Abstract
Objectives
To investigate the safety and effectiveness of pre-emptive vitrectomy in eyes with severe non-fibrotic proliferative diabetic retinopathy.
Methods
A multi-centre, retrospective, observational study. Pre-emptive vitrectomy was performed in non-fibrotic diabetic eyes with a visual acuity (VA) of 20/50 or better, where there was extensive persistent neovascularisation despite prior panretinal photocoagulation, and where the fellow eye had established sight loss despite vitrectomy for tractional complications. The primary outcome measure was the VA at last visit.
Results
Twenty patients were included. The mean age was 39 ± 14 years. Fifteen patients were Type 1 diabetic. The median baseline VA was 20/30 and remained stable at 20/28 at last visit (median follow-up period: 24 months). Eight eyes (40.0%) developed post-operative vitreous cavity haemorrhage; 4 of which required a vitreous cavity washout procedure. There were no post-operative retinal detachments. The index eye remained the significantly better eye at all time points bar one month post-surgery. Regression of retinopathy grading was observed in all eyes.
Conclusion
In this pilot study, we found no sight loss with pre-emptive diabetic vitrectomy. Better eye status was maintained in this high-risk group. Further study with larger number of patients and longer-term follow-up is indicated.
Subject terms: Retinal diseases, Outcomes research
Introduction
Proliferative diabetic retinopathy (PDR) remains a leading cause of blindness worldwide [1]. Standard management of PDR with panretinal photocoagulation (PRP) and more recently, intravitreal anti-vascular endothelial growth factor (anti-VEGF) agents reduce the risk of visual loss, but a third of these eyes will still suffer from PDR-related complications long term [2]. As illustrated in the DRCR.net protocol S study, only a third of eyes receiving either PRP or anti-VEGF showed regression of new vessels and improvement in diabetic retinopathy grading from PDR to non-PDR; half of these eyes also developed vitreous haemorrhage at some point during the study period; 6 to 15% developed some form of retinal detachment and 11 to 20% required vitrectomy over the 5-year follow-up period [3]. Standard treatment with PRP or anti-VEGF therefore does not seem adequate in preventing sight loss in certain high-risk eyes.
Parikh et al. found that on presentation with PDR and vitreous haemorrhage, the likelihood of requiring vitrectomy at 2 years was 34% [4]. The traditional indications for vitrectomy in eyes with PDR are ‘reactive’, usually in response to recurrent or non-clearing visually impairing vitreous haemorrhage and/ or fibrovascular traction affecting the macula. These indications have remained largely unchanged over the last 30 years since the Diabetic Retinopathy Vitrectomy Study (DRVS) showed up to a 40% risk of severe visual loss following such surgery [5]. In that period, the safety of the procedure has improved greatly with development in both equipment and techniques. Modern series using small gauge vitrectomy showed low rates of severe loss of vision and a high final rate of retinal re-attachment in these diabetic eyes [6, 7]. Whilst rates of anatomical success have improved, it remains the case that patients who require such surgery frequently have permanent visual loss. Jackson et al. found in their study that when delamination was required for a tractional retinal detachment, the eyes had often suffered irreversible visual loss. Despite an improvement in vision post-operatively, less than a third of the eyes in this registry study retained driving standard vision [8].
The natural history of PDR, and in particular the role of the angio-fibrotic switch from an early angio-proliferative phase, progressing to a late fibrotic phase is now better understood. Fundamental to this is the presence of a metaplastic fibrovascular posterior hyaloid complex in which unopposed connective tissue growth factor (CTGF) at the late stages of PDR can result in fibrous contraction and subsequent catastrophic visual loss in the form of tractional retinal detachment (TRD) and chronically distorted macular structure [9]. We also know that the need for fellow eye vitrectomy increases from 24% within 1 year, to 36% within 5 years after first eye vitrectomy [10] and that poor visual outcome following vitrectomy in first eye is predictive of a poorer outcome in the fellow eye due for vitrectomy, if the indication of vitrectomy remained a ‘reactive’ one [11].
A combination of the high risk of second eye vision loss in such patients and improved safety of modern vitrectomy has resulted in us hypothesising that pre-emptive vitrectomy in the fellow eye may result in an improved visual outcome compared to standard care.
