Increased Risk of Diabetes Observed Among Statin-Treated Patients

February 17, 2010

February 17, 2010 (Glasgow, Scotland) — New data from a large meta-analysis of major statin trials suggests the low-density lipoprotein (LDL)-cholesterol–lowering drugs slightly increase the risk of developing diabetes mellitus [1]. Investigators stress, however, that clinical practice should remain unchanged in patients with moderate or high cardiovascular risk, given the low absolute risk of developing diabetes, particularly when compared with the benefit of statins.

"We found that there was indeed a risk of diabetes, about 9%, but it isn't a worrying increase as had been suggested by other studies," said co–lead investigator Dr David Preiss (University of Glasgow, Scotland). "Then again, it wasn't a completely flat result. We did see something. Our message would be that people on statins should be those we think are at moderate to high cardiovascular risk in the future. If you look at that group of patients, then what we really want to see come out of the study is a reassuring message, because there is little question that the protective effects in reducing heart attacks, strokes, and so on heavily outweigh this risk of developing diabetes."

Dr Steven Nissen (Cleveland Clinic, OH), who was not involved in the meta-analysis, praised the researchers, calling their interpretation of the data "responsible." Most important, he agreed with their conclusions, stating that the benefits of statins exceed the risk of diabetes and that physicians should not alter clinical practice based on these findings.

"In all of these trials, the population with diabetes or the population with new-onset diabetes had the same benefit in terms of reduction in morbidity and mortality as did people who received statins who were not insulin resistant or who had prediabetes," Nissen told heartwire . "Whatever this effect is, it doesn't lessen the favorable effect of statins on clinical outcomes. I don't think people should hesitate to give prediabetic patients statins because they might develop diabetes a few weeks or a few months later and deny them all the other benefits of these drugs."

The results of the study, which was also led by Dr Naveed Sattar (University of Glasgow), are published online February 17, 2010 in the Lancet.

Diabetes Concern Raised in JUPITER

Speaking with heartwire , Preiss explained that previous statin trials have reported conflicting results with regard to the risk of developing diabetes in patients taking the lipid-lowering medication. In the West of Scotland Coronary Prevention Study (WOSCOPS), for example, there was a reduction in diabetes in patients treated with pravastatin, whereas in the large, highly publicized Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) study, previously reported by heartwire , researchers showed that significantly more patients in the rosuvastatin (Crestor, AstraZeneca) arm developed new diabetes after a median treatment of 1.9 years. It was this finding that raised some red flags, he noted.

To address the possibility of an increased risk of diabetes with statin therapy, Preiss and colleagues performed a meta-analysis of 13 statin studies. The group included statin studies with more than 1000 patients, those that had identical follow-up in both treatment arms, and those with more than one-year duration in follow-up. In total, 91 140 participants were included in the meta-analysis. During a mean follow-up of four years, 4278 individuals developed diabetes mellitus, including 2226 patients treated with statins and 2052 assigned to a control therapy.

Treatment with statins was associated with a statistically significant 9% increase in the risk of incident diabetes, with little heterogeneity between trials. As with the overall results, the increased risk was maintained when the analysis was restricted to placebo-controlled trials only and when JUPITER and the Management of Elevated Cholesterol in the Primary Prevention Group of Adult Japanese (MEGA) study, a Japanese patient population, were each excluded from the analysis.

Association Between Statins and Development of Diabetes

Statin Odds ratio (95% CI)
Overall (n=91 140) 1.09 (1.02–1.17)
Atorvastatin only (n=7773) 1.14 (0.89–1.46)
Simvastatin only (n=18 815) 1.11 (0.97–1.26)
Rosuvastatin only (n=24 714) 1.18 (1.04–1.33)
Pravastatin (n=33 627) 1.03 (0.90–1.19)
Lovastatin (n=6211) 0.98 (0.70–1.38)

Overall, the 174 more cases of diabetes among patients treated with statins translates into one additional case of diabetes per 255 patients taking statins for four years, report the researchers. Using the data from the Cholesterol Trialists' Collaboration, Preiss said this compares favorably with the 5.4 deaths or MIs that are avoided over four years for every patient treated with statins. This benefit is even larger when the reduction in revascularizations and strokes are accounted for with statin therapy.

