Corley D. Proton pump inhibitors, H2 antagonists, and risk of hip fracture: A large population-based study. Digestive Disease Week 2009; Abstract 414. Chicago, June 2009.
++++
DDW: Fracture Risk Associated with PPIs Confirmed, With Limits
By John Gever, Senior Editor, MedPage Today
Published: June 04, 2009
Reviewed by Robert Jasmer, MD; Associate Clinical Professor of Medicine, University of California, San Francisco and
Dorothy Caputo, MA, RN, BC-ADM, CDE, Nurse Planner
Action Points ÂÂ
• Explain to interested patients that PPIs and histamine-2 (H2) receptor antagonists are effective treatments for gastroesophageal reflux disease, which can have serious consequences if allowed to continue unabated.
• Explain that earlier studies had suggested a link between PPIs and fracture risk, although the mechanisms are unclear.
• Explain that this was a retrospective analysis of medical records, a relatively weak form of evidence. It is possible that other factors unaddressed in the study could account for the findings.
• Note that this study was published as an abstract and presented orally at a conference. These data and conclusions should be considered to be preliminary until published in a peer-reviewed journal.
CHICAGO, June 4 -- Another large retrospective study has found a strong link between use of proton pump inhibitors (PPIs) and hip fracture risk, but the excess risk appeared to be confined to certain types of patients, it was reported here.
Those patients appeared to have at least one other standard risk factor for hip fracture, such as renal impairment, diabetes, or glucocorticoid use, according to Douglas Corley, M.D., Ph.D., of Kaiser Permanente's research division in Oakland, Calif.
An analysis of the healthcare company's massive database found that the rate of hip fractures was increased by about 30% in patients using PPIs for two years or more prior to fracture, compared with nonusers, Dr. Corley told colleagues at in a presentation here at Digestive Disease Week.
The study focused on 33,752 Kaiser members who had suffered a hip or femur fracture along with more than 130,000 controls matched for age, sex, race, and length of Kaiser membership.
In men, the odds ratio for hip fracture with at least two years' PPI use was 1.34 (95% CI 1.18 to 1.51). The odds ratio for women was 1.28 (95% CI 1.17 to 1.39).
A significant but somewhat smaller increase in hip fractures was also found for another class of acid-suppressant medication, H2 receptor antagonists.
The odds ratio for hip fracture with two years or more of treatment with these drugs was 1.18 (95% CI 1.08 to 1.29).
But the excess risk from the two acid-suppressants was concentrated in patients who were already at risk for hip fractures.
"People who did not have other risk factors for hip fracture did not have an excess risk from taking these agents," Dr. Corley said.
Moreover, taking PPIs for longer periods did not appear to increase the fracture risk further.
"There was an increased risk after taking [a PPI] for a couple of years, but if you were taking it for three years, four years . . . up to over ten years, because we had a fair amount of follow-up with this, there was not a substantial increased risk with duration," Dr. Corley said.
But the Kaiser study did identify a relationship between daily doses and hip fractures for those on the drugs for at least two years.
Patients taking fewer than three-fourths of a pill daily showed no excess risk. Those with a daily dose of 0.75 to 1.49 pills had an odds ratio of 1.30 (95% CI 1.19 to 1.42). The odds ratio for higher doses was 1.41 (95% CI 1.21 to 1.64) (P for trend not reported).
The same patterns for dose and duration -- positive association for dose, none for duration -- were seen for H2 antagonists, Dr. Corley said.
The study also suggested that patients who took PPIs for a while but then stopped saw their fracture risk decline, though not all the way to the no-PPI baseline, he reported.
He said there was no indication that the results were confounded by other medications patients were taking. The analysis showed no associations with drugs such as ACE inhibitors, calcium channel blockers, or non-narcotic painkillers.
It did show increases in hip fracture risk for drugs that would be expected to have such an association, such as corticosteroids, estrogen, and bisphosphonates.
Adjusting for comorbidities did not change the overall study results, Dr. Corley said.
When the researchers looked at the effects of age, they found the greatest increase in risk among those 50 to 59 years old (OR 2.31, 95% CI 1.67 to 3.18).
A 2006 retrospective database study in Great Britain found a 44% increase in older patients with at least one year of PPI treatment, with much greater risks seen with long-term, high-dose use. (See Proton Pump Inhibitors Linked to Fracture Risk)
Another database study published last year indicated that PPI use for at least seven years led to a nearly doubled risk of osteoporotic fracture. (See Proton Pump Inhibitors Over Years Increase Osteoporotic Fracture Risks)
"PPIs are overused," commented Nicholas J. Shaheen, M.D., of the University of North Carolina in Chapel Hill, N.C., who moderated a press briefing on the study.
"Anybody with a pain between their chin and their knees who walks into a primary care office, there's a very good chance they're going to walk out with a PPI script," he remarked.
Not only is it sometimes inappropriate, he said, but treatment also may be unnecessarily long lasting.
"Once they're on [PPIs], they're oftentimes on forever," Dr. Shaheen said.
He added that the overall safety profile of PPIs is perceived as good, which may lull physicians into neglecting to re-evaluate the need for continuing treatment.
Kaiser Permanente funded the study.
Dr. Corley is an employee of Kaiser and also reported a relationship with Wyeth.
Dr. Shaheen reported relationships with AstraZeneca, Barrx Medical, CSA Medical, Procter & Gamble, Takeda, and TAP.
Primary source: Digestive Disease Week
Source reference:
Corley D, "Proton pump inhibitors, H2 antagonists, and risk of hip fracture: A large population-based study," DDW 2009; Abstract 414.
© 2004-2009 MedPage Today, LLC.