Primary vitrectomy or intravitreal antibiotics followed by early vitrectomy for acute endophthalmitis: A prospective observational study

Abstract

Background/aims: To compare outcomes of acute endophthalmitis (EO) managed with either primary vitrectomy (PV) or primary intravitreal antibiotics (vancomycin and ceftazidime) followed by early vitrectomy (PIAEV) combined with polymerase chain reaction (PCR)-based diagnostics.

Methods: This was a prospective, comparative observational study of acute EO cases admitted to a regional vitreoretinal service over 18 months. Depending on whether immediate vitrectomy (within 6 h) was achievable, the EO cases were treated with either (1) PV or (2) PIAEV. Microbiology samples were collected either (A) before or (B) after administration of intravitreal antibiotics. The samples were analysed with broad-range 16S PCR and culture.

Results: The study included 41 EO cases. There were 19 post-injection EO, 18 post-cataract EO, three post-vitrectomy EO, and one blebitis-related EO. Fifteen of 19 PV cases and 15 of 21 PIAEV had a clinically meaningful improvement in best-corrected visual acuity (BCVA) of at least 15 letters at 3 months (p = 0.58). One patient was lost to follow-up. Twenty-three cases were culture- and PCR-positive, and seven additional cases were culture-negative but PCR-positive (p = 0.02). PCR increased the diagnostic yield for samples collected both before and after administration of intravitreal antibiotics.

Conclusion: Primary vitrectomy or PIAEV allowed for vitrectomy for all cases of acute EO in a large region. Most eyes in both groups achieved a clinically meaningful improvement in BCVA. By combining culture with PCR in connection with the vitrectomy procedure, intravitreal antibiotics could be injected before microbiological sampling, thereby improving the door-to-treatment time without sacrificing microbial identification.

Keywords: PCR; antibiotics; culture; endophthalmitis; intravitreal; vitrectomy.

FIGURE 1

Visual function at admission and at 3 months. One patient was lost to follow‐up.

FIGURE 2

Proportions of positive and negative microbiological results for samples collected before (28 cases) and after (13 cases) administration of intravitreal antibiotics. Notably, no cases were PCR‐negative, but culture‐positive.