Purpose: To assess whether an association exists between systemic use of anti-inflammatory medications at baseline and the prevalence or incidence of either late or early age-related maculopathy (ARM) in a population-based cohort.
Methods: 3654 participants of the Blue Mountains Eye Study baseline examination (1992-94) were followed during 1997-99. Use of anti-inflammatory medication was recorded during the baseline interview. After excluding 543 persons who died since baseline, 2334 (75% of the surviving participants) attended 5-year follow-up examinations. Retinal photographs taken during both examinations were graded using the Wisconsin Age-Related Maculopathy Grading System. Prevalence refers to the proportion of participants having ARM at baseline. Incidence refers to the proportion of participants without ARM at baseline who developed it over the 5-year period. Known ARM risk factors were adjusted for when assessing the relationship between use of anti-inflammatory medications and ARM.
Results: At baseline, 1010 (27.6%) of 3654 participants were current users of non-steroidal anti-inflammatory drugs (NSAIDs), 514 (14.1%) were past users and 1282 (35.1%) were ever users. The corresponding numbers of subjects reporting current, past or ever use of corticosteroids (including inhaled steroids) were 225 (6.2%), 519 (14.2) and 564 (15.4), respectively. Late ARM was present in 72 participants (2.0%) and early ARM was present in 171 participants (4.9%) at the baseline examination. During the follow-up period, 25 participants (1.1%) developed incident late ARM and 192 (8.7%) developed incident early ARM. After adjusting for age, sex, family history of ARM and smoking, no significant associations were evident for the use of NSAIDs or corticosteroids and the prevalence of either late or early ARM. There were also no associations found between use of these medications at baseline and the 5-year incidence of either late or early ARM.
Conclusions: No association was found between use of systemic anti-inflammatory medications and either the cross-sectional prevalence or longitudinal incidence of ARM in this population.