Aspirin use linked to age-related macular degeneration

Regular use of aspirin was associated with a 2-fold greater risk of developing the neovascular subtype of age-related macular degeneration (AMD) than in non-users, according to research reported online in JAMA Internal Medicine.

Australian researchers’ finding of a longitudinal association-with temporal information in the association, ie, exposure to aspirin before the development of AMD-confirms an observation from a European cross-sectional survey (European Eye Study) that also reported a 2-fold higher prevalence of neovascular AMD among regular aspirin users compared with non-users (Ophthalmology. 2011;119:112–118).

“The reported association from the European Eye Study, however, lacked temporal information as the exposure and the disease outcome were measured simultaneously in the survey,” Paul Mitchell, MD, PhD, principal investigator of the study, called the Blue Mountains Eye Study (BMES), told Formulary. “Nevertheless, this confirmation suggests a consistency of the evidence on this association in 2 different population-based samples. We felt that it would be important to see if there is a longitudinal association present between regular use of aspirin and subsequent development of AMD. As the BMES had such longitudinal data available, we acted on this project without delay.”

Evidence from observational studies can usually only generate and/or support a study hypothesis but cannot resolve completely the controversy of the study question or hypothesis, mainly due to the lack of a proven mechanism(s) for the associations from observational studies, said Dr Mitchell, who is also professor in ophthalmology and director of the Westmead Millennium Institute’s Centre for Vision Research, Sydney, Australia.

The research team conducted a population-based study of 2,389 older Australians aged 50+ years 2 decades ago (1992-1994). They followed this sample over 15 years with 5-year intervals between each examination, and collected a range of data including retinal photographs from study participants.

 “We defined incident AMD using retinal photographic documentation with verification. We collected information on aspirin use via face-to-face interviews in multiple study visits. We used statistical models to assess the link between the 2 while adjusting for possible confounding factors that were also collected in our study, the Blue Mountains Eye Study,” Dr Mitchell said.

While aspirin is among the most effective cardiovascular disease (CVD) preventive therapies to reduce recurrent CVD events (secondary prevention), regular use over the longer term has been associated with a number of well-documented adverse side effects such as increased gastrointestinal, intracerebral, and extracranial hemorrhage. “If our finding is further replicated by other studies, the possibility of a deleterious effect on vision via an increase in the risk of AMD needs to also be considered for patients who are at high risk of AMD [eg, existing neovascular AMD in the other eye], in addition to other known adverse effects,” Dr Mitchell said. 

“Given the widespread use of aspirin, any increased risk of disabling conditions and morbidity would likely be important and may affect small subgroups of patients, in this case, patients who are at high risk of neovascular AMD,” he said. “However, the magnitude of this potential risk is relatively small-9.3% after 15 years-and needs to be balanced with the significant morbidity and mortality of suboptimally treated CVD.”

Currently there is insufficient evidence to recommend changing clinical practice, except perhaps in cases of patients with strong risk factors for neovascular AMD, in whom it could be appropriate to raise awareness among the patients and families of the potentially small, but increased, risk of incident neovascular AMD in patients on long-term aspirin therapy, according to Dr Mitchell.

“The increased risk of neovascular AMD was only detected after 10 or 15 years. This may suggest that cumulative dosage of aspirin may be important in pathogenesis,” he said.

“Our findings also suggest the specificity of the association: only aspirin use (but not other pain killers) was associated with neovascular AMD-but not associated with the early stage of AMD or geographic atrophy, another subtype of late-stage AMD,” Dr Mitchell ­concluded.