Proton pump inhibitors associated with fracture and C. difficile

by Charles Bankhead

Proton pump inhibitors (PPIs) significantly increased the risk of both fracture and recurrent infection with Clostridium difficile, investigators in separate studies reported.

PPI use increased the odds of spine, forearm/wrist, and total fractures by 25% to 50% over three years, but had no effect on the risk of hip fracture. Overall, the acid-fighters had a modest effect on bone mineral density (BMD), Seattle researchers reported in the May 10 Archives of Internal Medicine.

PPI use was associated with a 42% increase in the risk of recurrent C. difficile infection, with the greatest risk among patients older than 80 and in patients treated with antibiotics that were not targeted to the bacterium, Boston investigators reported in the same issue of the journal.

Both studies add to existing evidence of potentially harmful effects of chronic PPI therapy.

Several studies have demonstrated an association between PPI use and osteoporotic fractures, Shelly L. Gray, PharmD, of the University of Washington, and co-authors wrote.

However, other studies have shown no association in patients without other risk factors for fracture, or in a series of analyses limited to hip fracture.

Moreover, histamine (H2) receptor antagonists — an older class of acid-fighters — have likewise been linked to a modest increase in hip fracture risk, the authors continued.

Yet few studies have examined the relationship between PPI use and bone mass density (BMD). A better understanding of that relationship might provide insight into the biologic plausibility of an association between PPIs and fracture, the authors wrote.

To that end, they analyzed data from the Women’s Health Initiative (WHI). The primary outcome was three-year incidence of self-reported fracture (hip, spine, forearm, wrist, and total). For a subgroup of patients, the three-year change in BMD was also an outcome of interest.

Of 161,806 women in the WHI observational and clinical studies, 3,396 reported current use of a PPI, and 10,016 reported use of H2 receptor antagonists alone. During 1,005,126 person-years of follow-up, 1,500 hip fractures occurred, as did 4,881 fractures of the forearm or wrist, 2,315 clinical spine fractures, among 21,247 total fractures.

A multivariate analysis yielded the following hazard ratios for PPI users:

* Hip fracture, 1.00 (95% CI 0.71 to 1.40)
* Clinical spine fracture, 1.47 (95% CI 1.18 to 1.82)
* Forearm/wrist fracture, 1.26 (95% CI 1.05 to 1.51
* Total fracture, 1.25 (95% CI 1.15 to 1.36)

Use of H2 receptor antagonists did not increase the risk of hip, spine, or forearm/wrist fracture but modestly increased total fracture risk (HR 1.08, 95% CI 1.02 to 1.14).

PPI users and nonusers had similar baseline BMD, and baseline BMD did not vary by duration of PPI therapy, the authors wrote.

Overall, BMD increased during the three years of follow-up, reflecting some patients’ participation in clinical trials of anti-osteoporosis therapy. The difference in hip BMD was 0.74% in favor of patients not using PPIs (95% CI 0.01 to 1.51). BMD did not differ significantly between groups for any of the other sites assessed or for total BMD.

The findings do not resolve the controversy surrounding PPI use and fractures.

“Questions remain regarding the potential risk associated with long-term PPI use and fracture risk,” the authors wrote. “Thus, based on the accumulation of evidence, it is prudent for clinicians to periodically reevaluate the need for long-term PPI therapy.”

Meanwhile, PPIs and other acid-suppressing drugs have been associated with several types of infection, including C. difficile, Amy Linsky, MD, of Boston Medical Center, and co-authors wrote.

Specifically, studies have linked PPI use to an increased risk of community-acquired and nosocomial C. difficile infection, including some evidence of a two- to threefold increase in risk among PPI users.

“The exact relationship between PPI use and incident C. difficile infection remains elusive, and no causative pathway has been demonstrated,” the authors wrote.

The relationship between PPI use and recurrent C. difficile infection is even less clear, as the few available studies have produced conflicting findings, they continued.

In an effort to resolve some of the uncertainty, the authors performed a retrospective cohort analysis of data from the New England Veterans Healthcare System.

The data came from linkage of pharmacy and administrative records covering the period from Oct. 1, 2003 through Sept. 30, 2008. The authors identified 1,166 patients who had metronidazole- or vancomycin-treated C. difficile infections. Of those, 527 (45.2%) received oral PPIs within 14 days of diagnosis.

The primary outcome was recurrent C. difficile infection, defined as a positive toxin finding between 15 and 90 days after the initial infection.

PPI users had a recurrence rate of 25.2% compared with 18.5% among nonusers. The difference resulted in an adjusted hazard ratio of 1.42 for the patients treated with PPIs (95% CI 1.11 to 1.82).

Patients older than 80 had an even greater risk of recurrent infection (HR 1.86, 95% CI 1.15 to 3.01), as did patients who received antibiotics that were not targeted to C. difficile (HR 1.71, 95% CI 1.11 to 2.64).

“Given the morbidity and cost associated with recurrent C. difficile infection and the lack of readily modifiable risk factors, our findings have important clinical implications,” the authors concluded.

“The data presented herein support the need for critical assessment of PPI use in patients being treated for C. difficile infection as well as further research to test this association.”

Three other studies reported in the same issue of Archives of Internal Medicine provided additional support for potentially harmful effects of long-term, high-dose or inappropriate use of PPIs and other acid-suppressing drugs:

* A pharmacoepidemiologic study of more than 1,000,000 hospital discharges showed a dose-response association between the level of acid suppression and the risk of C. difficile infection (Arch Intern Med 2010; 170: 784-90)
* High-dose PPI therapy proved to be no more effective than lower doses for controlling bleeding ulcers (Arch Intern Med 2010; 170: 751-58)
* Introduction of a standardized guideline for prescribing practices reduced inpatient use of PPIs, but only among patients who were not on PPIs at admission (Arch Intern Med 2010; 170: 779-83)

Collectively, the five studies plus evidence from previous reports show that “for most patients the adverse effects of PPIs outweigh the benefits,” Mitchell H. Katz, MD, of the San Francisco Department of Public Health, wrote in accompanying editorial.

“Reducing the unnecessary use of these medications will require action by both physicians and patients. As physicians we should offer treatments other than PPIs for functional dyspepsia, prescribe short courses of PPI treatment (after disclosure of possible risks and benefits), and consider a trial of discontinuing PPI therapy in patients who are asymptomatic. Once our patients fully appreciate the adverse effects of PPIs, they themselves may prefer other treatments.”

“On a broader level,” he concluded, “the over prescription of PPIs should also remind us to critically evaluate our own treatment paradigms: ‘more is better’ or ‘do no harm’?”

Charles Bankhead is a MedPage Today staff writer.

Originally published in MedPage Today. Visit for more gastroenterology news.


✓ Join 150,000+ subscribers
✓ Get KevinMD's most popular stories