〔Abstract〕 Objective To analyze the risk factors for endophthalmitis requiring evisceration or enucleation. Methods The charts of 121 eyes of 121 inpatients with endophthalmitis were retrospectively reviewed, and the group that required evisceration or enucleation(24 patients) with those that received salvaging therapies(97 patients)were compared. Age, sex, medication history, past medical history, clinical manifestation, Leukocyte counts and treatment progression were retrospectively analyzed.Results Twenty four eyes(19. 8%)underwent enucleation or evisceration. The proportion of corneal ulcerative endophthalmitis( 33. 3%) and endogenous endophthalmitis(25. 0%) in evisceration or enucleation group was greater than that in the salvaging group(1. 0%, 4. 1%)(P<0. 001). The group of eviscerated or enucleated eyes was older(P<0. 05], had poorer initial visual acuity [(2. 9± 0. 2) LogMARvs(2. 3 ± 0. 5) LogMAR,P<0. 001], had longer duration before intervention(15. 8 dvs4. 6 d,P<0. 05), and had more Leukocyte counts [(12. 8 ± 5. 6) × 109/Lvs(9. 1 ± 3. 3) × 109/L,P<0. 005].With Logistic regression analysis, corneal ulcer, endogenous endophthalmitis, initial vision, leukocyte counts, duration before intervention were the risk factor for evisceration or enucleation(OR= 343. 283,OR= 22. 608,OR=1 920. 384,OR= 1. 341,OR= 1. 167).Conclusion The most common cause of evisceration or enucleation caused by infectious endophthalmitis are infections from corneal ulcer and from endogenous source. The risk factors for endophthalmitis requiring evisceration or enucleation would be considered to be corneal ulcer endophthalmitis,endogenous endophthalmitis, initial vision, leukocyte counts, and duration before intervention.