Perez Eli A, Scott Nathan L, Russell Jonathan F
Gavin Herbert Eye Institute, University of California, Irvine, Irvine, California; Department of Ophthalmology and Visual Sciences, Carver College of Medicine, University of Iowa, Iowa City, Iowa.
Shiley Eye Institute, University of California, San Diego, La Jolla, California.
Ophthalmology. 2024 Apr;131(4):412-421. doi: 10.1016/j.ophtha.2023.10.030. Epub 2023 Oct 29.
To independently validate the Retinal Detachment after Open-Globe Injury (RD-OGI) scoring system as a clinical prediction model for estimating the risk of retinal detachment (RD) after open-globe injury (OGI), and to characterize outcomes of RD stratified by RD-OGI risk category.
Retrospective, multi-institutional cohort study.
A validation cohort of 236 eyes with OGI managed at 2 academic institutions from 2017 through 2021 was compared with the predictions of the RD-OGI score model and to the derivation cohort of 893 eyes with OGI used to develop it.
RD-OGI scores were calculated, and patients were stratified into 3 risk groups (low, moderate, and high). The incidence of RD in the validation cohort was used to calculate performance metrics to evaluate predictive accuracy of the RD-OGI score.
Area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy of the RD-OGI score; anatomic success (retina attached or detached); and best-corrected visual acuity (BCVA) at last follow-up.
In the validation cohort, 92 of 236 eyes (39%) demonstrated RD after OGI. The RD incidence predicted by the RD-OGI score was correlated strongly with actual RD incidence in the validation cohort (Spearman ρ = 0.92; P < 0.001). The RD-OGI score demonstrated high discrimination with the validation cohort (AUC, 0.84; 95% confidence interval [CI], 0.79-0.89). An RD-OGI score of 2.5 was found to have 73% accuracy, 93% sensitivity, and 60% specificity for predicting no RD (RD-OGI score < 2.5) versus RD (RD-OGI score ≥ 2.5). Median BCVA at last follow-up for successfully reattached eyes was comparable for high-risk eyes (approximately 20/380) and moderate-risk eyes (approximately 20/300; P > 0.99), but both were significantly worse compared with low-risk eyes (approximately 20/25; P = 0.010).
The RD-OGI score reliably predicts the risk of developing RD after OGI. In eyes that develop RD, a higher RD-OGI score is correlated strongly with a greater risk for poor functional and anatomic outcomes. An RD-OGI score of 2.5 or more identifies eyes at substantial risk of RD and adverse outcomes and may serve as a useful cutoff for guiding referral to a vitreoretinal surgeon after primary globe closure.
FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
独立验证开放性眼球损伤后视网膜脱离（RD-OGI）评分系统作为评估开放性眼球损伤（OGI）后视网膜脱离（RD）风险的临床预测模型，并描述按RD-OGI风险类别分层的RD结局。
回顾性、多机构队列研究。
将2017年至2021年在2家学术机构接受治疗的236只OGI患眼的验证队列与RD-OGI评分模型的预测结果以及用于开发该模型的893只OGI患眼的推导队列进行比较。
计算RD-OGI评分，并将患者分为3个风险组（低、中、高）。使用验证队列中RD的发生率来计算性能指标，以评估RD-OGI评分的预测准确性。
RD-OGI评分的受试者工作特征曲线下面积（AUC）、敏感性、特异性和准确性；解剖学成功（视网膜附着或脱离）；以及最后一次随访时的最佳矫正视力（BCVA）。
在验证队列中，236只眼中有92只（39%）在OGI后发生了RD。RD-OGI评分预测的RD发生率与验证队列中的实际RD发生率密切相关（Spearman ρ = 0.92；P < 0.001）。RD-OGI评分在验证队列中显示出较高的区分度（AUC，0.84；95%置信区间[CI]，0.79-0.89）。发现RD-OGI评分为2.5时，预测无RD（RD-OGI评分<2.5）与RD（RD-OGI评分≥2.5）的准确性为73%，敏感性为93%，特异性为60%。成功复位的高风险眼和中风险眼在最后一次随访时的中位BCVA相当（约20/380），但与低风险眼相比均显著更差（约20/25；P = 0.010）。
RD-OGI评分可靠地预测了OGI后发生RD的风险。在发生RD的眼中，较高的RD-OGI评分与较差的功能和解剖学结局的更大风险密切相关。RD-OGI评分为2.5或更高可识别出有显著RD风险和不良结局的眼，并且可作为在初次眼球闭合后指导转诊至玻璃体视网膜外科医生的有用临界值。
在本文末尾的脚注和披露中可能会发现专有或商业披露信息。
