Use of COX-2 inhibitors in patients with retinal venous occlusive disease

PURPOSE: To determine the prevalence of prior or current usage of COX-2 inhibitors among patients with retinal venous occlusion (RVO). METHODS: Records of all patients with RVO and control patients matched by age and gender without the diagnosis of RVO seen in a retina referral practice between May 1999 and October 2004 were reviewed. Prevalence of COX-2 inhibitor usage was compared. Multivariable analysis was used to assess the independent correlation of COX-2 inhibitor usage with RVO. RESULTS: A total of 111 consecutive patients with RVO and 316 controls without RVO were identified. There was no significant difference in race or presence of hypertension between cases and controls. Ten of the RVO patients (9%) had a history of using COX-2 inhibitors. Of these 10 patients, one had a central RVO, one had a hemi-central RVO, and eight had a branch RVO. Thirty-nine of the 321 controls (12%) had a history of COX-2 inhibitor use. The prevalence of COX-2 inhibitor usage among RVO patients was not significantly different from that of controls (9% versus 12%; P = 0.37). In a multivariable analysis adjusting for effects of age, gender, hypertension, diabetes, and cardiovascular disease, the association of COX-2 inhibitor usage and RVO was still not significant (P = 0.48). CONCLUSION: A few patients with RVO had prior or concurrent use of COX-2 inhibitors. The prevalence of COX-2 inhibitor usage does not appear to be significantly higher in patients with RVO.