Related Experiment Video
Updated: Sep 30, 2026
Intravitreal Injection and Quantitation of Infection Parameters in a Mouse Model of Bacterial Endophthalmitis
Published on: February 6, 2021
Early Vitrectomy Within Three Days of Symptom Onset is Associated with Greater Visual Recovery in Exogenous
Chia-An Hsu1, Meng-Jou Chen1, Chang-Chi Weng1,2
1Department of Ophthalmology, Taipei Veterans General Hospital, Taipei, Taiwan, ROC.
Purpose:
To evaluate whether pars plana vitrectomy (PPV) performed within 72 hours of symptom onset is associated with greater visual recovery compared with initial tap-and-inject therapy in exogenous infectious endophthalmitis.
Methods:
Retrospective cohort study of consecutive patients with exogenous infectious endophthalmitis at a tertiary referral centre (2009-2018). Eyes were classified according to whether PPV was performed within 72 hours of symptom onset, irrespective of preceding intravitreal antibiotics injection; eye without PPV during this window comprised the injection-first comparison group. The primary endpoint was logMAR best corrected visual acuity (BCVA) change at three months; secondary endpoints were absolute BCVA and BCVA change at one year. Multivariable linear regression identified independent outcome predictors.
Results:
Eighty-four eyes were included (20 early PPV; 64 tap-and-inject). Despite worse baseline BCVA in the early PPV group (2.18 ± 0.36 vs. 1.62 ± 0.82 logMAR), three-month BCVA improvement was significantly greater (-1.28 ± 0.78 vs. -0.64 ± 0.89 logMAR; p = 0.01). The benefit persisted at one year (-1.51 ± 0.77 vs. -0.79 ± 0.96 logMAR; p = 0.012). Multivariable regression confirmed early PPV as an independent predictor of better three-month BCVA change (β = -0.647; p = 0.003) and greater one-year improvement (β = -0.578; p = 0.022).
Conclusion:
Early vitrectomy within a 72-hour window is associated with significantly greater visual recovery in exogenous infectious endophthalmitis, independent of presenting BCVA. Both presenting visual acuity and time from symptom onset should jointly inform treatment decisions, supporting a management approach that might extend beyond conventional EVS criteria.
