Abstract & Commentary
Synopsis: Use of ß-blockers is associated with reduced risk of fracture. The magnitude of reduction is similar to that seen with use of thiazides.
Source: Schlienger RG, et al. JAMA. 2004;292:1326-1332.
The General Practice Research Database (GPRD) holds medical information on 3 million Britons, including demographic data, diagnoses, hospitalizations, and deaths. The GPs whose practices are part of this research network use computers to write prescriptions, and these prescriptions are also part of the database. In this case-controlled study, the researchers extracted data on all patients with a fracture diagnosis from January 1993 to December 1999 (30,601cases) and searched for correlates.
Inclusion criteria were ages 30 to 79 years and in the database for at least 3 years. Excluded were patients with osteoporosis, osteomalacia, Paget’s disease, cancer (other than non-melanoma skin cancer), or alcoholism or who used bisphosphonates. Randomly matched controls (120,819) were selected with up to 4 controls per case. The ratio of men to women, 40:60, was the same among cases and controls. Schlienger and colleagues looked at exposure to ß-blockers, thiazide or thiazide-like diuretics, and other drugs with known associations to bone health. Each case and the matched controls were classified as current, recent, and past users of a drug based on the date of their last prescription for the drug (1-59 days, 60-119 days, and > 119 days, respectively, from the date of the case’s fracture). They then divided the subjects into 3 groups: ß-blockers only, thiazides only, and ß-blockers plus thiazides.
Among current users of ß-blockers only, the odds ratio (OR) of a fracture for patients with 3-19 prescriptions was 0.63 (95% confidence interval [CI], 0.55-0.73). Among patients with > 19 prescriptions, the OR was 0.83 (CI, 0.76-0.91). When these 2 prescription groups were subdivided by gender, the ORs for men were 0.59 and 0.69, respectively, and for women, 0.67 and 0.92.
The ORs for thiazide only and for ß-blockers plus thiazide did not show this sex difference. When the 2 prescription groups were combined, the OR for current users of ß-blockers only was 0.77 (CI, 0.64-0.79), for thiazide only 0.80 (CI, 0.74-0.86), and for ß-blockers plus thiazides 0.71 (CI, 0.64-0.79). No difference was found when ß-blockers were segregated into hydrophilic versus lipophilic or into ß-blockers with intrinsic sympathomimetic activity (ISA) vs those without ISA.
Other antihypertensive medications were examined. Users of ACE-inhibitors had a decreased risk (OR, 0.81), but users of angiotensin receptor blockers, loop diuretics, or potassium-sparing diuretics had no change in risk. Corticosteroids, benzodiazepines, antipsychotics, and antiepileptics increased risk, and estrogens and statins reduced it.
Comment by Allan J. Wilke, MD
What a rich resource for data mining! These patients are very similar to those that we see in our practices. As a case-controlled study, these results can only be considered associations, not causal. Additionally, although the database records that a patient received a prescription, it is unknown whether the drug was actually taken. Even with these caveats, however, the sheer size of the database makes these findings compelling.
The study does raise some intriguing questions. The different ORs by gender suggest that there may be a possible sex-related difference in prolonged use of ß-blockers. Why would this be true? Schlienger et al do not provide a definitive answer. ß-blockers are not the only pharmaceuticals that are associated with bone health. Last year, Schoffs reported (and Internal Medicine Alert reviewed)1,2 that thiazide use for at least one year is associated with a decreased incidence of hip fractures, but that decrease disappears four months after discontinuation. This study reports a decreased risk among thiazide only users and the risk is further reduced when thiazides are combined with ß-blockers. The mechanism of action for thiazides is presumed to be decreased urinary calcium excretion and possibly a direct effect on osteoclasts.
How do ß-blockers reduce fracture risk? Schlienger et al cite animal and in vitro studies that suggest that the sympathetic nervous system has a catabolic effect on bones. They point out that bones have sympathetic innervation, and osteoclasts and osteoblasts have adrenergic receptors.
How should we use this information? Before we place any patient on a ß-blocker (or for that matter, a thiazide) to prevent or treat osteoporosis, we need confirmation of these findings from well-designed prospective, randomized, controlled studies. On the other hand, these results support the JNC-7 recommendation of making ß-blockers and thiazides first-line drugs for treating hypertension.
Dr. Wilke, Associate Professor of Family Medicine, Medical College of Ohio, Toledo, OH, is Associate Editor of Internal Medicine Alert.
1. Schoffs MWCJ, et al. Ann Intern Med. 2003;139: 476-482.
2. Wilke AJ. Internal Medicine Alert. 2003;25:145-146.
3. Chobanian AV, et al. JAMA. 2003;289:2560-2572.