The aim of this first in class pilot study was to investigate the safety and effectiveness of pre-emptive vitrectomy in patients with good vision and extensive non-fibrotic retinal neovascularisation despite prior PRP in the index eye, where the patient’s fellow eye had suffered loss of vision due to tractional complications of PDR.
Methods
This was a multi-centre, retrospective, observational study of consecutive patients managed at St Thomas’ Hospital (UK), The Maidstone Hospital (UK), Norfolk and Norwich Hospital (UK), Sunderland Eye Infirmary (UK), Eye Clinic Sulzbach, Knappschaft Hospital Saar (Germany), and Department of Ophthalmology, Aalborg University Hospital (Denmark) from 1st Jan 2014 to 30th June 2019. The study was defined as a service evaluation and accordingly, formal ethical approval was not required.
The inclusion criteria were: 1. Extensive, persistent neovascularisation elsewhere (NVEs) or of the disc (NVDs) in the index eyes despite prior PRP, defined subjectively as a persistent area or areas of non-fibrotic perfused neovascularisation totalling 3 or more disc areas; 2. No evidence of established pre-retinal fibrosis or traction in any area of the index eye; 3. Visual acuity (VA) equivalent to 20/50 or better in the index eye; and 4. Previous ‘reactive’ vitrectomy for tractional complications in the fellow eye, with or without non-clearing, dense vitreous haemorrhage.
Patient case notes, electronic patient records and imaging databases were reviewed to identify eyes that fulfilled the inclusion criteria within the study period. A pre-designed proforma was then used to collect data including baseline demographic, ocular co-morbidities, surgical procedure, intraoperative and post-operative complications, VA at 1, 3, 6, and 12 months, and the ETDRS diabetic retinopathy grading at final visit. Clinic acuities were converted to logarithm of the minimum angle of resolution (logMAR) for statistical analysis. Counting finger vision and hand movement vision were assigned an acuity score of 2.0 and 3.0 logMAR, respectively [12].
Intravitreal anti-VEGF (bevacizumab, 1.25 mg/0.05 ml) was administered 1 to 10 days prior to surgery in the absence of ischaemic events within 6 months and where the injection can be administered under local anaesthetic. A standard 3-port pars plana vitrectomy (PPV) was performed using either 23 g, 25 g or 27 g instrumentation by experienced vitreoretinal surgeons. Measures were taken to ensure the posterior hyaloid face was meticulously separated from the retina during surgery. Segmentation and delamination were performed as indicated. At the end of surgery, complete endo-PRP was applied up to the ora serrata. Intraocular tamponade was employed if required in cases with intraoperative iatrogenic breaks.
The primary outcome measure was the VA at last follow-up visit.
Statistical analysis was performed using the Stata software (StataCorp, Texas, USA) and a p < 0.05 was considered statistically significant. VA for index and fellow eyes were investigated for normality using a Shapiro-Wilk test, demonstrating deviation from a normal distribution (p < 0.01). A Wilcoxon signed-rank test was run to investigate the difference in VA in index eyes prior to intervention and at the final visit. A Mann-Whitney U test was used to investigate difference in VA between index and fellow eyes at each visit (baseline, month 1, month 3, month 6, month 12, and final visit).
Results
A total of 20 eyes of 20 patients were included in the study. All patients completed at least 12 months follow-up. Baseline characteristics are presented in Table 1. The mean age of patients was 39 years (range 26–68 years) and 15 (75.0%) patients were Type 1 insulin-dependent diabetic. All eyes had undergone PRP prior to surgery. Eighteen (90.0%) eyes received intravitreal anti-VEGF injection at a median of 2 days (range 1-8 days) before surgery.
Table 1.
Baseline characteristics.
| n = 20 n (%) | |
|---|---|
| Age (mean ± SD) | 39.2 ± 13.7 years |
| Female: Male | 10 : 10 |
| Laterality | |
| - Right | 13 (65) |
| - Left | 7 (35) |
| Diabetes | |
| - Type 1 | 15 (75) |
| - Type 2 | 5 (25) |
| Ocular co-morbidities | |
| - Diabetic macular oedema | 3 (15) |
| - Cataract | 2 (10) |
| - Pseudophakic | 1 (5) |
| Previous PRP | 20 (100) |
| Pre-op anti-VEGF | 18 (90) |
PRP Panretinal photocoagulation, VEGF vascular endothelial growth factor.
