (0.62e0.83) 0.66 (0.53e0.82)
p 0.012 0.001 <0.001 0.002 0.149 0.004 <0.001 <0.001 <0.001† <0.001 <0.001 <0.001† <0.001 0.001 <0.001† <0.001 <0.001
* Dose equivalent to the strength of simvastatin 20 mg according to the Food and Drug Administration recommendation. † Log-rank p for trend.
Table 3 Characteristics among statin users according to the pattern of statin therapy Variable Age, yr Women Hypertension Coronary heart disease Coronary revascularization Myocardial infarction Ischemic stroke Heart failure Peripheral arterial disease Charlson index Ambulatory care visit, per yr Plasma glucose test, per yr Daily equivalent dosage, %* Statin exposure, day
Statin User After Prediabetes, n ¼ 366 (%)
Early Nonpersistent User, n ¼ 2,157 (%)
Early Persistent User, n ¼ 546 (%)
p Value
68 8 219 (60) 308 (84) 167 (46) 0 (0) 4 (1) 26 (7) 33 (9) 10 (3) 22 15 14 2.0 5.0 62 44 59 72
66 9 1,171 (54) 1,620 (75) 828 (38) 2 (<1) 34 (2) 117 (5) 173 (8) 67 (3) 22 16 14 1.6 3.6 76 44 343 386
66 9 302 (55) 450 (82) 255 (47) 1 (<1) 18 (3) 22 (4) 51 (9) 7 (1) 22 15 15 1.8 3.8 77 50 815 806
<0.001 0.142 <0.001 <0.001 0.680 0.015 0.128 0.545 0.066 0.004 0.601 0.086 <0.001 <0.001
* Percentage of dose equivalent to the strength of simvastatin 20 mg according to the Food and Drug Administration recommendation.
constructed with comparisons between curves by log-rank tests. Cox proportional hazards models were used to estimate the hazard ratio (HR) and 95% confidence interval (CI). To assess the impact of different patterns of statin therapy, multivariate Cox models adjusted for baseline characteristics were further applied. The Mantel-Haenszel test evaluated whether the linear relation between statin therapy and the composite of hospitalizations and emergency visits existed. The data were linked and processed by Microsoft SQL Server 2008, and the statistical analyses were performed with SPSS 16 (SPSS Inc., Chicago, Illinois). A 2-tailed p value <0.05 was considered statistically significant for all analyses.
Results There were 9,055 subjects enrolled in this study. Of them, 3,288 subjects had received statins. Table 1 reports baseline characteristics before and after propensity score matching. The cardiovascular and noncardiovascular background of 3,069 pairs of subjects after matching was similar as opposed to the entire cohort. During the mean follow-up of 4.1 2.5 years, NODM and MACE occurred in 23.5% and 16.7%, respectively, of nonusers and 28.5% and 12.0%, respectively, of users. Statin therapy was associated with the increased occurrence of NODM (HR 1.20, 95% CI 1.08 to 1.32, p <0.001) and the reduction in MACE (HR 0.70, 95%
4
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Kaplan-Meier plots of survival free from NODM (A) and MACE (B) stratified by the patterns of the statin therapy.
CI 0.61 to 0.80, p <0.001) compared with nonusers (Figure 1). The most commonly prescribed statins were lovastatin, followed by atorvastatin and simvastatin. The average dose was 10 mg/day for atorvastatin, 46 mg/day for fluvastatin, 21 mg/day for lovastatin, 11 mg/day for pravastatin, 17 mg/day for simvastatin, and 8 mg/day for rosuvastatin. Table 2
summaries the relation of the characteristic use of stains with NODM and MACE. The commonly prescribed statins were more pronouncedly associated with NODM development. All statins, except rosuvastatin, were consistently associated with a lower rate of MACE. Statin therapy constantly correlated with the increased occurrence of NODM and the reduction in MACE across dosages and
Preventive Cardiology/Statins in Prediabetes
5
depicts the distributions of the morbidity assessed by overall hospitalizations and emergency visits. Discussion
Figure 3. The morbidity indexed by overall hospitalizations and emergency department visits stratified by the patterns of the statin therapy. There were 30%, 12%, and 18% of statin nonusers who had 1 to 2, 3 to 4, and 5 hospitalizations or emergency visits, respectively. The proportions of subjects who had 1 to 2, 3 to 4, and 5 hospitalizations or emergency visits were 31%, 12%, and 14% in statin users after prediabetes, 28%, 11%, and 13% in early nonpersistent users, and 28%, 10%, and 12% in early persistent users, respectively.
