ObjectiveTo investigate the risk factors of postoperative vitreous hemorrhage after minimal vitrectomy without endotamponade for proliferative diabetic retinopathy (PDR).MethodsFrom June 2015 to June 2017, 103 eyes of 103 patients with PDR diagnosed and underwent minimalvitrectomy in Henan Provincial People's Hospital were enrolled in the study. There were 58 males and 45 females, with the average age of 58.37±10.14 years and diabetes duration of 8.7±7.2 years. Baseline systemic parameters including sex, age, diabetes duration, hypertension, HbA1c, creatinine, whether received anticoagulants, ocular parameters including whether combined with vitreous hemorrhage, whether finished panretinal photocoagulation (PRP), whether received treatment of anti-VEGF, whether combined with iris neovascularization (NVI), lens status preoperatively, whether hypotony postoperatively and intraoperative parameters including whether disc neovascularization (NVD) bleeding, whether fibrovascular membrane (FVM) residual, laser points, whether combined with cataract phacoemulsification were identified by multivariate logistic regression analysis.ResultsTwenty-nine of 103 eyes (28.15%) developed PVH in 1 day to 6 months after surgery, with self absorption of 18 eyes and reoperation of 11 eyes. Univariate analysis showed there were significant differences in age (t=2.124, P=0.036), anti-VEGF(χ2=7.105, P=0.008), NVD bleeding (χ2=10.158, P=0.001) and FVM residual(χ2=8.445, P=0.004) between patients with and without postoperative vitreous hemorrhage. Sex (χ2=0.021, P=0.884), diabetes duration (t=0.87, P=0.386), hypertension (χ2=2.004, P=0.157), HbA1c (t=1.211, P=0.229), creatinine (t=0.851, P=0.397), preoperative oral anticoagulants (χ2=0.985, P=0.321), preoperative vitreous hemorrhage (χ2=0.369, P=0.544), PRP (χ2=1.122, P=0.727), NVI (χ2=2.635, P=0.105), lens status (χ2=0.172, P=0.679), hypotony postoperatively (χ2=1.503, P=0.220), laser points (χ2=1.391, P=0.238) and combined phacoemulsification surgery (χ2=0.458, P=0.499) were not associated with PVH. Multivariate logistic regression analysis revealed the more PVH appeared in younger (OR=1.065, P=0.009) and NVD bleeding (OR=6.048, P=0.001) patients.ConclusionYounger age and NVD bleeding are the important risk factors for PVH after minimal vitrectomy without endotamponade in PDR.
ObjectiveTo observe the effect of preoperative intravitreal ranibizumab injection (IVR) on the operation duration of vitrectomy and postoperative vision for the treatment of proliferative diabetic retinopathy (PDR). MethodsA prospective study was carried out with the 90 PDR patients (90 eyes) who underwent vitrectomy. The 90 patients(90 eyes)were assigned to the vitrectomy only group(43 eyes) and the IVR combined with vitrectomy group (47 eyes). The IVR was performed 5-13 days prior to vitrectomy in the IVR combined with vitrectomy group. There were 15 eyes with fibrous proliferation PDR (FPDR), 16 eyes with advanced PDR (APDR) without involving the macular and 16 eyes with APDR involving the macular in the vitrectomy only group. There were 14 eyes with FPDR, 15 eyes with APDR without involving the macular and 14 eyes with APDR involving the macular patients in the IVR combined with vitrectomy group. All the eyes in the two groups were regularly operated by the same doctor to complete the vitrectomy. The start and end time of vitrectomy were recorded. The average follow-up time was 10 months. The changes of best corrected visual acuity (BCVA) before and 1, 3 and 6 months after surgery were compared between the two groups. ResultsThe duration of operation of the FPDR type (t=-8.300) and the APDR involving the macular type (t=-2.418) in the IVR combined with vitrectomy group was shorter than vitrectomy only group (P < 0.05). The comparison of duration of operation of the APDR without involving the macular type in the two groups has no statistically significant difference (t=-1.685, P > 0.05). At 1 month after surgery, the comparison of BCVA of the IVR combined vitrectomy group and the vitrectomy only group in APDR involving the macular type has no statistically significant difference (t=0.126, P > 0.05). At 3, 6 months after surgery, the BCVA of the IVR combined vitrectomy group in APDR involving the macular type was significantly better than the BCVA of the vitrectomy only group (t=8.014, 7.808; P < 0.05). At 1, 3, and 6 months after surgery, the BCVA of the IVR combined vitrectomy group in FPDR type (t=3.809, 1.831, 0.600) and APDR without involving the macular type (t=0.003, 1.092, 3.931) compared with pre-treatment, the difference were not statistically significant (P > 0.05); the BCVA in APDR without involving the macular type compared with pre-treatment, the difference was distinctly statistically significant (t=2.940, 4.162, 6.446; P < 0.05); the BCVA in APDR involving the macular type (t=0.953, 1.682, 1.835) compared with pre-treatment, the difference were not statistically significant (P > 0.05). ConclusionPreoperative IVR of PDR can shorten the operation duration and improve the BCVA of APDR involving the macular type.