Intraoperative surgical details can be found in Table 2. The incidence of intraoperative iatrogenic breaks was 10% (2 out of 20 eyes) and these 2 eyes had sulphur hexafluoride gas tamponade (SF6) inserted at the end of surgery. No other intraoperative complications were recorded.
Table 2.
Intraoperative surgical details.
| n = 20 n (%) | |
|---|---|
| Gauge (n, %) | |
| 23 G | 8 (40) |
| 25 G | 11 (55) |
| 27 G | 1 (5) |
| Combined cataract surgery | 0 |
| Internal limiting membrane peel (n, %) | 3 (15) |
| Retinopexy (n, %) | |
| - Endolaser PRP | 20 (100) |
| - Cryopexy | 2 (10) |
| Intraoperative complications (n, %) | |
| - Iatrogenic breaks | 2 (10) |
| - Intraoperative retinal detachment | 0 |
| - Suprachoroidal haemorrhage | 0 |
| - Lens touch | 0 |
| Tamponade | |
| - BSS | 16 (80) |
| - Air | 2 (10) |
| - Gas | 2 (10) |
| - Oil | 0 |
PRP Panretinal photocoagulation, BSS balanced salt solution.
Visual and retinopathy grading outcome
The median baseline VA in the index eye was 20/30, which remained stable at 20/28 (p = 0.14) at last visit over a median follow-up period of 24 months (IQR = 19–36 months). No patients lost more than 2 logMAR lines at the last visit compared to baseline and 18 patients (90.0%) retained 20/40 or better in their operated eye. Figure 1 shows the mean changes of VA over the follow-up period in index and fellow eyes over the first 12 months of follow-up. The median VA was significantly better in the index eye than the fellow eye at all recorded time points (p-value < 0.05) throughout the study period bar month 1 (p = 0.31) after surgery. (Table 3).
Fig. 1.
Graph showing the mean best-corrected visual acuity in index and fellow eyes over 12 months follow-up period.
Table 3.
Comparison between the visual acuity for index and fellow eyes throughout different follow-up time points.
| Median VA (IQR) logMAR | |||
|---|---|---|---|
| Time point | Index eye | Fellow eye | p-value |
| Baseline | 0.20 (0.20–0.30) | 0.90 (0.23–2.00) | 0.001 |
| 1 month | 0.30 (0.20–0.70) | 0.65 (0.20–1.23) | 0.31 |
| 3 months | 0.20 (0.20–0.40) | 0.55 (0.28–0.93) | 0.04 |
| 6 months | 0.16 (0.02–0.40) | 0.50 (0.20–0.80) | 0.01 |
| 12 months | 0.16 (0–0.20) | 0.44 (0.30–0.70) | 0.003 |
| Final visit | 0.15 (0.05–0.23) | 0.50 (0.20–0.80) | 0.0003 |
VA visual acuity, IQR interquartile range.
Treated eyes were found to have achieved stable diabetic retinopathy grading at last visit and resolution of retinal haemorrhages was observed. No new areas of elevated neovascularisation or fibrosis were identified in these eyes at last visit. (Fig. 2a, b).
Fig. 2. Pre- and post- operative colour fundal photographs.
a Severe NVEs and NVDs despite PRP; vision was good at 20/20. b Stable retinopathy with resolution of retinal haemorrhages was observed 2 years after vitrectomy and vision was stable at 20/20. (NVD new vessels of the disc, NVE new vessels elsewhere, PRP panretinal photocoagulation).
Post-operative complications
Eight eyes (40.0%) developed post-operative vitreous cavity haemorrhage (PVCH), of which 6 had bled within the first 4 weeks after surgery. Four of these 8 eyes (20% overall) required 1 further operation in the form of PPV and vitreous cavity washout. All eyes had a clear vitreous cavity at the last visit. No eye underwent more than 2 vitrectomy procedures.
