COMMENTARY

Smoking, PPI Use, and Hip Fracture: What's the Bottom Line?

David A. Johnson, MD

Disclosures

April 26, 2012

Use of Proton Pump Inhibitors and Risk of Hip Fracture in Relation to Dietary and Lifestyle Factors: A Prospective Cohort Study

Khalili H, Huang ES, Jacobson BC, Camargo CA Jr, Feskanich D, Chan AT
BMJ. 2012;344:e372

Study Summary

Concern over the potential association between long-term use of proton pump inhibitors (PPIs) and bone fractures (hip, neck, and wrist) has been increasing. The US Food and Drug Administration (FDA) issued a warning about this association in 2010 and prompted a product label change for all manufactures of prescription PPIs.[1]

Suggested explanations for this increased risk have included pH-related alterations of ionized calcium absorption, alteration of osteoclastic proton pump activity, and hypergastrinemia related to hyperparathyroidism. However, none of these pathways has been established as the mechanism of increased fracture risk. Furthermore, to date, the studies have been retrospective database analyses with limited ability to adjust for other risk factors for bone density loss or fracture.

The Nurses' Health Study is a prospective cohort study that began in 1976 and involved 121,700 female registered nurses (ages 30-55), who were then followed sequentially every 2 years using validated questionnaires to assess health, activity, and dietary intakes. Compliance with returning the information has been remarkably strong (> 90% to date). In the current study, the investigators analyzed bone fracture history in this cohort, excluding the 15% who reported trauma in association with fracture (eg, skiing or falling down steps). The use of PPIs was assessed, along with other risks such as menopausal status, smoking, hormone replacement therapy, obesity, activity levels, known osteoporosis, and use of other medications (steroids, calcium supplements, bisphosphonates, and thiazide diuretics).

In the 565,786 person-years of follow-up, 893 nontrauma-related hip fractures were reported. Compared with women who did not use PPIs, the risk for hip fracture was 35% higher in those who used PPIs for at least 2 years (age-adjusted hazard ratio [HR], 1.35; 95% confidence interval [CI], 1.13-1.62), with longer PPI use associated with increasing risk. Adjustment for other risk factors (body mass index, physical activity, and calcium intake) did not alter this association. Among current and former smokers, the HR for fracture was 1.51 (95% CI, 1.20-1.91). In women who had never smoked, however, there was no associated risk (HR, 1.06; 95% CI, 0.77-1.46). The investigators also performed a meta-analysis of 10 previous studies, including a total of 1,562,862 individuals, that examined PPI use and fracture risk. The pooled odds ratio (using a random-effects meta-analysis) of hip fracture associated with PPI use was 1.28 (95% CI, 1.19-1.37).

Viewpoint

This study differs from most of the preceding reports on this topic in that it is a prospective evaluation and considers many other risks for bone fracture. This study is illustrative of how covariate risks may potentially skew the interpretation of harm -- in this case, relative to PPI and bone fracture risks.

Cigarette smoking is a known risk factor for osteoporosis. Research has shown an association between tobacco use and reduced bone density.[2,3] However, it is difficult to determine whether reduced bone density is a consequence of smoking itself or of other risk factors common among smokers. Factors such as physical activity, diet, and age at menopause can influence a smoker's risk for osteoporosis, in addition to tobacco use.

The study by Khalili and colleagues suggests a causal relationship between PPIs and bone fracture that is applicable only to smokers. However, they attribute a risk for bone fracture to smoking and taking PPIs that parallels the risk associated with smoking alone. Patients should be made aware of the data and the FDA recommendations, but also that the evidence for harm associated with PPI use is weak.

Abstract

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....