treatment duration. Furthermore, there was a consistent trend of dose-proportional diabetogenic risk of statins (both maximal and mean daily doses and exposure duration). Table 3 lists baseline characteristics according to the patterns of statin therapy. Except for demographic differences, early persistent users had a larger proportion of taking high-potency statins, using greater daily dosages, and receiving longer exposure. Figure 2 shows the survival free from NODM and MACE according to the patterns of statin therapy. The highest rate of NODM occurred in the early persistent use (7.2% per year) followed by the early nonpersistent use (6.9% per year) and statin use after prediabetes (6.3% per year). However, the lowest rate of MACE occurred in the early persistent use (2.7% per year) followed by the early nonpersistent use (2.8% per year) and statin use after prediabetes (3.9% per year). After multivariate adjustment for baseline characteristics, the HR for NODM was 1.05 (95% CI 0.84 to 1.30), 1.19 (95% CI 1.07 to 1.33), and 1.36 (95% CI 1.16 to 1.59) for statin users after prediabetes, early nonpersistent users, and early persistent users, respectively; the HR for MACE was 0.85 (95% CI 0.65 to 1.12), 0.66 (95% CI 0.56 to 0.76), and 0.63 (95% CI 0.50 to 0.81) for statin users after prediabetes, early nonpersistent users, and early persistent users, respectively. There were 5,878 hospitalizations and 8,382 emergency visits during the follow-up. The number of hospitalizations was 1.1 2.2, 1.0 2.0, 0.8 1.7, and 0.8 1.5 for statin nonusers, statin users after prediabetes, early nonpersistent users, and early persistent users, respectively (p <0.001 for trend). The number of emergency visits was 1.6 3.8, 1.2 2.3, 1.2 3.0, and 1.0 2.3 for statin nonusers, statin users after prediabetes, early nonpersistent users, and early persistent users, respectively (p <0.001 for trend). Figure 3
Our study indicates that statin therapy was associated with an increased risk of NODM in prediabetic subjects. The clinical benefits assessed by MACE and morbidity outweighed such disadvantage. In addition, the earlier and more persistent statin therapy correlated with the greater reduction in MACE, hospitalizations, and emergency visits. Our finding is in line with the results of the meta-analysis targeting subjects with CVD that the benefits of statins were not offset by the potential risk of NODM.4 Furthermore, the assessment of morbidity indexed by hospitalizations and emergency visits empowered our investigation to capture the most potential events to ascertain the therapeutic advantage in this lower event rate cohort.11 The relation between diabetogenic consequences and statins has been reported from the analyses of randomized trials with the incremental risk of 9%.3 The risk was dose proportional and was particularly increased in those carrying unfavorable metabolic profiles.3,4,6e8 However, the heterogeneity among trials selected in the meta-analyses and statins per se has been argued.12 Our study investigating prediabetic subjects, who were more susceptible to DM conversion, indicated that the risk was further amplified by 20% compared with statin nonusers. The commonly prescribed statins were significantly associated with NODM, whereas the least-used statins had broader CIs, which suggested a statistically underpowered trend (HR 1.11, 95% CI 0.98 to 1.27, p ¼ 0.108, n ¼ 1,245 from the pooled analysis of fluvastatin, pravastatin, and rosuvastatin). The smaller statin dose used in our study was commonly observed in other investigations based on the practice survey.13 Despite the lower dose, the trend of dose-proportional diabetogenic risk of statins was consistent. The analysis from the Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) trial indicated that rosuvastatin was associated with the increased risk for DM by 28% in subjects carrying 1 DM risk factor.6 Over a median of 