硅油取出手术后视网膜再脱离的发生原因和影响因素

Causes and risk factors of recurrent retinal detachment after silicone oil removal

目的探讨复杂视网膜脱离眼硅油取出手术后视网膜再脱离（RRD）的影响因素及发生原因。方法回顾性研究。选择行玻璃体切割联合硅油填充手术的连续患者455例458只眼纳入研究。对所有患者行玻璃体切割手术，手术中根据眼内情况酌情行重水、剥膜、视网膜切开或部分切割、眼内激光光凝或冷冻、气液交换或直接油液交换等操作。对伴多发裂孔、陈旧性视网膜脱离、增生及牵拉病变较严重者行环扎手术98只眼。手术完毕时玻璃体腔内填充硅油。硅油取出手术中眼底检查发现有需处理的视网膜前膜者，行切断、剥膜或切除及360。预防性激光光凝治疗等操作，手术中发现裂孔或可疑裂孔者行眼内激光光凝或冷冻治疗。硅油取出手术后1周内及随诊期间均采用与手术前相同的设备和方法行视力、眼压、裂隙灯显微镜、检眼镜等检查。依据硅油取出手术后有无RRD将患眼分为复位组、再脱离组，分别为419、39只眼。对患者年龄、眼轴长度、玻璃体切割手术前最小分辨角对数（LogMAR）最佳矫正视力（BCVA）和眼压、硅油取出手术前LogMARBCVA和眼压、视网膜裂孔数目、硅油填充时间、随访时间、硅油取出手术后眼压和视力以及玻璃体切割手术和硅油取出手术中相关因素进行记录。统计年龄、性别、高度近视、巨大裂孔、下方裂孔、黄斑裂孔、无晶状体眼、增生性玻璃体视网膜病变（PVR）C3级及以上、既往视网膜脱离手术失败史、360。预防性激光光凝、联合环扎、角膜穿刺取硅油与硅油取出手术后发生RRD的关系。计算年龄〈40岁、性别等因素的比值比（OR）及其95％可信区间（CI）。将高度近视眼、联合巩膜环扎及经角膜穿刺硅油取出纳人多元回归方程。结果硅油取出手术后患眼平均LogMARBCVA为0．86±0．63，复位组、再脱离组平均LogMARBCVA分别为0．82±0．59、0．99±0．70，两组平均LogMARBCVA比较，差异无统计学意义（F=1．559，P〉0．05）。复位组和再脱离组高度近视眼分别为116、22只眼，分别占为27．7％、56．4％，差异有统计学意义（X2=13．984，P〈0．01）。玻璃体切割手术中联合环扎手术患者中发生RRD3只眼，占3．1O，未行环扎手术患者发生RRD36只眼，占10．0％，两者RRD发生率比较，差异有统计学意义（X2=4．761，P〈0．05）。手术后RRD的发生率与手术前PVR程度、既往视网膜脱离手术失败史、无晶状体眼以及预防性视网膜激光光凝等因素均无关性（1_626、1．699、1．986、0．709，95％CI：0．836～3．162、0．832～3．658、0．921～4．279、0．268～1．875，P〉0．05）。与高度近视、联合环扎有相关性（OR一3．380、0．284，95％CI：1．733～6．595、0．086～0．944，P〈0．05）。通过角膜穿刺取硅油的风险无统计学意义（OR一2．119，95％CI：1．043～4．306，P〉0．05）。硅油取出手术后RRD的发生率为8．5％，其中35．9％源于新裂孔的形成，5．1％源于PVR，69．2％与新裂孔相关，51．3％与PVR相关。结论高度近视眼是硅油取出手术后发生RRD的独立危险因素；联合巩膜环扎手术是硅油取出手术后发生RRD的保护性因素。对于行玻璃体切割及硅油填充手术后视网膜稳定复位眼而言，硅油取出手术后发生RRD的主要原因是新裂孔的形成。PVR可能是RRD后的继发改变。

