Purchase this article with an account.
J. G. Santiago, J. K. Sun, P. S. Silva, Z. A. Haddad, L. P. Aiello; Risk Factors for Intraocular Ischemia and Neovascularization in Central Retinal Vein Occlusion (CRVO) in Diabetic versus Nondiabetic Patients. Invest. Ophthalmol. Vis. Sci. 2009;50(13):3720.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To identify diabetes-specific risk factors associated with intraocular ischemia and neovascularization (NV) in patients with CRVO.
A retrospective chart review of all patients seen at the Beetham Eye Institute from 1/1/04 to 11/15/08 recorded demographic, systemic and ophthalmic findings on standardized forms for the subgroup diagnosed with CRVO at presentation, 3-6 months later and at last follow-up.
A total of 76 eyes of 72 patients with diabetes mellitus (DM) and 28 eyes of 27 nonDM patients with CRVO were identified from a total 13,571 diabetic (0.5%) and 6,077 nonDM (0.4%) patients. With CRVO, mean age was 65+14 (range: 30-95) y, 55% female and 79% white. In CRVO with DM, mean duration of DM was 16+10 (1.5-45) y, HbA1c was 7.8+1.3 [5.0-11.3] % and 19% were type 1 (T1DM). Compared to all patients with DM, CRVO with DM had higher median age (66 vs 57 y, p<0.0001) and shorter DM duration (12.8 vs. 15.1 y, p=0.007). NonDM CRVO patients were more likely to use glaucoma medications prior to CRVO diagnosis (63 vs 13%, p=0.002) and had lower median IOP (14 vs 16 mmHg, p=0.01) than DM with CRVO. Ischemic CRVO occurred in more DM than nonDM eyes (13.9 vs 5.6%, p = 0.36) and more DM than nonDM eyes developed NV of the disc (21 vs 0%, p=0.05). DM eyes were also more likely to develop any NV, including NV elsewhere and on the iris, although this was not statistically significant (49 vs 29%, p = 0.15). Panretinal photocoagulation (PRP) after CRVO was more common in DM (49%) than nonDM eyes (21%, p=0.01). By comparison, only 15% of non-CRVO DM eyes required PRP during this period. No patient with optic nerve head collateral vessels (CVs) at baseline had ischemic CRVO whereas 12.5% of patients without CVs had ischemic CRVO. CVs at baseline were associated with lower subsequent rates of PRP (14 vs 46%, p=0.03). CRVO patients with T1DM were more likely to have PRP (79 vs 42%, p=0.01), retinal NV (29 vs 4%, p=0.004) at endpoint, and less likely to have CVs (23 vs 56%, p = 0.04) than T2DM.
Diabetic patients with CRVO have higher rates of ischemic CRVO and NV, and are more likely to require subsequent PRP than are nonDM patients. T1DM patients are even more prone to NV than are T2DM. The finding that nonDM patients are more likely to use glaucoma medications prior to CRVO suggests that glaucoma is a stronger risk factor for CRVO in nonDM patients, or that vascular pathology inherent in DM potentiates other CRVO risk factors. Finally, the role of CV formation in reducing retinal ischemia and subsequent need for laser treatment warrants further investigation.
This PDF is available to Subscribers Only