ObjectiveTo investigate the risk factors for neovascular glaucoma (NVG) after vitrectomy in proliferative diabetic retinopathy (PDR) patients. MethodsThree hundred and one patients (301 eyes) with PDR who underwent vitrectomy between January 2008 and December 2013 in our hospital were retrospectively evaluated. Risk factors for NVG after vitrectomy were identified by multivariate Logistic regression analysis. ResultsTwelve of 301 patients (4.0%) developed postoperative NVG in 2 to 18 months after vitrectomy. The incidence of postoperative NVG peaked in 2 to 6 months after vitrectomy (7 eyes, 58.3%). Logistic regression analysis showed that postoperative retinal detachment was a significant risk factor for postoperative NVG in eyes with PDR (P < 0.001). Eyes with postoperative retinal detachment were more likely to develop NVG after vitrectomy than those without postoperative retinal detachment (OR=17.826). Gender, age, duration of diabetes, preoperative serum creatinine levels, glycated hemoglobin levels, preoperative intraocular pressure, preoperative lens status, combined phacoemulsification surgery and tamponade were not associated with postoperative NVG (P > 0.05). ConclusionPostoperative retinal detachment is a major risk factor for NVG after vitrectomy in PDR.
Objective To investigate the risk factors associated with neovascular glaucoma (NVG) after pars plana vitrectomy (PPV) in eyes with proliferative diabetic retinopathy (PDR). Methods Retrospective study. One hundred and thirty-seven patients (137 eyes) with PDR who underwent PPV were recruited. There were 85 males and 52 females. The average age was (60.1±8.8) years old. The duration of diabetes was (10.2±3.6) years. There were 49 patients with ipsilateral carotid artery stenosis. Fifty-three eyes underwent intravitreal ranibizumab or conbercept injection before PPV. All eyes were treated with 23G standard three-port PPV. The average follow-up time after PPV was 11.5 months. Fundus fluorescein angiography (FFA) was conducted in postoperative 4-6 weeks to observe non-perfused retinal areas. Risk factors, such as ipsilateral carotid artery stenosis, the presence of non-perfusion in retina after PPV and the application of anti-vascular endothelial growth factor (VEGF) drugs before PPV, were identified by logistic regression. Results Twenty of 137 patients (14.6%) developed postoperative NVG after PPV. Ipsilateral carotid artery stenosis [odds ratio (OR) =5.048, 95% confidence interval (CI) 2.057-12.389,P=0.000] and the presence of non-perfusion in retina after PPV (OR=4.274, 95%CI 1.426-12.809,P=0.009) were significant risk factors for postoperative NVG, while the application of anti-VEGF drugs was not (OR=1.426, 95%CI 0.463-4.395,P=0.536). But the time from PPV to the onset of NVG varies significantly between the two groups of injection of anti-VEGF drugs or not (t=−4.370,P=0.000). Conclusions Risk factors associated with NVG after PPV in eyes with PDR included ipsilateral carotid artery stenosis and the presence of non-perfusion in retina after PPV. The application of anti-VEGF drugs before PPV can delay the onset of NVG in PDR eyes after vitrectomy.