1.9 years, the DM incidences were 1.7% to 2.1% in the group carrying unfavorable metabolic profiles. The DM conversion rates were higher in our subjects (5.9% in the overall cohort and 6.4% in the propensity scoreematched cohort). The difference might be because the JUPITER trial enrolled subjects with median fasting glucose of 94 mg/dl, whereas subjects enrolled from the clinical practice database might have worse glucose homeostasis, which agreed with the previous analyses.14,15 Additionally, our results are similar to those of the analysis of trials involving atorvastatin, which showed the conversion rate generally exceeded 10% over nearly 5 years among subjects with fasting glucose >100 mg/dl.8 Prediabetes is a major cardiovascular risk factor beyond being serving as a pioneering signal of DM.14,16 The similar approach to prevent CVD in DM was advocated for prediabetic subjects.17e19 Results of long-term follow-up of statin trial participants indicated the benefit of early treatment extended beyond the ends of trials.20 Our analysis
The American Journal of Cardiology (www.ajconline.org)
6
showing the benefit of the earlier and more persistent statin therapies to reduce adverse outcomes echoed the legacy effect demonstrated by randomized trials. The legacy effect observed in our study was not compromised by the augmented risk of NODM. However, the weaknesses of our retrospective analysis should be acknowledged. First, the personal information including significant predictors for NODM (e.g., family history of DM, body weight, caloric intake, and exercise habits), smoking condition, and laboratory results were not available in the claim database. Second, the accuracy of the DM but not prediabetes diagnosis in NHIRD has been validated. There was possibility that subjects at the early stage of DM under lifestyle interventions miscoded with prediabetes were enrolled. Meanwhile, the variations in the prediabetic population (e.g., subjects complicated with obesity or metabolic syndrome) could skew the data independently of the statin effects. Furthermore, the use of statins was not randomly assigned, and the factors attributed to prescriptions could not be fully assessed and accounted. The propensity scoreematched analysis enabled a rigorous adjustment for factors actually measured. However, the selection bias and confounding inherent to observational studies can only be partially detected and controlled. It requests further investigations whether the risk-benefit balance of statin therapy differs in the greater risk population (e.g., metabolic syndrome). Finally, the average dose was relatively smaller than doses used in randomized trials. Nevertheless, the trend of dose-proportional diabetogenic risk was concordant with results of the published analysis.4
4.
5.
6.
7.
8.
9.
10.
11.
Acknowledgment: The study was based in part on data from the NHIRD provided by the Bureau of National Health Insurance, Department of Health and managed by the National Health Research Institutes. The interpretation and conclusions contained herein do not represent those of the aforementioned offices.
12.
13.
Disclosures The authors have no conflicts of interest to disclose. Supplementary Data
14.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2013.10.043.
15.
1. Kearney PM, Blackwell L, Collins R, Keech A, Simes J, Peto R, Armitage J, Baigent C. Efficacy of cholesterol-lowering therapy in 18, 686 people with diabetes in 14 randomised trials of statins: a metaanalysis. Lancet 2008;371:117e125. 2. Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, Peto R, Barnes EH, Keech A, Simes J, Collins R. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010;376: 1670e1681. 3. Sattar N, Preiss D, Murray HM, Welsh P, Buckley BM, de Craen AJ, Seshasai SR, McMurray JJ, Freeman DJ, Jukema JW, Macfarlane PW, Packard CJ, Stott DJ, Westendorp RG, Shepherd J, Davis BR, Pressel SL,
16. 17. 18. 19. 20.
