Re-Evaluation of Artery-Vein Relationships as a Risk Factor in Branch Retinal Vein Occlusion

Abstract

Purpose: : To confirm or refute the findings of previous studies which showed that artery-anterior-to-vein patterns of arteriovenous (AV) crossing increase one’s risk of branch retinal vein occlusion (BRVO) using a larger sample size.

Methods: : A case-control study was performed on digital color fundus photo images of 263 eyes with BRVO (249 patients) referred for retinal photography in a tertiary retinal referral academic center taken over a 5 year period. The AV crossing sites in BRVO patients were studied. All occlusion sites were classified into two groups: Those involving 1st order veins (166) and those involving 2nd order veins (97). The corresponding crossings on the opposite temporal arcade in study eyes and on the same arcade in fellow eyes represented controls 1 and 2 respectively. As another control, 149 corresponding AV crossings of 50 eyes of age-related macular degeneration (ARMD) patients without BRVO were studied. The percentage of AV relationship patterns were compared between the study crossings and control 1 and 2 crossings in each vein order group and the vein order groups combined, as well as with the ARMD control group.

Results: : In the 1st order vein group, 78% of BRVO crossings were of the artery-anterior-to-vein pattern, compared to 76% in control 1 (P>0.75) and 81% in control 2 (P>0.5). In the 2nd order vein group, 80% of BRVO crossings were of the artery-anterior-to-vein pattern, compared to 78% in control 1 (P>0.5) and 70% in control 2 (P>0.1). In the entire group, 79% of BRVO crossings were of the artery-anterior-to-vein pattern, compared to 77% in control 1 (P>0.5) and 78% in control 2 (P>0.9). In the ARMD control group, 65% of crossings were of the artery-anterior-to-vein pattern (P<0.005).

Conclusions: : This study of BRVO eyes showed that there were no significant differences in AV relationships in the comparisons among the BRVO eyes. These findings are in contrast to previous studies, which showed a significantly higher artery-anterior-to-vein pattern in BRVO AV crossings compared to controls among the BRVO eyes. This study does support previous findings in that there was a significant difference in AV relationships between BRVO crossings and ARMD, non-BRVO, crossings. Ours is the largest study on this topic to date. Our findings may differ because of the smaller sample size of previous studies and the choice of controls. The previously accepted belief that artery-anterior-to-vein patterns of AV crossing are a risk factor for BRVO needs to be reconsidered in light of our findings. Further studies are required to settle this debate.

Keywords: vascular occlusion/vascular occlusive disease • retina • imaging/image analysis: clinical