Meta-analysis: Atorvastatin reduces CV events and increases new-onset diabetes in patients with coronary disease

In a meta-analysis including 15,000 patients from two randomized controlled trials (taking 80 mg/day of atorvastatin, 10 mg/day of atorvastatin, or 20 to 40 mg/day of simvastatin), those with two to four risk factors for diabetes were more likely to have new-onset diabetes (NOD) in the high-dose statin group. Patients with no or one risk factor had no increase in NOD. However, the number of cardiovascular (CV) events was significantly lowered by the higher dose in both risk-factor groups.

The study was published in the January 15 Journal of the American College of Cardiology. The following commentary by Sanjum S. Sethi, MD, and Michael E. Farkouh, MD, MSc, was published in the ACP Journal Club section of the July 16 Annals of Internal Medicine.

Statins are a mainstay of therapy for patients with established CAD [coronary artery disease], but recent reports suggest that high doses may increase risk for NOD. Waters and colleagues address this issue in a meta-analysis of 2 pivotal trials. Overall, the number needed to harm for NOD is comparable to the number needed to treat to prevent 1 additional CV event. Analysis by the number of baseline risk factors for diabetes showed that the risk for NOD was greater in patients with 2 to 4 risk factors.

The mechanism by which NOD is related to statin use is unknown but may be linked to increased muscle insulin resistance secondary to statin-induced myopathy. Risk factors for NOD in patients treated with statins are not dissimilar to those in patients who are not—statins may simply hasten the onset of diabetes in patients already at high risk.

The increased risk for microvascular complications from NOD should be weighed against the benefit of reducing CAD risk with statins in each patient with dysglycemia. For younger patients with CAD, NOD may have important implications since diabetic complications increase with time, whereas the preventive effects of statins may be more stable. We recommend that patients at risk for diabetes be informed about the risk for NOD and its complications. It is reasonable to treat those at highest risk with moderate-dose statins to mitigate some of the risk for NOD while other interventions are being pursued. Overall, this doesn't change the playing field for most patients with CAD since all patients, regardless of the number of risk factors for NOD, had a reduction in major CV events.