Cataract surgery was performed in 2 eyes (10%) during the follow-up period. Three eyes (15.0%) were treated for diabetic macular oedema in the follow-up period. No eyes developed retinal detachment post-operatively. Two eyes had pre-existing ocular hypertension controlled with topical anti-hypertensive drops and remained stable throughout the follow-up period.
Discussion
Patients who have irreversibly lost vision in one eye due to tractional complications of PDR are known to be at high risk of second eye visual loss from the same causes, with a 5-year cumulative incidence of up to 25% reported in a recent series [13]. Whilst it is possible to reactively improve the retinal structure and function with current surgical techniques, good vision is only rarely restored once traction has developed [8]. It, therefore, seems reasonable to hypothesise that pre-emptive vitrectomy in the phase prior to the onset of any cicatricial change may offer a better outcome. To the best of our knowledge, there are no studies in the current literature investigating this.
The included eyes in this study were defined as eyes that had persistent, multiple areas of actively perfused extensive NVDs/ NVEs despite PRP without evidence of fibrous proliferation, where the fellow eye had previously lost vision despite vitrectomy for tractional complications. These eyes shared some clinical features with what is termed florid diabetic retinopathy (FDR). FDR was originally defined by Kohner et al. [14]. and Beaumont et al [15]. It is characterised by bilateral, rapidly progressive PDR associated with severe ischaemia in patients with poor glycaemic control. FDR is extremely rare, representing less than 1% of diabetic patients [15]. The authors had refrained from using the term florid diabetic retinopathy in this report as rapid progression of the retinopathy and poor glycaemic control were difficult to ascertain due to the retrospective nature of the study. Instead, the definition used in this study included not only extensive new vessels in the index eye despite prior pan-retinal laser and also a history of loss of vision in fellow eye requiring vitrectomy. Diabetic retinopathy is usually a symmetric disease and studies have shown that poor visual outcome following vitrectomy in fellow eye was predictive of a poorer outcome in the eye due for vitrectomy [11]. Of note, the inclusion criteria used in our study were different in many ways compared to that used in the DRVS Report 3 study, where eyes included had evidence of either fibrous proliferation or red vitreous haemorrhage in addition to the modified Airlie House definition of severe or moderate neovascularization [5]. DRVS Report 3 also included patients with ‘useful vision’ which was defined as better than 10/200 Snellen acuity, rather than the 20/50 or better vision which we included in this series.
One of the concerns with diabetic vitrectomy is the high rate of complications reported particularly in early studies [5, 16]. Development of techniques, equipment and the use of surgical adjuncts have greatly reduced those risks. We found in our series that the overall intraoperative complications were lower than reported in the UK National Ophthalmological Database (UKNOD) Study for diabetic vitrectomy (10% vs. 23%) [8]. The main intraoperative complication recorded was iatrogenic breaks (10%) which had a comparable rate to that reported in the UKNOD series where delamination was not required, but lower than in the parallel series where delamination was performed (27.7%) [8]. The lower rate of complication might be explained by a less traumatic surgical separation of the vitreous from the retina due to a relatively normal vitreoretinal interface in our patients compared to eyes that had undergone an ‘angiofibrotic switch’ where fibrovascular complexes with retinal traction were evident. Only 10% of our patients required gas tamponade and no patients required silicone oil tamponade. On the other hand, 60-100% of eyes required a tamponade agent when delamination was performed in ‘reactive’ diabetic vitrectomy and silicone oil was used in up to 20% of these cases [8, 17, 18]. The reduced need for tamponade agents in these eyes may have also mitigated any tamponade-related complications such as raised intraocular pressure and silicone oil toxicity that could further compromise the visual outcome [19]. Similarly, post-operative complications were also lower in our study with no incidence of retinal detachment and only 10% of our eyes required cataract surgery compared to up to 50% in other studies [8, 20]. Within the limits of this study, pre-emptive vitrectomy as performed here did not appear to be high-risk surgery.