Objective To analyze the risk factors of no light perception (NLP) after vitreoretinal surgery for proliferative diabetic retinopathy (PDR). Methods Retrospectively analyzed the follow-up data of 882 patients (1000 eyes) with PDR who had undergone vitreoretinal surgery. The standard of NLP was: in a darkroom, one eye was covered, and the other one could not catch the candlelight 30 cm in front of the eye. The number of eyes with NLP was counted and the clinical data of the eyes with or without NLP were analyzed and compared. chi;2 test was used to analyze the risk factors of NLP. Results In these 1000 eyes with PDR,the postoperative visual acuity was NLP in 22 eyes (2.2%) and light perception in 978 eyes (97.8%). Comparing with the patients with light perception, the patients with NLP had severer disease condition, including ante-operative neovascular glaucoma (NVG)(36.4%), tension combined with retinal detachment 50%, and a need for lens excision during the surgery (45.5%) and for silicone oil filling at the end of the operation (63.6%). After the surgery, NVG was found in 14 eyes, un-reattached retina in 5 eyes (before the surgery was VI stage of PDR), and optic nerve atrophy and retinal vessel atresia in 3 eyes, which significantly differed from which in the patients with light perception (Plt;0.001，P=0.004, (Plt;0.001). The differences of sex, diabetes type and PDR stage between the NLP group and non-NLP group were not significant (P=0.136, P=0.681, P=0.955). Conclusions The incidence of NLP after vitreoretinal surgery for proliferative diabetic retinopathy is low. The direct causes were NVG, optic nerve atrophy, retinal vessel atresia and retinal redetachment, while the sex, type of diabetes mellitus and stage of PDR show no statistical relation to the occurrence of NLP after surgery. (Chin J Ocul Fundus Dis,2007,23:244-247)
ObjectiveTo observe the clinical effect of intravitreal ranibizumab (IVR) combined with vitrectomy in treating proliferative diabetic retinopathy (PDR). MethodsThis is a prospective non-randomized controlled clinical study. A total of 62 patients (70 eyes) who underwent vitrectomy for PDR were enrolled and divided into IVR group (30 patients, 34 eyes) and control group (32 patients, 36 eyes).IVR group patients received an intravitreal injection of 0.05 ml ranibizumab solution (10 mg/ml) 3 or 5 days before surgery. The follow-up time was 3 to 18 months with an average of (4.5±1.8) months. The surgical time, intraoperative bleeding, iatrogenic retinal breaks, use of silicone oil, the best corrected visual acuity (BCVA) and the incidence of postoperative complications were comparatively analyzed. ResultsThe difference of mean surgical time (t=6.136) and the number of endodiathermy during vitrectomy (t=6.128) between IVR group and control group was statistically significant (P=0.000, 0.036). The number of iatrogenic retinal break in IVR group is 8.8% and control group is 27.8%, the difference was statistically significant (χ2=4.154, P=0.032). Use of silicone oil of IVR group is 14.7% and control group is 38.9%, the difference was statistically significant (χ2=5.171, P=0.023). The incidence of postoperative vitreous hemorrhage in 3 month after surgery was 11.8% and 30.6% respectively in IVR group and control group. The differences were statistically significant (χ2=3.932, P=0.047). The 6 month postoperative mean BCVA of IVR group and control group have all improved than their preoperative BCVA, the difference was statistically significant (t=4.414, 8.234; P=0.000).But there was no difference between the mean postoperative BCVA of two groups (t=0.111, P=0.190). There was no topical and systemic adverse reactions associated with the drug after injection in IVR group. ConclusionsMicroincision vitreoretinal surgery assisted by IVR for PDR shorten surgical time, reduces the intraoperative bleeding and iatrogenic retinal breaks, reduces the use of silicon oil and the postoperative recurrent vitreous hemorrhage. But there was no significant relationship between vision improvement and IVR.
Objective To observe the clinical features and outcomes of vitrectomy for diabetic retinopathy (DR) with central retinal vein occlusion (CRVO) in type 2 diabetes mellitus (T2DM). Methods A total of 192 patients (241 eyes) with proliferative DR (PDR) who underwent vitrectomy were enrolled in this study. All the patients were diagnosed as vitreous hemorrhage (VH) because of suddenly decreased vision. There were 93 eyes with tractional retinal detachment (TRD) and six eyes with neovascularization of iris (NVI). The patients were divided into PDR with CRVO group (group A, 41 eyes) and PDR group (group B, 200 eyes) according to the results of fundus examination. All patients received vitrectomy with silicone oil and C3F8 gas tamponade. There were 138 eyes with silicone oil tamponade which including 30 eyes in group A and 108 eyes in group B. The difference of number in silicone oil-filled eyes in two groups was statistically significant (chi;2=5.110,P＜0.05). There were 38 eyes with C3F8 gas tamponade which including six eyes in group A and 32 eyes in group B. There was no difference in C3F8 gas-filled eyes numbers in two groups (chi;2=0.048, P＞0.05). The follow-up ranged from one to 60 months, with the mean of (28.69plusmn;17.28) months. The corrected vision, retinal reattachment, persisting macular edema (ME), neovascular glaucoma (NVG) and repeated VH after surgery were comparatively analyzed. Results Of 241 eyes, there were 41 eyes (17.0%) with CRVO. Before surgery, the differences of corrected vision (Z=-0.138), intraocular pressure (t=0.966), whether there was TRD or not (chi;2=0.412), whether underwent panretinal photocoagulation or not (chi;2=1.416) were not statistically significant (P＞0.05), but the difference of whether NVI were present or not was statistically significant (chi;2=31.724,P＜0.05) between two groups. After surgery, the corrected vision improved in both two groups (Z=2.319, 4.589; P＜0.05). There was no difference of corrected vision after surgery between two groups (Z=0.782,P＞0.05). Postoperative complications occurred in 94 eyes, including 26 eyes in group A and 68 eyes in group B. The differences of incidence of reoperation (chi;2=0.498), retinal reattachment (chi;2=0.818), persisting ME (chi;2=2.722) between two groups after surgery were not statistically significant (P＞0.05). The incidence of repeated VH (chi;2=5.737) and NVG (chi;2=6.604) in group A were higher than those in group B (P＜0.05). Conclusions CRVO is commonly found to coexist with DR in T2DM patients with VH. Combined with CRVO patients are more likely to suffer NVI. Vitrectomy can improve the visual function in PDR with CRVO patients.
