Abstract
PURPOSE To investigate the clinical importance of changes in diabetic retinopathy severity score (DRSS) in patients with diabetic macular edema (DME) treated with intravitreal ranibizumab. DESIGN Post hoc analysis of the phase III RIDE and RISE studies of ranibizumab for treatment of DME. PARTICIPANTS Four hundred sixty-eight eyes treated with ranibizumab from randomization with gradable DRSS on baseline fundus photographs. METHODS Visual and anatomic outcomes were examined in eyes grouped according to DRSS change from baseline to month 24. MAIN OUTCOME MEASURES Mean best-corrected visual acuity (BCVA) letter score change, proportion of patients with 15 or more Early Treatment Diabetic Retinopathy Study (ETDRS) letter score change, mean contrast sensitivity change, proportion of patients with resolved macular edema, and leakage on fluorescein angiography. RESULTS Most (56.8%) patients treated with ranibizumab experienced 1-step or more improvement in DRSS from baseline to month 24; 40.0% had no change, and 3.2% experienced DRSS worsening. Patients with DRSS stability or improvement had greater mean BCVA letter score changes (+15.1, +14.2, +11.3, and +11.2 letters for ≥3-step improvement, ≥2-step improvement, 1-step improvement, and no DRSS change, respectively) compared with +5.0 letters in patients who had any DRSS worsening. Best-corrected visual acuity letter score gain of 15 letters or more was more common in patients with 2-step or 3-step or more DRSS improvement (51.9% and 44.6%, respectively) compared with those with a 1-step DRSS improvement, no change, or worsening (37.9%, 39.6%, and 26.7%, respectively). A loss of 15 letters or more in BCVA was more common in patients with any DRSS worsening (13.3%) compared with patients who had stable or improved DRSS (0%-2.8%). Resolution of macular edema was more common in patients with DRSS improvement: 84.2%, 87.7%, and 92.3% of patients with 1-step, 2-step or more, and 3-step or more improvement in DRSS achieved central foveal thickness of 250 μm or less, compared with 65.2% and 53.3% of patients who had no DRSS change or any DRSS worsening. CONCLUSIONS These findings provide further support that improvement in DRSS is a clinically important outcome that should be evaluated as a measure of treatment effectiveness in future studies of diabetic eye disease.