Although the overall rate of complications was low in our cohort, 40% of the patients developed post-operative vitreous cavity haemorrhage (PVCH). PVCH is a recognised complication following vitrectomy, with a reported rate between 20 to 75% [21–23]. The use of pre-operative anti-VEGF agents has been associated with a reduction in this risk to 10–20% [21, 24]. The incidence of PVCH in our study was 40%, in spite of the use of pre-operative anti-VEGF injections in 90% of the cases. We postulate that this high incidence was due to the presence multiple pegs linking the retinal and neovascular circulations, which in turn were at risk of post-operative bleeding following the induction of posterior vitreous detachment and cutting or avulsion of the neovascular origins. Seventy-five percent of the PVCH in our study happened in the first 4 weeks, accounting for the poorer vision in the first 3 post-operative months and 1 in 2 of these patients with PVCH required cavity washout. It is therefore important to warn patients of the risk of transient reduction in vision during the early post-operative phase and the possibility of a second operation before the vision stabilises in the long term. The low incidence of late PVCH also suggests that proliferation of new vessels in the anterior hyaloid was uncommon in our series [21].
In addition to a lower complication rate, our patients also retained a stable vision of 20/28 at last visit over a follow-up period of 24 months. No patient had an acuity recorded to be more than 2 logMAR lines worse than baseline at last visit. Eighteen of the 20 patients also retained 20/40 or better in their operated eye, an acuity which approximates to the driving standard in many countries. In contrast, when ‘reactive’ diabetic vitrectomy was performed following loss of vision, the visual improvement after surgery did not always translate into driving standard vision. The UKNOD Study for diabetic vitrectomy found that the mean final VA after diabetic vitrectomy was 0.50 logMAR.8 In a series reported from the US, eyes with established diabetic tractional retinal detachment had a final post-operative VA of 0.93 logMAR, one of the worse visual outcome in this series probably reflecting the baseline cohort characteristics [8, 17].
The limitations of this study included its retrospective nature and a small sample size. In addition, a number of surgeons were involved in this study and some of the observed outcomes might be surgeon related. Although the inclusion criteria were pre-defined in this study, the clinical features that confer the very ‘high-risk’ status for serious PDR-related complications remain contentious, possibly contributing to some degree of selection bias in the study. While our study showed a good visual outcome from ‘pre-emptive’ vitrectomy in the index eye compared to the poor outcome that had arisen with ‘reactive’ vitrectomy in the fellow eye, with the small numbers and no control group undergoing standard of care laser treatment, the authors could not demonstrate and do not claim proof of superiority of pre-emptive vitrectomy in the management of such eyes. Equally, we could not demonstrate superiority of pre-emptive vitrectomy over repeated anti-VEGF treatment or further laser due to the lack of head-to-head studies in this area. However, caution is required when using anti-VEGF in these high-risk eyes as the incidence of TRD has been reported to vary from 1.5% to 18.4% due to induction of the angiofibrotic switch. It is also worth noting that these studies were limited by small numbers [25, 26]. The risk of TRD has also been reported to be higher in eyes with PDR refractory to PRP such as those included in our study [27].
In conclusion, pre-emptive diabetic vitrectomy in eyes with severe, non-fibrotic retinal neovascularisation with good vision was found to be safe and effective within the limitations of this small pilot study. We found better visual outcomes and a lower rate of complications overall compared to conventional ‘reactive’ diabetic vitrectomy that had been performed in the fellow eye of all patients. The operated eye remained the better eye at all early post-operative time points bar month 1 due to PVCH. The results of this first-in-class small pilot study appear to justify further investigation of the relative merits of pre-emptive vitrectomy against the more conservative standard care in this clinical context.
Summary
What was known before
Standard management of proliferative diabetic retinopathy with panretinal photocoagulation may not be sufficient in preventing sight loss in high-risk eyes.
‘Reactive’ vitrectomy in diabetic eyes where patients had already lost vision result in overall poor visual outcome.
What this study adds
Pre-emptive vitrectomy in high risk proliferative diabetic retinopathy eyes where vision was still good was found to be safe and effective.
Author contributions
All authors meet the following criteria:1. Conceived and/or designed the work that led to the submission, acquired data, and/or played an important role in interpreting the results. 2. Drafted or revised the manuscript. 3. Approved the final version. 4. Agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Funding
The authors received no funding for the study.
Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Resnikoff S, Pascolini D, Etya’ale D, Kocur I, Pararajasegaram R, Pokharel GP, et al. Global data on visual impairment in the year 2002. Bull World Health Organ. 2004;82:844–51. [PMC free article] [PubMed] [Google Scholar]
- 2.Gross JG, Glassman AR, Jampol LM, Inusah S, Aiello LP, Antoszyk AN, et al. Panretinal photocoagulation vs intravitreous ranibizumab for proliferative diabetic retinopathy: a randomized clinical trial. JAMA. 2015;314:2137–46. doi: 10.1001/jama.2015.15217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gross JG, Glassman AR, Liu D, Sun JK, Antoszyk AN, Baker CW, et al. Five-year outcomes of panretinal photocoagulation vs intravitreous ranibizumab for proliferative diabetic retinopathy: a randomized clinical trial. JAMA Ophthalmol. 2018;136:1138–48. doi: 10.1001/jamaophthalmol.2018.3255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Parikh R, Shah RJ, VanHouten JP, Cherney EF. Ocular findings at initial pan-retinal photocoagulation for proliferative diabetic retinopathy predict the need for future pars plana vitrectomy. Retina. 2014;34:1997–2002. doi: 10.1097/IAE.0000000000000192. [DOI] [PubMed] [Google Scholar]
- 5.Early vitrectomy for severe proliferative diabetic retinopathy in eyes with useful vision. Results of a randomized trial–Diabetic Retinopathy Vitrectomy Study Report 3. The Diabetic Retinopathy Vitrectomy Study Research Group. Ophthalmology. 1988;95:1307–20. [DOI] [PubMed]
- 6.Khan MA, Samara WA, Hsu J, Garg S. Short-term outcomes of hybrid 23-, 25-, and 27-gauge vitrectomy for complex diabetic tractional retinal detachment repair. Retin Cases Brief Rep. 2019;13:244–7. doi: 10.1097/ICB.0000000000000571. [DOI] [PubMed] [Google Scholar]
- 7.Ozone D, Hirano Y, Ueda J, Yasukawa T, Yoshida M, Ogura Y. Outcomes and complications of 25-gauge transconjunctival sutureless vitrectomy for proliferative diabetic retinopathy. Ophthalmologica. 2011;226:76–80. doi: 10.1159/000328407. [DOI] [PubMed] [Google Scholar]
- 8.Jackson TL, Johnston RL, Donachie PH, Williamson TH, Sparrow JM, Steel DH. The Royal College of Ophthalmologists’ National Ophthalmology Database Study of Vitreoretinal Surgery: Report 6, Diabetic Vitrectomy. JAMA Ophthalmol. 2016;134:79–85. doi: 10.1001/jamaophthalmol.2015.4587. [DOI] [PubMed] [Google Scholar]
- 9.Shea M. Early vitrectomy in proliferative diabetic retinopathy. Arch Ophthalmol. 1983;101:1204–5. doi: 10.1001/archopht.1983.01040020206006. [DOI] [PubMed] [Google Scholar]
- 10.Hwang JC, Sharma AG, Eliott D. Fellow eye vitrectomy for proliferative diabetic retinopathy in an inner city population. Br J Ophthalmol. 2013;97:297–301. doi: 10.1136/bjophthalmol-2012-302233. [DOI] [PubMed] [Google Scholar]
- 11.Sulak M, Urbancic M, Petrovic MG. Predicting visual outcomes of second eye vitrectomy for proliferative diabetic retinopathy. Retina. 2018;38:698–707. doi: 10.1097/IAE.0000000000001589. [DOI] [PubMed] [Google Scholar]
- 12.JT H. Visual acuity measurements. J Cataract Refractive Surg. 2004;30:287–90. doi: 10.1016/j.jcrs.2004.01.014. [DOI] [PubMed] [Google Scholar]
- 13.Schreur V, Brouwers J, Van Huet RAC, Smeets S, Phan M, Hoyng CB, et al. Long-term outcomes of vitrectomy for proliferative diabetic retinopathy. Acta Ophthalmol. 2021;99:83–9. doi: 10.1111/aos.14482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kohner EM, Hamilton AM, Joplin GF, Fraser TR. Florid diabetic retinopathy and its response to treatment by photocoagulation or pituitary ablation. Diabetes. 1976;25:104–10. doi: 10.2337/diab.25.2.104. [DOI] [PubMed] [Google Scholar]