Objective To evaluate the long-term results of vitreoretinal surgery without use of intraocular silicone oil or gas in patients with diabetic tractional retinal detachment (DTRD). Methods The clinical interventional caseseries study included 104 patients (112 eyes) with DTRD, who were consecutively treated by pars plana vitrectomy without use of intraocular silicone oil or gas. Among the eyes, there were 6 eyes with iris neovascularization (INV), 1 eye with neovascular glaucoma (NVG) and 50 eyes with macular retinal detachment. There were no preexisting retinal holes or breaks prior to surgery nor any iatrogenic retinal breaks developed during vitrectomy. Cataract removal combined with intraocular lens implant surgeries were performed on 15 eyes. Followup duration varied from 12 to 65 months (mean: 29 months). Results Subretinal fluid was completely absorbed within 2 months after surgery. In 107 eyes (95.54%), the retina reattached after surgery and remained attached till the end of followup period. Best corrected visual acuity (BCVA) improved in 79 eyes (70.53%), remained unchanged in 14 eyes (12.50%) and got worse in 19 eyes (16.79%). The BCVA improving rate was lower in the macular detached group (33 eyes/50 eyes, 66.00% Vs 46 eyes/62 eyes, 74.19%,chi;2=0.89, P=0.344). No obviously aggravated opacity of lens was observed after vitreoretinal surgeries in the eyes without cataract surgeries. Seven (6.25%) eyes showed INV (5 new onset eyes), and none of them developed into NVG. In multivariate logistic regression, factors associated with postoperative rubeosis iridis were pre-existing rubeosis iridis ［adjusted odds ratio (OR)=10.2］, low preoperative BCVA (OR=11.1) and low postoperative BCVA (OR=16.7). Conclusions Vitreoretinal surgery for DTRD may not necessarily be combined with silicone oilor gas tamponade if there are no preoperative or intraoperative retinal breaks, and only using irrigation fluid could access a good longterm prognosis result.
ObjectiveTo compare the effects of intravitreal tamponade of C3F8 with silicon oil on postoperative vitreous hemorrhage and visual prognosis after vitrectomy for proliferative diabetic retinopathy (PDR). MethodsThe clinical data of 121 patients (127 eyes) who underwent primary vitrectomy due to PDR were analyzed retrospectively. All the patients were divided into two groups according to different intravitreal tamponade, including C3F8 tamponade group (53 patients with 56 eyes) and silicone oil tamponade group (68 patients with 71 eyes). There was no difference of gender (χ2=0.956), age (t=1.122), duratiion of diabetes (t=0.627), fasting blood glucose (t=1.049), systolic pressure (t=1.056), diastolic pressure (t=0.517), history of hypertension (χ2=0.356), nephropathy (χ2=1.242), preoperative laser photocoagulation (χ2=1.225) and All the patients underwent three port pars plana vitrectomy. The mean follow-up was 2 years ranging from 6 months to 4 years. And then the incidence and onset time of postoperative vitreous hemorrhage and postoperative BCVA of the two groups were compared. ResultsPostoperative vitreous hemorrhage occurred in 14 of 56 eyes (25.00%) in C3F8 tamponade group. The average onset time of postoperative vitreous hemorrhage were (64.64±59.09) days ranging from 7-225 days and mostly were within 30-60 days (35.71%, 5/14). Postoperative vitreous hemorrhage also occurred in 7 of 71 eyes (9.89%) of silicone oil tamponade group after silicone oil removal with an average onset time of (25.29±20.46) days ranging from 3-65 days and were mostly within 15-30 days (42.86%, 3/7). There was a significant difference in the incidence of postoperative vitreous hemorrhage between the two groups (χ2=5.200, P<0.05). BCVA of the two groups was improved significantly after operation (Z=2.472, 3.114; P<0.05). Postoperative BCVA of silicone oil tamponade group was poorer than C3F8 tamponade group (Z=1.968, P<0.05). ConclusionBoth C3F8 and silicone oil tamponade can improve the visual acuity after vitrectomy for PDR. Compared with C3F8, silicone oil tamponade had lower incidence and late onset of postoperative vitreous hemorrhage after vitrectomy for PDR.