AGS: Proton Pump Inhibitors Don't Add Pneumonia Risk to Older Adults

SEATTLE, May 4 -- Gastric acid suppression with proton pump inhibitors appears to hold no excess risk of pneumonia in hospitalized older adults, researchers reported here.

The retrospective study's results counter those from a much larger population-based study of ambulatory adults, said Jen-Tzer Gau, M.D., Ph.D., of Ohio University in Athens, and colleagues.

On the basis of the results physicians might feel more comfortable using proton pump inhibitors in older hospitalized patients, Dr. Gau said at the American Geriatrics Association meeting.

Acid suppressing medications had been found in some studies to reduce risk of upper respiratory infections by reducing aspiration of acid reflux. Other studies, though, found the opposite possibly due to increased pathogen colonization of the stomach as pH increases.

Then in 2004, a large case-control study published in the Journal of the American Medical Association created a stir by finding a negative impact of proton pump inhibitors on pneumonia risk.

It reported that proton pump inhibitors users in a prescription database in Holland had an adjusted relative risk for pneumonia of 1.89 compared with former users.

So, Dr. Gau and colleagues reviewed records for all 350 adults ages 65 and older treated at a community hospital over a year to see whether the same association would appear.

Among these patients, 79 were discharged with a diagnosis of pneumonia on the basis of chest radiographs and clinical presentation and 148 were proton pump inhibitors users prior to admission.

The researchers found no significant difference in the prevalence of pneumonia by PPI therapy (26% PPI users versus 20% nonusers, P=0.192).

Likewise, patients with pneumonia were not significantly more likely to have PPIs in their medication history than patients without pneumonia (49% versus 40%, P=0.1476).

"Our study does not reveal an association between the prevalence of pneumonia and gastric acid suppression therapy use," the researchers concluded.

"Gastric acid suppressive therapy may decrease the frequency of reflux episodes leading to a diminished prevalence of microbial aspiration," they posited.

The explanation for the difference between their findings and those of large population study may be the smaller sample size and the different study settings.

There is a possibility that older adults may have less aspiration with acid reflux, said Utkarsh Acharya, a medical student at Ohio University and a coauthor on the paper.

Nevertheless, "our study has more strength," Dr. Gau said, because it used diagnostic criteria for determining pneumonia rather than just antibiotic prescription.

One of the challenges with a population-based database study is the inability to determine confounding differences in patient characteristics, which may have played a role in differences between the studies as well.

In Dr. Gau's study, there were significant differences between patient groups. These included:

• Proton pump inhibitors users had a significantly higher prevalence of gastroesophageal reflux disease (60% versus 10%, P<0.0001), chronic heart failure (P=0.0002), anemia (P=0.006), and antihypertensive use (P=0.008).
• Proton pump inhibitors users had significantly higher mean serum potassium levels compared with nonusers (4.3 versus 4.1 mEq/L, P=0.031).
• Patients with proton pump inhibitors use prior to admission also had longer hospital stays than non-PPI users (5.5 versus 4.3 days, P=0.003).
• The pneumonia group had significantly higher prevalence of smoking, chronic heart failure, coronary artery disease, prior history of pneumonia, and chronic obstructive pulmonary disorder compared with older patients hospitalized for nonpneumonia diagnoses.
• Patients with pneumonia were significantly more likely to use anti-psychotropic medication, calcium channel blockers, and albuterol inhalers compared with controls.
• The pneumonia group had higher white blood cell counts during hospitalization and longer hospital stays (P<0.0001 and P=0.0002, respectively).

Whatever the reason for the difference between the two studies, the new findings are good news for patients, Acharya said, particularly because "gastroesophageal reflux disease and GI problems are highly rampant in the geriatric population."

The researchers said the findings are not conclusive evidence and further study will be needed to confirm the lack of pneumonia risk with proton pump inhibitors in this population.