Osteoporosis Treatment: Decreased Mortality Too? Free

In the key trial of hip fracture patients treated with zoledronic acid (1), not only were there fewer fractures in the treated patients compared to the placebo group, but there were fewer deaths. Even though hip fracture in elderly individuals is clearly associated with increased mortality, the decrease in mortality in the zoledronic acid–treated patients was due to factors other than prevention of a second fracture. This plus other evidence has led to the hypothesis that drugs for osteoporosis, particularly bisphosphonates, might have salutary effects on tissues other than bone. For example, Reid et al (2) reported on women with osteopenia (some with a history of fragility fracture) who were randomly assigned to receive 4 infusions of zoledronic acid over 6 years or 4 placebo infusions. The women receiving active drug had fewer cardiovascular events and fewer cancers. On the other hand, in a meta-analysis of 38 randomized clinical trials of osteoporosis medications, Cummings et al (3) reported that bisphosphonates were not associated with lower mortality, although the authors suggested that zoledronic acid requires further study to determine if it can decrease mortality. Randomized trials often have restrictive inclusion and exclusion criteria, and they may focus on somewhat younger patients, so the typical osteoporosis patient encountered in practice might not qualify for inclusion. Thus, findings from clinical trials may not reflect what happens in normal clinical settings. Observational studies can assess larger numbers of patients, who may be followed for much longer periods of time. The most rigorous ones can be of great value, despite well-known limitations of observational research.

In this issue of the Journal, Wu et al (4) have taken advantage of the remarkable Taiwan National Health Insurance Research Database, which includes almost all people living in Taiwan, and studied them over a 10-year period. The final group of individuals older than 40 years with a fracture was almost 47 000 (80% female) and mean follow-up was close to 5 years. The authors had access to prescriptions of osteoporosis medications approved by the Taiwan Food and Drug Administration. Fracture patients treated with bisphosphonates or denosumab were compared to fracture patients treated with raloxifene based on a study by Grady et al (5) that actually demonstrated a 10% reduction in all-cause mortality in women treated with raloxifene. The choice of the comparison group makes the effect of bisphosphonates that Wu and colleagues report even greater.

For all fracture patients, compared to those treated with raloxifene, oral bisphosphonates (alendronate and risedronate), and intravenous zoledronic acid were clearly associated with lower overall mortality rates. Interestingly, denosumab was also associated with lower mortality. For people at least age 65 years, the association of these medications with decreased mortality was also significant, with hazard ratios of 0.79 (zoledronic acid) to 0.86 (denosumab).

The likelihood of unmeasured confounders is a potential problem for all observational studies, and in the study by Wu and colleagues (4), measures of 25-hydroxyvitamin D and lifestyle factors such as smoking and alcohol intake were not assessed. Instead, the authors used the Charlson Comorbidity Index as a method to at least partly control for these potential confounders. The size of the study improves the chances that the individuals in each of the drug categories were reasonably balanced in their characteristics. Nonetheless, just like randomized controlled trials, observational studies must be understood with their strengths and weaknesses.

Hence, consistency of results across studies would strengthen the conclusion of the Wu report that there are pleotropic effects of osteoporosis drugs that may affect mortality. A completely different type of observational study was reported by Lee et al (6). In this study, mortality in patients admitted to a busy intensive care unit (ICU) was evaluated. Of 7830 people admitted to an ICU over 11 years, only 245 had been treated with bisphosphonates (for any indication other than malignancy) before admission. Such patients were older (average age, 66 vs 58 years) and had more comorbidities according to the Charlson Index. Yet, those who had been exposed to bisphosphonates had lower hospital mortality (mortality ratio 0.41, and after further adjustments 0.39).

The conclusion from real-world studies is that antiresorptive osteoporosis medications may have beneficial effects beyond the decrease in fracture risk. Will such benefits alter the current crisis in osteoporosis (7), in which patients at high risk for fracture do not start or quickly discontinue osteoporosis medications because of concerns about side effects (8)? If patients can be educated about the additional effects of such medications, will clinicians be better able to convince them that osteoporosis drugs are, on the whole, safe and effective? Clinicians caring for patients at risk for fracture need as much information as possible to provide the strongest benefit-risk assessment to help lower the continuing toll of osteoporotic fractures that will only get worse as the population ages.

Disclosures

The author has nothing to disclose.

References

Abbreviation

 
• ICU

  intensive care unit