Objective To investigate the main causes and risk factors of recurrent retinal detachment （RRD） after silicone oil removal （SOR） in eyes with complex retinal detachment. Methods It was a retrospective case series study. A total of 458 eyes of 455 consecutive patients who underwent pars plana vitrectomy with silicone oil tamponade were recruited in this study. All patients underwent vitrectomy operation. Additionally, they were given heavy water, membrane peeling, retinotomy or partial cutting, intraocular laser photocoagulation or frozen, gas-liquid exchange or direct oil exchange operation accordingly. Ninety-eight eyes with multiple holes, old retinal detachment, hyperplasia and serious traction lesions underwent scleral buckling surgery simultaneously. Infravitreal silicone oil was padded at the end of operation. Cutting, stripping or resection and 360% preventive laser photocoagulation were applied while the epiretinal membrane was found and need treatment during SOR. Holes or suspicious hiatus underwent intraocular laser photocoagulation or cryotherapy during the operation. One week after SOR and during follow-up, the visual acuity, intraocular pressure （IOP）, slit lamp microscope, and ophthalmoscope examination were examined with the same technique and methods as preoperation. The eyes were divide into two groups based on the attachment status of retina after SOR, which were reattached group （419 eyes） and redetached group （39 eyes） respectively. The following data were recorded： the age of patients, ocular axial length, logarithm of minimum angle of resolution （logMAR） best corrected visual acuity （BCVA） and IOP be{ore vitrectomy operation and before and after SOR, the number of retinal breaks, the duration of silicone oil filling, the duration of follow-up, and the related %actors during vitreetomy operation and SOR. The relation of age, sex, high myopia, the size and location of holes, aphakic eye, proliferative vitreoretinopathy （PVR） C3 level and above, previous history of failed retinal detachment operation, 360% preventive laser photocoagulation, assistant scleral buckling surgery, SOR via corneal puncture to RRD after SOR were analyzed. Odds ratio （OR） and its 95% confidence interval （CI） were calculated for the age 〈40 years old and gender. High myopia, assistant scleral buckling surgery and SOR via corneal puncture were further analyzed by multiple regression equation. Results After SOR operation, the total average logMAR BCVA was 0.86 % 0.63. The average logMAR BCVA was 0.82 i- 0.59 and 0.99 % 0.70 respectively for the reattached and redetached groups, which was not statistically different （F= 1. 559, P〉O. 05）. The number of high myopia eyes in the reattached and redetached groups were 116 and 22 eyes, respectively, accounted for 27.7% and 56.40%, and the difference was statistically significant （X2 =13. 984, P〈0.01）. Three eyes underwent vitrectomy with scleral buckling occured RRD, accounting for 3.1% % while 36 eyes underwent vitrectomy without scleral buckling occured RRD, accounting for 10.0%. The incidence of RRD between them was statistically significant （2 =4. 761, P〈0.05）. The incidence of RRD was not retated to the PVR levels before the operation, previous history of failed retinal detachment operation, aphakic eye and preventive laser photocoagulation （OR= 1. 626, 1. 699, 1. 986, 0. 709; 950//00 CI% O. 836 - 3. 162, 0. 832 - 3.658, 0.921-4.279, 0.268 1.875; P〉0.05） . RRD had a close relation with high myopia and assistant scleral buckling surgery （OR=3. 380, 0. 284; 95%CI： 1. 733 - 6. 595, 0. 086 - 0. 944% P〈0.05）. The raise of risk derived from SOR via corneal puncture had no statistical significance （OR%2. 119% 95%CI： 1. 043 - 4. 306% P〉0.05）. The incidence of RRD after SOR was 8.5% of which, 35.9% originated from new breaks and 69.20% were related to new breaks, in contrast, only 5.1% originated from PVR but 51.3% were related to PVR. Conclusions High myopia is an independent prognostic risk factor of RRD after SOR. Combined scleral buckling surgery is a protective factor of RRD after SOR. To the well reattached eyes before SOR, the new breaks seems to be the main cause of RRD, wheras PVR was probably a secondary phenomenon.