Factors predicting short–term improvement in visual acuity following intravitreal triamcinolone injection for diabetic macular edema refractory to focal laser therapy

Abstract

Abstract: : Purpose: Despite focal laser treatment for diabetic macular edema (DME), many patients fail to improve vision. Intravitreal triamcinalone (IVTA) has become an increasingly common treatment for DME refractory to focal laser, but many patients still fail to respond. We sought to determine what factors predict improvement in visual outcome based on baseline clinical, fluorescein angiographic, and optical coherence tomography (OCT) findings. Methods: Charts of 28 eyes in 24 patients were retrospectively reviewed. All patients in this study experienced no improvement in vision despite 1–3 sessions of focal laser therapy. All patients underwent intravitreal injection of 4 mg of triamcinalone acetonide. Data was collected regarding visual acuity, intraocular pressure, clinical appearance of macular edema (cystoid, diffuse, traction, lipid), OCT thickness and morphology, and fluorescein angiographic features (extent and location of leakage, non–perfusion). Patients who experienced improvement in their visual acuity (at three months follow–up) greater than or equal to 0.3 logmar units (or doubling of the visual angle) were defined as "responsive"; all other patients were deemed "unresponsive". Data was compared between these two groups. Results: 17 (61%) eyes were classified as responsive, while 11 (39%) were unresponsive. Responsive patients improved vision 0.50 + 0.25 logMAR units, while unresponsive patients decreased vision –0.13 + 0.36 logMAR units. At baseline, all patients had vision of 20/40 or worse, and thus all patients were capable of improving by at least three lines of vision. Pre–injection visual acuity was significantly worse in the responsive compared to unresponsive eyes (logmar 0.52 + 0.24 vs. 1.4 + 0.5, Snellen equivalent 20/476 vs. 20/67, p<0.0001). Initial review of the color photographs and fluorescein angiograms suggest that foveal lipid and mild to moderate degrees of macular ischemia do not correlate with responsiveness to IVTA. Conclusions: With increasing numbers of patients being treated with IVTA for DME, and given that the treatment is not without risk, it is important to identify factors which will predict which patients will improve. This study, though limited by its retrospective design, suggests that poor initial visual acuity predicts responsiveness to IVTA. Fluorescein angiography findings do not appear to be predictive of responsiveness to IVTA. Further study appears warranted.

Keywords: diabetic retinopathy • corticosteroids • macula/fovea