Anti-Inflammatory Drugs Do Not Prevent Dementia, Alzheimer's Disease

Pauline Anderson

April 23, 2009

April 23, 2009 — Contrary to previous reports, a new study suggests that nonsteroidal anti-inflammatory drugs (NSAIDs) do not prevent dementia, including Alzheimer's disease (AD), but may simply delay disease onset.

The study showed that compared with nonuse of NSAIDs, heavy use of these drugs was associated with a 66% increased risk for dementia.

The authors speculate that earlier research found less dementia in NSAID users because those studies included relatively young populations, and therefore the true action of NSAIDs delays the onset of dementia rather than suppressing the disorder.

However, at this point this is merely a hypothesis, and the question of why this study shows NSAIDs may cause dementia while other research shows the opposite is still an open one, said study author John C. S. Breitner, MD, from the Department of Veterans Affairs and the University of Washington School of Medicine, in Seattle.

"There's no escaping the fact that this study found that NSAID users have more dementia in their 80s than nonusers. But this and earlier studies do not form a basis for clinical-practice decisions," Dr. Breitner told Medscape Neurology.

The study is published online April 22 in Neurology.

Sophisticated Database

Researchers analyzed NSAID use and development of dementia in the ongoing, prospective, community-based Adult Changes in Thought (ACT) study. Between 1994 and 2003, 3392 cognitively intact older subjects who were members of Group Health (GH), a large healthcare-delivery system in King County, Washington, were enrolled in ACT.

Of these participants, 3026 met this study's requisite 10 years of membership in the GH and 2736 completed at least 1 year of follow-up. About one-quarter of the study subjects were 80 years of age or older.

At baseline, the researchers collected data on body-mass index (BMI), APOE genotype, and self-rating of health, physical activity, alcohol use, and smoking status. They determined subjects' cognitive function using the Cognitive Abilities Screening Instrument during biennial visits that continued for 12 years.

To evaluate exposure to NSAIDs, researchers used both patient self-reports and pharmacy information. They had access to a sophisticated electronic pharmacy-dispensing database that kept detailed prescription information, including drug name, strength, and amount dispensed, going as far back as 1977.

To determine average daily drug doses over time, researchers calculated what they called standard daily doses (SDDs) associated with each NSAID dispensed. They considered all prescriptions filled within a 2-year window, but to avoid NSAID use being influenced by dementia, their analysis did not include the last year before dementia onset.

They classified NSAID exposures into light or no use (less than 60 SDDs); moderate use (60 to 499 SDDs); and heavy use (500 or more SDDs) within 2 years.

Intriguing Hypothesis

Of the subjects who had at least 1 year of follow-up, 1380 were light or nonusers of NSAIDs at baseline while 351 subjects were heavy users, and another 107 became heavy users.

During the study, 476 patients developed dementia, and most (356) were diagnosed with AD. Heavy NSAID use was associated with a 66% increased risk for dementia compared with nonuse (adjusted hazard ratio, 1.66; 95% CI, 1.24 – 2.24).

These results underline the complex nature of the relationship between NSAIDs and AD, said Dr. Breitner, adding that there are perhaps 3 possible explanations for the study's findings.

One is that the methods he and his colleagues used were more comprehensive than those of previous studies. The prescription database, for example, was more sophisticated than using just patient reports.

A second possible explanation is that this study's cohort was considerably older than even the 70- or 75-year-old subjects included in other studies. "These drugs might do different things in younger people than in older people," said Dr. Breitner.

But the most "intriguing" hypothesis, he said, is that these drugs postpone the onset of dementia. "Postponing does not mean preventing. It stands to reason that postponing means there will be fewer cases among a younger population and more cases among older people."

Delayed Onset Equal to Prevention

In late-onset diseases such as AD, delaying and preventing can often amount to the same thing, said Dr. Breitner. "You might be delaying dementia to age 90, but a lot of people don't live to age 90 and die without ever getting it; that's pretty close to preventing it."

Viewing related research as a whole, Dr. Breitner said the delay hypothesis makes sense. "In studies of relatively young people, you would see less dementia in users; in an intermediate aged cohort it's kind of a wash; and here, in this still-older cohort, we have more dementia in the users."

Earlier research had shown that people with rheumatoid arthritis (RA) do not develop dementia to the same degree as the general population. It was this finding that first led researchers to suspect that anti-inflammatory drugs used to treat RA might have a preventive effect.

The bottom line is that this study should not change the way doctors prescribe NSAIDs to the elderly, said Dr. Breitner. "People should not be taking these drugs and doctors should not be prescribing these drugs at this point for the prevention of AD — or for delaying its onset. To have that as a clinical indication, we really have to have clinical trials."

Asked to comment on the findings, Ronald C. Petersen, MD, a neurologist who heads up the Mayo Clinic Study of Aging and a member of the American Academy of Neurology, agreed that the relatively old age of the study cohort might play a role. The results may reflect a "catch-up phenomenon," he said.

"You might be seeing more of the disease because it has sort of been delayed; now that you're looking at an older cohort of individuals, you're actually going to see more of the disease. So it doesn't prevent the disease; it just might delay it."

The authors report no disclosures.

Neurology. Published online April 22, 2009.


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.