Baruch D. Kuppermann, MD, PhD (Irvine, CA),* Connie Chou, PhD (Irvine, CA),* David V. Weinberg, MD (Milwaukee, WI),* Scott M. Whitcup, MD (Irvine, CA),* Julia A. Haller, MD (Baltimore, MD),* Mark S. Blumenkranz, MD (Stanford, CA),* George A. Williams, MD (Royal Oak, MI)*


To evaluate the response of different patterns of persistent diabetic macular edema (DME) unresponsive to prior medical or laser therapy to treatment with a novel intravitreal sustained-release dexamethasone posterior-segment drug-delivery system (dexamethasone DDS).


Prospective, multicenter, masked, randomized, controlled clinical study. Eyes with central macular edema due to diabetic retinopathy unresponsive to prior medical or laser therapy for at least 3 months with a best-corrected ETDRS visual acuity (BCVA) of 20/40 to 20/200 were randomized to observation or treatment with 350 μg or 700 μg dexamethasone DDS (n=57
per group). At baseline, DME was characterized as focal (n=96); cystoid (n=66); diffuse (n=94); or combined cystoid and diffuse (cystoid-diffuse; n=51) based on the clinical exam. (Note: patients could exhibit more than one pattern of DME). Outcome measures included BCVA and central retinal thickness measured by OCT.


Efficacy was seen across all patterns of DME, but there was a significant treatment-by-DME pattern interaction (P=.038) favoring the 700 μg dose when cystoid-diffuse DME was present. At day 90, an improvement of =10 letters was seen in 42% (11/26) of cystoid-diffuse eyes in the 700μg group (P=.001 vs observation), 17.2% (5/29) in the 350 μg group (P=.123 vs observation), and 4% (1/25) of eyes in the observation group. Other patterns of ME showed =10 letter improvement in 32-39% of eyes. (P=.04 to .004 vs observation) in the 700 μg group and in 18-24% of eyes in the 350 μg group (P=.283 to .165 vs observation). Changes in central retinal thickness for all DME patterns ranged from -132 to -183 μm in the 700 μg group (P=.005 to <.001 vs observation), and from +4.6 to -45 μm in the 350 μg group (P=.589 to .061 vs observation) while the among-group treatment-by-DME pattern interaction was borderline statistically significant (700 μg group only) at P=.058.


Despite failing prior therapy, eyes with all patterns of DME had significant improvements in BCVA and central retinal thickness following treatment with 700 μg dexamethasone in a novel intravitreal sustainedrelease posterior-segment drug-delivery system. Eyes with combined cystoid-diffuse DME showed a more favorable response than other patterns of
* Financial interest disclosed