[Efficacy and safety of intravitreal dexamethasone implant in diabetic macular edema patients with and without prior vitrectomy: a retrospective cohort study].

To evaluate the effectiveness and safety of intravitreal dexamethasone implant (IDI) in diabetic macular edema (DME) patients with and without prior vitrectomy. A retrospective cohort study was conducted on DME patients who received IDI treatment at the Aier Eye Hospital, Beijing from March 2018 to August 2020. Patients were divided into two groups according to whether they had undergone vitrectomy or not. Clinical and follow-up data, including best-corrected visual acuity (BCVA), intraocular pressure (IOP), central macular thickness (CMT), occurrence of ocular and systemic complications, and time to DME recurrence and retreatment, were collected before and after IDI injection at 15 days, 1, 2, 3, and 6 months. Statistical analyses were performed using -test, Mann-Whitney -test, test or Fisher's exact test, and generalized estimating equations. Thirty-six patients (41 eyes) were included, with 19 patients (21 eyes) in the vitrectomy group and 17 patients (20 eyes) in the non-vitrectomy group. Compared with baseline, BCVA of eyes in the vitrectomy group was significantly improved at 15 days after IDI injection, with values of 1.00 (0.52, 1.31) and 0.61 (0.30, 1.00), respectively (=-2.10, =0.036); BCVA of eyes in the non-vitrectomy group was significantly improved at 1 month after IDI injection, with values of 0.76 (0.60, 1.35) and 0.52 (0.10, 0.70), respectively (=-2.24, =0.025). Compared with baseline, CMT of eyes in both groups was significantly reduced at all follow-up time points after 15 days of IDI injection (all <0.05). In the vitrectomy group, CMT before and 15 days after injection were 487 (438, 661) μm and 389 (340, 553) μm, respectively (=-3.45, <0.001); in the non-vitrectomy group, CMT before and 15 days after injection were 486 (410, 641) μm and 323 (290, 396) μm, respectively (=-4.07, <0.001). There were no statistically significant differences in BCVA and CMT between the two groups at all follow-up time points (all >0.05). The time to DME recurrence was 3.0 (3.0, 4.0) months in the vitrectomy group and 5.0 (4.0, 5.0) months in the non-vitrectomy group, with no significant difference between the two groups (=0.675). Four eyes (19.0%) in the vitrectomy group and three eyes (15.0%) in the non-vitrectomy group had high IOP, with no significant difference (=0.529). No severe ocular or systemic complications were observed in any patients. IDI treatment is safe and effective in DME patients with and without prior vitrectomy, with similar efficacy, but with faster onset of action in patients with prior vitrectomy. There was no significant difference in DME recurrence within 6 months after IDI injection between the two groups.