Skip navigation

Statins reduce risk of fractures

Studies
Results
Comment

Statins are widely prescribed because they lower cholesterol levels in blood and reduce the risk of cardiovascular mortality and mortality in hyperlipidaemia. Since the early 1990s some of them have also been known to increase bone formation and bone volume in laboratory animals. There is also preliminary evidence that statins increase bone density in humans. We now have three observational studies [1-3] that statins are likely to reduce fractures.

Studies


All three studies had similar case-control architectures. From databases in the UK and USA patients with fractures were matched with controls and the use of statins and other lipid lowering drugs examined. Possible confounding factors were carefully taken into account, and in all studies patients with conditions that could confuse the analyses (like osteoporosis, cancer, and alcoholism) were excluded. Two studies looked at women and men, and one at women only. Two looked at almost all fractures, and one at hip fracture. All the patients were 50 years or older.

Results


The main characteristics and results of the three studies are shown in Table 1. There was a remarkable similarity between the results, with an overall reduction in fractures, and perhaps particularly hip fractures. Current use and amount of statins consumed in the recent past both gave greater statistical significance. There was a use-response relationship for hip fracture in one study [2]. Use of lipid lowering agents other than statins was not associated with reduced risk of fracture in any study.

Table: Summary of observational studies

Reference Base population Cases Controls Main results (odds ratios with 95% CI)
Meier et al [1] Women and men aged 50-89 years in the UK from GPRD database (late '80s to 1998) taking statins, other lipid lowering drugs, with diagnosis of hyperlipidaemia not on treatment, plus randomly selected patients with none of these indications. 3940 individuals with a first time fracture of femur, humerus, hand, wrist, lower arm, vertebrae, clavicle, foot, or unspecified site 23379 controls without fracture matched on a 1:6 basis with cases Current use 0.55 (0.44 to 0.69) Fractured femur/current use 0.12 (0.04 to 0.41) Significantly reduced risk for any number of prescriptions, any type of fracture, and recent use, but not past use (no use for 90 days or more). No effect with other lipid lowering drugs
Wang et al [2] Women and men aged 65 years and older years in New Jersey from Medicare or Medicaid or Pharmacy programs for aged. 1222 cases with hip fracture in 1994 with no prior hip fracture 4888 controls without fracture matched on a 1:4 basis with cases Current use 0.29 (0.10 to 0.81) Use in past 6 months 0.50 (0.33 to 0.76) Use in past 3 years 0.57 (0.40 to 0.82) Significant use-response, with lower risk with greater statin use over 3 years, with risk for highest use quartile of 0.37 (0.17 to 0.82). No effect with other lipid lowering drugs
Chan et al [3] Women aged 60 years or older in October 1994 with continuous benefits to 1996 in six HMOs in the USA 928 cases of fracture of hip, humerus, tibia, vertebrae or wrist over one year 2747 matched controls without fracture For women with more than 13 dispensings of statins there was a statistically reduced rate of 0.48 (0.27 to 0.83). Lower numbers of dispensings were not statistically different. No effect with other lipid lowering drugs
Note that all the results were generated after adjusting for a variety of confounding factors using a number of different methods

Comment


It looks as if use of statins reduces the risk of fracture by about half, and perhaps hip fracture by more than half. The crude fracture rate for people taking any lipid lowering drug was half that for people with untreated hyperlipidaemia or matched controls who neither had hyperlipidaemia and never took lipid lowering drugs (Figure) in the UK [1].

Figure: Crude fracture risk by treatment [1]


There is a biological plausibility for the effect, both because statins are active in the complex biochemistry of bone regulation, and because they have been demonstrated to increase bone formation in animal experiments [4]. But this is not a green light to prescribe statins to prevent fractures. All it does for now is to give us an additional warm glow that the statins we prescribe may be doing more good than we thought, while remembering that an effect seen in observational studies is not always replicated in randomised trials. Not all statins might be the same, and perhaps doses and delivery routes may have to be different to get effects on bone. Right now it is a case of 'watch this space'.

References:

  1. CR Meier et al. HMG-CoA reductase inhibitors and the risk of fractures. JAMA 2000 283: 3205-3210.
  2. PS Wang et al. HMG-CoA reductase inhibitors and the risk of hip fractures in elderly patients. JAMA 2000 283: 3211-3216.
  3. KA Chan et al. Inhibitors of hydroxymethylglutaryl-CoA reductase and risk of fracture among older women. Lancet 2000 355: 2185-2188.
  4. SR Cummings, DC Bauer. Do statins prevent both cardiovascular disease and fracture? JAMA 2000 283: 3255-3257.
previous or next story in this issue