Injury-to-admission interval and preoperative endophthalmitis after open-globe injury: a matched case–control study

Delayed presentation after open-globe injury (OGI) is commonly defined using a fixed binary cutoff, an approach that may misclassify a continuous timing exposure, obscure its dose–response pattern, and make conclusions dependent on the selected threshold. We evaluated the injury-to-admission interval as a graded exposure in relation to preoperative endophthalmitis. This retrospective matched case–control study included 344 patients with OGI treated surgically at a tertiary eye hospital in central China between 2017 and 2024: 172 patients with preoperative endophthalmitis and 172 propensity-score-matched controls. The injury-to-admission interval extended from the recorded injury time to emergency registration on arrival at the study hospital. It was modeled primarily as a log2-transformed continuous exposure and additionally examined using prespecified multicategory and dichotomous definitions and restricted cubic splines. Multiple-imputation, complete-case, and matched-pair analyses assessed robustness after adjustment for demographic, systemic, and ocular-trauma characteristics. The median injury-to-admission interval was 48.0 h (IQR, 24.0–96.0) in patients with preoperative endophthalmitis and 8.0 h (IQR, 5.0–12.5) in controls ( P < 0.001). In the primary model, each doubling of the interval was associated with higher odds of preoperative endophthalmitis (OR, 2.30; 95% CI, 1.71–3.10; P < 0.001). The direction and magnitude of the association were consistent across sensitivity analyses. For each doubling of the injury-to-admission interval, the adjusted odds of preoperative endophthalmitis after OGI were 2.30 times as high. Earlier admission was associated with lower odds of endophthalmitis after OGI, and no single time cutoff should be interpreted as defining a safe interval.