- 15.Beaumont P, Hollows FC. Classification of diabetic retinopathy, with therapeutic implications. Lancet. 1972;1:419–25. doi: 10.1016/S0140-6736(72)90868-9. [DOI] [PubMed] [Google Scholar]
- 16.Two-year course of visual acuity in severe proliferative diabetic retinopathy with conventional management. Diabetic Retinopathy Vitrectomy Study (DRVS) report #1. Ophthalmology. 1985;92:492–502. [DOI] [PubMed]
- 17.Sokol JT, Schechet SA, Rosen DT, Ferenchak K, Dawood S, Skondra D. Outcomes of vitrectomy for diabetic tractional retinal detachment in Chicago’s county health system. PLoS One. 2019;14:e0220726. doi: 10.1371/journal.pone.0220726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mikhail M, Ali-Ridha A, Chorfi S, Kapusta MA. Long-term outcomes of sutureless 25-G+ pars-plana vitrectomy for the management of diabetic tractional retinal detachment. Graefes Arch Clin Exp Ophthalmol. 2017;255:255–61. doi: 10.1007/s00417-016-3442-7. [DOI] [PubMed] [Google Scholar]
- 19.Pichi F, Hay S, Abboud EB. Inner retinal toxicity due to silicone oil: a case series and review of the literature. Int Ophthalmol. 2020;40:2413–22. doi: 10.1007/s10792-020-01418-0. [DOI] [PubMed] [Google Scholar]
- 20.Gupta B, Sivaprasad S, Wong R, Laidlaw A, Jackson TL, McHugh D, et al. Visual and anatomical outcomes following vitrectomy for complications of diabetic retinopathy: the DRIVE UK study. Eye. 2012;26:510–6. doi: 10.1038/eye.2011.321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Smith JM, Steel DH, Anti-vascular endothelial growth factor for prevention of postoperative vitreous cavity haemorrhage after vitrectomy for proliferative diabetic retinopathy. Cochrane Database Syst Rev. 2015:CD008214. [DOI] [PMC free article] [PubMed]
- 22.Yang CM. Surgical treatment for diabetic retinopathy: 5-year experience. J Formos Med Assoc. 1998;97:477–84. [PubMed] [Google Scholar]
- 23.Khuthaila MK, Hsu J, Chiang A, DeCroos FC, Milder EA, Setlur V, et al. Postoperative vitreous hemorrhage after diabetic 23-gauge pars plana vitrectomy. Am J Ophthalmol. 2013;155:757–63. doi: 10.1016/j.ajo.2012.11.004. [DOI] [PubMed] [Google Scholar]
- 24.Tan SZ, Dellʼ Aversana Orabona G, Robins JJ, Kumaran N, Wong R. “Delamination Plus”: A Technique to Reduce Immediate Postoperative Diabetic Cavity Hemorrhage. Retina. 2020. Online ahead of print. [DOI] [PubMed]
- 25.Torres-Soriano ME, Reyna-Castelán E, Hernández-Rojas M, García-Aguirre G, Kon-Jara V, Diaz-Rubio JL, et al. Tractional retinal detachment after intravitreal injection of bevacizumab in proliferative diabetic retinopathy. Retin Cases Brief Rep. 2009;3:70–3. doi: 10.1097/ICB.0b013e3181578dd8. [DOI] [PubMed] [Google Scholar]
- 26.Oshima Y, Shima C, Wakabayashi T, Kusaka S, Shiraga F, Ohji M, et al. Microincision vitrectomy surgery and intravitreal bevacizumab as a surgical adjunct to treat diabetic traction retinal detachment. Ophthalmology. 2009;116:927–38. doi: 10.1016/j.ophtha.2008.11.005. [DOI] [PubMed] [Google Scholar]
- 27.Arevalo JF, Sanchez JG, Lasave AF, Wu L, Maia M, Bonafonte S, et al. Intravitreal Bevacizumab (Avastin) for Diabetic Retinopathy: The 2010 GLADAOF Lecture. J Ophthalmol. 2011;2011:584238. doi: 10.1155/2011/584238. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.