Penicillin may help prevent strokes in the elderly

American Heart Association
Friday, 8 August 2003

Hypertension patients taking penicillin were half as likely to have a stroke as other hypertension patients, researchers report in today's rapid access issue of Stroke: Journal of the American Heart Association.

"Penicillin showed a consistent and significant protective association with the occurrence of stroke in elderly hypertensive subjects," says lead researcher Paul Brassard, M.D., MSc, a clinical-researcher at McGill University Health Centre's Department of Medicine in Montreal.

While researchers found an overall protective trend with other antibiotics, it was not a consistent significant association.

Many studies have linked infections to heart disease and stroke. Atherosclerosis (fatty build up in artery walls) may be a chronic inflammatory condition with a treatable infective cause. Treating the infection, might prevent the start of cardiovascular disease, halt its progression, or stop the cascade of events resulting in a stroke. Previous studies have had inconsistent results.

Brassard and colleagues evaluated the effect of antibiotic use in preventing strokes among elderly patients being treated for hypertension. They analyzed the health records of 29,937 patients (average age 78) treated for hypertension between 1982 and 1995. Researchers looked back at potential exposure to antibiotics before the stroke occurred. They focused on 1,888 patients discharged from hospitals between 1987 and 1995 with a primary diagnosis of stroke. The researchers then randomly selected a control group of 9,440 patients who were discharged without having had a stroke during the same period. All were followed for an average of five years.

Information for each outpatient prescription for antibiotics was available, including the nature, quantity, strength and dosage of the drug and its dispensing date and prescribed duration. Each patients' history of antibiotic use was available for up to 14 years before they were admitted to the hospital. The researchers noted whether antibiotics had been taken within 30 days of hospital admission, described as current use; within 30 days to a year of admission, described as recent use; and more than a year before, described as past use.

Those who had a stroke were compared to the control group based on antibiotic use. Researchers adjusted for potential confounders, such as diabetes, previous heart disease, cardiac arrhythmias and use of aspirin and cortico-steroids, among others.

With penicillin, current, recent and past use all showed a statistically significant protective benefit, but one that was stronger in people currently using the antibiotic. Current penicillin users were 47 percent less likely to have a stroke. Recent use was associated with a 27 percent reduction in risk and past use was linked with a 14 percent reduction in risk.

Other classes of antibiotics studied included cephalosporins, macrolides, quinolones, and tetracylclines. Although the other antibiotics failed to show a consistent protective benefit against stroke, recent use of antibiotics showed a significant association with stroke prevention. Even though this association with stroke prevention was not statistically significant for most antibiotics, a trend toward protecting against strokes was found but lessened over time, researchers say.

The researchers caution that several limitations to this study should be considered. "First, antibiotic use is only an indirect marker of the presence of infection," write the researchers. Therefore, they can't directly link antibiotic use to the frequency of stroke. They also acknowledge they don't know if all patients took their antibiotics for the prescribed duration, which influences their effectiveness.

Because this was an observational study, it could not consider other important factors such as smoking and socioeconomic status. The researchers note that another limitation is the lack of information on potential strokes occurring before their study period, which would have limited their ability to account for differences between stroke patients and controls with respect to pre-existing heart disease. Finally, random error or chance in these study findings is possible.

However, they say this "intriguing association between penicillin use and stroke should be explored further."

Co-authors are: Chantal Bourgault, Ph.D.; James Brophy, M.D., Ph.D.; Abbas Kezouh, Ph.D.; and Samy Suissa, Ph.D.

For more information, or to contact American Heart Association, see their website at: www.americanheart.org