Purchase this article with an account.
Bianca Gerendas, Christian Simader, Gabor Gy Deak, Sonja Gudrun Prager, Jan Lammer, Sebastian M Waldstein, Michael Kundi, Ursula Schmidt-Erfurth; Morphological parameters relevant for visual and anatomic outcomes during anti-VEGF therapy of diabetic macular edema in the RESTORE trial. Invest. Ophthalmol. Vis. Sci. 2014;55(13):1791.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Identification of relevant morphologic factors in multimodal imaging during anti-VEGF therapy for diabetic macular edema (DME) and prediction of visual and anatomical outcomes.
In a subanalysis of a prospective randomized phase III clinical trial, images of 345 patients with DME were systematically analyzed. Patients were randomized to receive 0.5mg Ranibizumab (RZ), 0.5mg Ranibizumab plus laser (RZ+L) or laser alone (L). After an initial loading phase (LP) of 3 injections in the RZ(+L) arms and one laser treatment at baseline (BL) in the (RZ+)L arms, patients were treated as needed (PRN). After standardized image evaluation at the Vienna Reading Center parameters of optical coherence tomography (OCT), fluorescein angiography (FA) and color fundus (CF) images were correlated with best corrected visual acuity (BCVA).
At BL mean BCVA was 64±10.5 letters (RZ: 65, RZ+L: 63, L: 62); change in BCVA from BL to month 12 (M12) was 5.5±10.0 letters (RZ: 7.5, RZ+L: 7.7, L: 0.9). Mean central retinal thickness (CRT) at BL was 418±121µm (RZ: 428µm, RZ+L: 417µm, L: 409µm); change in CRT from BL to M12 was 105±123µm (RZ: 126µm, RZ+L: 126µm, L: 59µm). An overall trend for correlation between BCVA gain and CRT decrease was observed during the LP but lost afterwards. IRC at BL were associated with a lower BL BCVA in all arms but had no influence on BCVA values at M12; however, patients with IRC at BL had a larger BCVA gain in RZ which resulted in the same BCVA at M12 for groups with and without IRC at BL. The same groups showed significantly (p=0.036) different CRT values at M12: in the group with IRC at BL it was 317µm, in the group without IRC at BL it was 284µm. In RZ+L IRC of ≤380µm in height were continuously associated with significantly (p<0.001) better BCVA from BL to M12. SRF at BL was not associated with a worse BCVA at BL. However, patients with SRF at BL had a significantly (p=0.004) higher BCVA gain from BL to M12 in RZ which also resulted in higher final BCVA levels. No significant impact on BCVA and anatomical outcomes was found for parameters derived from FA and CF.
Morphologic evaluation of multimodal images may allow predicting functional and anatomical response to anti-VEGF therapy. In particular, active disease at BL (i.e. IRC and SRF) is associated with higher BCVA gain, while CRT alone cannot predict BCVA response.
This PDF is available to Subscribers Only