Abstract：
OBJECTIVE: To measure the visual outcomes, proliferative vitreoretinopathy (PVR) and retinectomy rates following primary rhegmatogenous retinal detachment (RRD) repair, comparing silicone oil (SO) and heavy SO (Densiron).
METHODS: Retrospective, continuous comparative study from January 2017 to May 2021 of all primary RRD. Multivariable linear (logMAR gain) and binary-logistic (PVR-C and retinectomy rate) regression models to compare tamponade were performed. Covariates included age, gender, ocular co-morbidities, high myopia, macula-status, giant-retinal-tear (GRT), pre-op vision, PVR-C, oil type, perfluorocarbon-use, combined scleral buckle/vitrectomy, combined phaco-vitrectomy, 360-degrees-endolaser and oil duration. Cases with trauma or less than six-month follow-up were excluded.
RESULTS: A total of 259 primary RD were analysed. There were 179 SO patients and 80 Densiron patients that had six-month primary re-detachment in 18 (10.1%) and 8 (10.0%) respectively (p = 1.000). No difference in logMAR gain was detected between tamponade choice on multivariable linear regression. Subsequent glaucoma surgery was 5 (2.8%) and 4 (5.0%) for SO and Densiron patients respectively (p = 0.464). On multivariate binary-logistic regression we found no difference in development of PVR-C between oil tamponades. However, SO had significantly higher subsequent retinectomy rate compared to Densiron (odds ratio 15.3, 95% CI 1.9-125.5, p = 0.011). Duration of oil tamponade was not linked to differences in logMAR gain, PVR-C formation or increased retinectomy rate.
CONCLUSIONS: We report no difference in primary anatomical success, number of further RRD surgeries, subsequent glaucoma surgery, visual outcomes, PVR-C between both tamponades on multivariable models. Densiron oil was found to be more retinectomy sparing relative to SO.
METHODS: Retrospective, continuous comparative study from January 2017 to May 2021 of all primary RRD. Multivariable linear (logMAR gain) and binary-logistic (PVR-C and retinectomy rate) regression models to compare tamponade were performed. Covariates included age, gender, ocular co-morbidities, high myopia, macula-status, giant-retinal-tear (GRT), pre-op vision, PVR-C, oil type, perfluorocarbon-use, combined scleral buckle/vitrectomy, combined phaco-vitrectomy, 360-degrees-endolaser and oil duration. Cases with trauma or less than six-month follow-up were excluded.
RESULTS: A total of 259 primary RD were analysed. There were 179 SO patients and 80 Densiron patients that had six-month primary re-detachment in 18 (10.1%) and 8 (10.0%) respectively (p = 1.000). No difference in logMAR gain was detected between tamponade choice on multivariable linear regression. Subsequent glaucoma surgery was 5 (2.8%) and 4 (5.0%) for SO and Densiron patients respectively (p = 0.464). On multivariate binary-logistic regression we found no difference in development of PVR-C between oil tamponades. However, SO had significantly higher subsequent retinectomy rate compared to Densiron (odds ratio 15.3, 95% CI 1.9-125.5, p = 0.011). Duration of oil tamponade was not linked to differences in logMAR gain, PVR-C formation or increased retinectomy rate.
CONCLUSIONS: We report no difference in primary anatomical success, number of further RRD surgeries, subsequent glaucoma surgery, visual outcomes, PVR-C between both tamponades on multivariable models. Densiron oil was found to be more retinectomy sparing relative to SO.
摘要：
目的：为了测量视觉结果，增生性玻璃体视网膜病变（PVR）和原发性孔源性视网膜脱离（RRD）修复后的视网膜切除术率，比较硅油(SO)和重质SO(Densiron)。
方法：回顾性，2017年1月至2021年5月对所有主要RRD进行连续比较研究。使用多变量线性（logMAR增益）和二元逻辑（PVR-C和视网膜切除术率）回归模型来比较填塞。协变量包括年龄，性别,眼部合并症，高度近视,黄斑状态,巨大视网膜撕裂(GRT),术前视野,PVR-C,油类,全氟化碳用途,联合巩膜扣/玻璃体切除术，超声乳化术联合玻璃体切除术，360度-内激光和油持续时间。排除外伤或随访少于6个月的病例。
结果：共分析了259例原发性RD。有179例SO患者和80例Densiron患者分别有18例（10.1％）和8例（10.0％）的六个月原发性再脱离（p=1.000）。在多变量线性回归中，填塞选择之间的logMAR增益没有差异。SO和Densiron患者的后续青光眼手术分别为5（2.8％）和4（5.0％）（p=0.464）。在多元二元逻辑回归中，我们发现油塞之间PVR-C的发展没有差异。然而,与Densiron相比，SO的后续视网膜切除术率明显更高（比值比15.3，95％CI1.9-125.5，p=0.011）。油填塞的持续时间与logMAR增益的差异无关，PVR-C形成或增加视网膜切除率。
结论：我们报告的主要解剖学成功没有差异，进一步RRD手术的数量，随后的青光眼手术，视觉结果，多变量模型上两种填塞物之间的PVR-C。发现Densiron油相对于SO更节省了视网膜切除术。
方法：回顾性，2017年1月至2021年5月对所有主要RRD进行连续比较研究。使用多变量线性（logMAR增益）和二元逻辑（PVR-C和视网膜切除术率）回归模型来比较填塞。协变量包括年龄，性别,眼部合并症，高度近视,黄斑状态,巨大视网膜撕裂(GRT),术前视野,PVR-C,油类,全氟化碳用途,联合巩膜扣/玻璃体切除术，超声乳化术联合玻璃体切除术，360度-内激光和油持续时间。排除外伤或随访少于6个月的病例。
结果：共分析了259例原发性RD。有179例SO患者和80例Densiron患者分别有18例（10.1％）和8例（10.0％）的六个月原发性再脱离（p=1.000）。在多变量线性回归中，填塞选择之间的logMAR增益没有差异。SO和Densiron患者的后续青光眼手术分别为5（2.8％）和4（5.0％）（p=0.464）。在多元二元逻辑回归中，我们发现油塞之间PVR-C的发展没有差异。然而,与Densiron相比，SO的后续视网膜切除术率明显更高（比值比15.3，95％CI1.9-125.5，p=0.011）。油填塞的持续时间与logMAR增益的差异无关，PVR-C形成或增加视网膜切除率。
结论：我们报告的主要解剖学成功没有差异，进一步RRD手术的数量，随后的青光眼手术，视觉结果，多变量模型上两种填塞物之间的PVR-C。发现Densiron油相对于SO更节省了视网膜切除术。
