Study Findings Highlight Need for Caution in Prescribing Newer Drugs

There's no disputing the fact that some drugs currently marketed in the United States have been found to have unintended consequences -- even after the clinical testing mandated by the FDA as part of the new drug application process. A pair of recently published studies on drugs from two separate classes, both of which are used to treat common chronic conditions, illustrate this point. Both studies appeared in the April 28 issue of Archives of Internal Medicine.

A related editorial in the same issue assesses the validity of each of the two studies and discusses the respective risk-benefit profiles of the drugs involved. The editorial also addresses how clinicians should interpret these findings and how they can apply them in clinical practice.

Alendronate, a bisphosphonate indicated to treat postmenopausal osteoporosis, was the subject of a recent population-based controlled study. Two earlier studies had suggested that use of zoledronic acid or alendronate sulfate for treatment of postmenopausal osteoporosis raised the risk of serious atrial fibrillation, or AF, events. This adverse effect had not previously been reported in users of bisphosphonates.

In the new study, "Use of Alendronate and Risk of Incident Atrial Fibrillation in Women," researchers found that women who had used alendronate at any time in the past (defined as having received at least two alendronate prescriptions) had an increased risk of incident AF (1.86 odds ratio) compared with women who had never taken the drug. Adjustment for multiple variables the researchers considered -- such as race, presence of diabetes and body mass index, or BMI -- only slightly altered the odds ratio to 1.83.

Both the unadjusted and adjusted odds ratios were slightly higher in women who had taken alendronate in the past than in current users, but the difference was not statistically significant. The risk of AF was not affected by the total amount of alendronate a woman took.

The risk of sustained AF associated with alendronate use was higher than the risk of transitory or intermittent AF. Of patient subgroups considered (based on the presence of specific variables), the risk of AF associated with alendronate use was higher among patients who had diabetes and in those who took statins.

In a population-based controlled study titled "Use of Thiazolidinediones and Fracture Risk," researchers looked at the risk of fractures in patients taking insulin-sensitizing thiazolidinediones to treat type 2 diabetes. These relatively new drugs are widely used to treat clinical conditions characterized by insulin resistance.

Researchers compared the number of osteoporotic fractures in men and women ages 30-89 years who took the thiazolidinediones rosiglitazone maleate and pioglitazone hydrochloride with the number of fractures seen in individuals who used other oral antidiabetic agents or insulin. Results showed that current use of either thiazolidinedione in patients who had type 2 diabetes was associated with a two- to threefold increased risk of hip and nonvertebral osteoporotic fractures.

Specifically, although short-term use of these agents (defined as receiving one to seven prescriptions) did not alter the relative risk of fracture, current users who had taken thiazolidinediones for at least 12-18 months (eight or more prescriptions) had almost twice the fracture risk compared with controls. The fracture risk was highest among current users who had taken thiazolidinediones for at least two years (15 or more prescriptions); it was almost three times the fracture risk of patients who did not take the drugs.

This finding held true for both men and women, in contrast to previous studies suggesting that increased fracture risk is limited to women. The researchers further determined that fracture risk in patients taking thiazolidinediones was independent of such variables as BMI, duration of diabetes, comorbidities and use of other oral antidiabetic drugs.

An accompanying editorial offers a clinical practice perspective on the issues raised by the two studies. After discussing the relative risks of osteoporosis and diabetes in the general population, the editorial's authors weigh those risks against the potential adverse effects of each of the drugs studied.

When discussing alendronate, the authors point out that "the competing risk of AF in bisphosphonate users has to be weighed against the substantial benefits of reducing fractures, especially in individuals with osteoporosis." They further note that the finding seen in this study has not been consistently reported and that its biological plausibility is uncertain.

Given that the FDA continues to monitor reports of AF in patients who are taking drugs in this class, and because the overall relative risk of AF remains fairly small, the authors conclude that the benefits of bisphosphonate treatment outweigh the risk of AF.

For the thiazolidinedione study, the risk of fracture is strong and correlates well with other studies. In addition, the authors view the finding reported as being biologically plausible. Moreover, they state, other possible adverse effects of thiazolidinedione use include an increased risk of cardiovascular disease, weight gain, hepatotoxicity, congestive heart failure and fluid retention. These drugs also are costly and have not proven to be more effective in lowering glycemia than older oral hypoglycemic agents.

In the absence of long-term clinical trial data showing the benefit of thiazolidinedione therapy in reducing adverse clinical outcomes, the editorial's authors conclude that older hypoglycemic agents (i.e., second-generation sulfonylureas and metformin) should be the preferred treatment for patients with type 2 diabetes.