Marchioli R, Marfisi RM, Maggioni AP, Tavazzi L, Tognoni G, Kjekshus J, Pedersen TR, Cook TJ, Gotto AM, Clearfield MB, Downs JR, Nakamura H, Ohashi Y, Mizuno K, Ray KK, Ford I. Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials. Lancet 2010;375:735e742. Preiss D, Seshasai SR, Welsh P, Murphy SA, Ho JE, Waters DD, DeMicco DA, Barter P, Cannon CP, Sabatine MS, Braunwald E, Kastelein JJ, de Lemos JA, Blazing MA, Pedersen TR, Tikkanen MJ, Sattar N, Ray KK. Risk of incident diabetes with intensive-dose compared with moderate-dose statin therapy: a meta-analysis. Jama 2011;305:2556e2564. Wang KL, Liu CJ, Chao TF, Huang CM, Wu CH, Chen SJ, Chen TJ, Lin SJ, Chiang CE. Statins, risk of diabetes, and implications on outcomes in the general population. J Am Coll Cardiol 2012;60: 1231e1238. Ridker PM, Pradhan A, MacFadyen JG, Libby P, Glynn RJ. Cardiovascular benefits and diabetes risks of statin therapy in primary prevention: an analysis from the JUPITER trial. Lancet 2012;380: 565e571. Waters DD, Ho JE, Boekholdt SM, Demicco DA, Kastelein JJ, Messig M, Breazna A, Pedersen TR. Cardiovascular event reduction versus new-onset diabetes during atorvastatin therapy: effect of baseline risk factors for diabetes. J Am Coll Cardiol 2013;61:148e152. Waters DD, Ho JE, DeMicco DA, Breazna A, Arsenault BJ, Wun CC, Kastelein JJ, Colhoun H, Barter P. Predictors of new-onset diabetes in patients treated with atorvastatin: results from 3 large randomized clinical trials. J Am Coll Cardiol 2011;57:1535e1545. Karter AJ, Schillinger D, Adams AS, Moffet HH, Liu J, Adler NE, Kanaya AM. Elevated rates of diabetes in Pacific Islanders and Asian subgroups: The Diabetes Study of Northern California (DISTANCE). Diabetes Care 2013;36:574e579. Tillin T, Hughes AD, Mayet J, Whincup P, Sattar N, Forouhi NG, McKeigue PM, Chaturvedi N. The relationship between metabolic risk factors and incident cardiovascular disease in Europeans, South Asians, and African Caribbeans: SABRE (Southall and Brent Revisited)—a prospective population-based study. J Am Coll Cardiol 2013;61: 1777e1786. Anker SD, McMurray JJ. Time to move on from ‘time-to-first’: should all events be included in the analysis of clinical trials? Eur Heart J 2012;33:2764e2765. Navarese EP, Buffon A, Andreotti F, Kozinski M, Welton N, Fabiszak T, Caputo S, Grzesk G, Kubica A, Swiatkiewicz I, Sukiennik A, Kelm M, De Servi S, Kubica J. Meta-analysis of impact of different types and doses of statins on new-onset diabetes mellitus. Am J Cardiol 2013;111:1123e1130. Kim HS, Wu Y, Lin SJ, Deerochanawong C, Zambahari R, Zhao L, Zhang Q, Yan P. Current status of cholesterol goal attainment after statin therapy among patients with hypercholesterolemia in Asian countries and region: the Return on Expenditure Achieved for Lipid Therapy in Asia (REALITY-Asia) study. Curr Med Res Opin 2008;24: 1951e1963. Grundy SM. Pre-diabetes, metabolic syndrome, and cardiovascular risk. J Am Coll Cardiol 2012;59:635e643. Nichols GA, Hillier TA, Brown JB. Progression from newly acquired impaired fasting glucose to type 2 diabetes. Diabetes Care 2007;30: 228e233. Ford ES, Zhao G, Li C. Pre-diabetes and the risk for cardiovascular disease: a systematic review of the evidence. J Am Coll Cardiol 2010;55:1310e1317. Garber AJ. Hypertension and lipid management in prediabetic states. J Clin Hypertens (Greenwich) 2011;13:270e274. Bitzur R. Diabetes and cardiovascular disease: when it comes to lipids, statins are all you need. Diabetes Care 2011;34(Suppl 2):S380eS382. DeFronzo RA, Abdul-Ghani M. Assessment and treatment of cardiovascular risk in prediabetes: impaired glucose tolerance and impaired fasting glucose. Am J Cardiol 2011;108:3Be24B. Sever PS, Chang CL, Gupta AK, Whitehouse A, Poulter NR. The Anglo-Scandinavian Cardiac Outcomes Trial: 11-year mortality follow-up of the lipid-lowering arm in the U.K. Eur Heart J 2011;32: 2525e2532.