Postoperative endophthalmitis treatment with antibiotics associated or not with pars plana vitrectomy: a randomized clinical trial

Abstract

Background: Postoperative endophthalmitis (PSE) is a severe ocular complication that can lead to irreversible vision loss or even globe atrophy. The Endophthalmitis Vitrectomy Study (EVS) historically guided PSE management but is increasingly questioned due to advances in pars plana vitrectomy (PPV) techniques and its narrow focus on cataract surgery. This study aimed to compare PPV followed by intravitreal antibiotic injection at the end of surgery (PPV + IVAIES) with intravitreal antibiotic injection alone (IVAI) in managing PSE.

Methods: This randomized clinical trial included 35 pseudophakic patients with PSE following cataract extraction, anti-vascular endothelial growth factor (anti-VEGF) injections, or glaucoma surgeries. Participants were randomized to receive either PPV + IVAIES (n = 12) or IVAI (n = 23). Best-corrected visual acuity (BCVA) was assessed at baseline and days 7, 30, 60, and 90 post-intervention. Clinical worsening, defined as lack of improvement or progression of symptoms within 48-72 h, guided retreatment protocols. Group A (PPV + IVAIES) received repeat IVAI if required, while Group B (IVAI) underwent delayed PPV with repeat IVAI. Statistical significance was assessed using repeated measures ANOVA and logistic regression.

Results: Both groups showed significant BCVA improvement (p < 0.001). PPV + IVAIES resulted in faster recovery, with superior BCVA at day 7 (p = 0.019) and day 30 (p = 0.041). Retreatment was required in 39.1% of the IVAI group but not in the PPV + IVAIES group (p = 0.015). Subgroup analysis indicated a trend toward better early outcomes with early PPV (p = 0.029).

Conclusions: Early PPV + IVAIES provides faster visual recovery and reduces retreatment rates compared to IVAI alone. Multicenter studies are warranted to confirm these findings and refine clinical guidelines. Trial registration ClinicalTrials.gov identifier: NCT04192994.

Keywords: Endophthalmitis; Endophthalmitis vitrectomy study; Intravitreal antibiotic injection; Pars plana vitrectomy.

Fig. 1

Flowchart of the randomized clinical trial evaluating the management of postoperative endophthalmitis (PSE). A total of 56 patients were assessed for eligibility, with 5 excluded for not meeting inclusion criteria or declining participation. Fifty-one patients were randomized into two groups: Group A (Intravitreal Antibiotic Injection—IVAI, n = 25) and Group B (Pars Plana Vitrectomy with Intravitreal Antibiotic Injection at the End of Surgery—PPV + IVAIES, n = 26). Following withdrawals during follow-up, the final analysis included 12 patients in Group A and 23 in Group B

Fig. 2

Changes in visual acuity over time in patients treated with intravitreal antibiotic injection (IVAI) versus pars plana vitrectomy with intravitreal antibiotic injection at the end of surgery (PPV + IVAIES). The y-axis represents best-corrected visual acuity (BCVA) in logMAR, where lower values indicate better vision. The x-axis indicates follow-up intervals (baseline—D0, and days 7, 30, 60, and 90 post-intervention). The dashed orange line represents the PPV + IVAIES group, while the solid yellow line represents the IVAI group. Statistically significant differences in BCVA between groups were observed at days 7 (p = 0.019) and 30 (p = 0.041), but not at later intervals. Error bars represent standard deviations

Fig. 3

Comparison of visual acuity improvement over time between early pars plana vitrectomy (PPV) and delayed PPV in the management of postoperative endophthalmitis (PSE). The y-axis shows best-corrected visual acuity (BCVA) in logMAR, where lower values indicate better vision. The x-axis represents follow-up intervals (baseline—D0, and days 7, 30, 60, and 90 post-intervention). The dashed yellow line represents early PPV, while the solid orange line represents delayed PPV. A statistically significant difference in BCVA was observed at day 7 (p = 0.029), favoring early PPV. Error bars indicate standard deviations

Fig. 4

Visual acuity improvement over time in patients treated with early pars plana vitrectomy (PPV) compared to intravitreal antibiotic injection (IVAI) alone. The y-axis represents best-corrected visual acuity (BCVA) in logMAR, where lower values indicate better vision. The x-axis represents follow-up intervals (baseline—D0, and days 7, 30, 60, and 90 post-intervention). The dashed yellow line corresponds to the early PPV group, while the solid orange line represents the IVAI group. No statistically significant differences in BCVA were observed at any time point (p > 0.05). Error bars indicate standard deviations