Investigators report that there was no clear difference between hydrophilic statins (pravastatin and rosuvastatin) and lipophilic statins (atorvastatin, simvastatin, and lovastatin), although there was a statistically significant increased risk among individuals treated with rosuvastatin.

To heartwire , Preiss said the risk appears to be a class effect and that no one statin should be singled out. The 18% increased risk of diabetes in rosuvastatin-treated patients included two heart-failure trials, he noted, a patient population already at an increased risk of diabetes. Similarly, Nissen noted that the PROSPER trial first raised the diabetes issue when investigators observed a statistically significant 32% increased risk of diabetes among pravastatin-treated patients.

Continue to Treat With Statins, Urge Clinicians

Also commenting on the results for heartwire , Dr Richard Karas (Tufts Medical Center, Boston, MA), who was not part of the study, said the concern about diabetes is an important question that needs to be addressed given the widespread use of the drugs worldwide.

This is another opportunity for the media to really scare the pants off people, and I think their findings are very carefully worded--there is a statistically significant but slight increase in the risk of diabetes.”

"The possibility of small risks, when multiplied by the number of people who take statins, is an extremely important issue," said Karas. "I think the paper is well done and careful in its interpretation. This is another opportunity for the media to really scare the pants off people, and I think their findings are very carefully worded--there is a statistically significant but slight increase in the risk of diabetes. It's a provocative finding, but it's also of modest strength because it's information culled from studies not designed to answer this question."

Karas pointed out that a similar issue exists with niacin, in that there are concerns about its use among patients at risk for diabetes. Despite the drugs' adverse metabolic effect in patients at risk for diabetes, clinicians are still very aggressive in using statins and niacin because these are patients who are at an increased risk of cardiovascular events. "It's very important that clinicians don't get the message to shy away from treating patients most in need of lipid-altering therapies," Karas told heartwire .

Preiss reiterated those concerns, saying that the benefits of statin therapy are outweighed by the reduction of cardiovascular events, although the risk/benefit trade-off applies only to patients studied in these 13 trials, those with moderate to high risk of cardiovascular disease, and that the risk/benefit ratio is less clear in low-risk patients treated with statins.

Rosuvastatin Approved Last Week

Last week, the Food and Drug Administration agreed to broader labeling for rosuvastatin based on the results of the JUPITER study. Asked about the decision in light of the newly identified diabetes risks, particularly since more patients will now be eligible for statins, Nissen told heartwire that the patients captured by the new label are at moderate to high risk given their age, elevated C-reactive protein (CRP) levels, and one additional risk factor. As such, the degree that they stand to benefit from statins is greater than the risk they face for developing diabetes, he said

In an editorial accompanying the published study [2], Dr Christopher Cannon (Brigham and Women's Hospital, Boston, MA) calls the increased diabetes risk "almost paradoxical" given statins' benefit of reducing cardiovascular events in patients with known diabetes. He points out, though, that other cardiovascular drugs, including thiazide diuretics, beta blockers, and niacin have similar associations. He writes that this newly identified risk warrants monitoring and that "it seems reasonable to add glucose to the list of tests to monitor in older patients who are on statins."

Asked about mechanisms, no one who spoke with heartwire was sure why there is an increased risk of type 2 diabetes. It is possible, noted Preiss, that the risk is the result of an off-target effect of treatment, or even an indirect result, where those experiencing muscle pain or weakness caused by statins are less likely to exercise, resulting in more diabetes. However, he emphasized more study is needed. Next up, he said, is an investigation of the effects of high-dose vs low-dose statins on the risk of diabetes.

Preiss declares no conflicts of interest. Sattar reports consulting and lecture fees from Merck, Pfizer, and AstraZeneca and has received grant support from Pfizer. Cannon receives research support from Accumetrics, AstraZeneca, Bristol-Myers Squibb/Sanofi-Aventis, GlaxoSmithKline, InteKrin, Merck, Novartis, and Takeda and holds equity in Automedics Medical Systems. Nissen consults for numerous pharmaceutical companies but directs honoraria to charity. Karas reports that he has received research grants from AstraZeneca and Kos Pharmaceuticals; has served on the speakers' bureaus and/or received honoraria from Kos, AstraZeneca, Merck, and Pfizer; and has served as a consultant to Kos.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: