A series of 64 eyes, in 64 patients with giant retinal tears, treated by vitrectomy and fluid/silicone-oil exchange, has been reviewed after 5 years. Anatomical success was achieved in 73% of cases and visual function in successful cases ranged from 6/6 to NPL; 66% achieved acuities of 6/60 or better and 32% had vision of 6/18 or better. These results compare favourably with those in a previous series reported after 6 months and 18 months. Epiretinal membrane proliferation and shortening accounted for the majority of retinal redetachments while macular abnormalities, especially pucker, were responsible for a poor visual outcome. Glaucoma is the most serious long-term complication of the surgical method and occurs most frequently in aphakic eyes. Retention of the crystalline lens, whenever possible, the creation of a 6 o'clock iridectomy in aphakic eyes, early removal of silicone oil and a conservative approach to the use of scleral buckles are recommended.

A study was made of a consecutive series of 47 cases of rhegmatogenous retinal detachment treated by pars plana vitrectomy in which no holes were identified preoperatively. The view of the fundus during preoperative examination varied from being totally clear to completely obscured by media opacities. The role of pars plana vitrectomy in finding retinal holes peroperatively is considered. The incidence of discovering holes and the locations of those found at the time of surgery are presented. The significance of these findings is discussed. Where the preoperative view was good and the extent of proliferative vitreoretinopathy (PVR) did not exceed grade C2, retinal reattachment was achieved in 75% of the cases. A review made of a similar group of patients treated with conventional retinal buckling before the introduction of pars plana vitrectomy revealed that successful retinal reattachment was achieved in 70% of cases. The study concludes that pars plana vitrectomy, while being necessary for cases of rhegmatogenous retinal detachment when the view of the retina is obscured, will not always reveal the presence of a retinal break. If the preoperative view of the retina was good and the extent of PVR did not exceed grade C2, pars plana vitrectomy did not seem to offer obvious advantages over conventional buckling procedures.

outcome of primary surgery for rhegmatogenous retinal detachment. II. Clinical outcomes AbstractPurpose This national study was designed to audit anatomical outcome and complications relating to primary surgery for rhegmatogenous retinal detachments. This paper presents success and complication rates, and examines variations in outcome. Methods Sampling and recruitment details of this nationwide cross-sectional survey of 768 patients of 167 consultant ophthalmologists having their first operation for rhegmatogenous retinal detachment have been described. The main clinical outcomes detailed here are anatomical reattachment at 3 months after surgery and complications related to surgery. Consultants with a declared special interest in retinal surgery and able to perform pars plana vitrectomy were designated specialists for the analyses. Results Overall reattachment rate with a single procedure was 77% (95% CI 73.9-80.2). There were significant differences in reattachment rates between specialists and non-specialists. Without allowing for casemix, specialists had a reattachment rate of 82% (95% CI 77.9-85.7) with a single procedure and non-specialists 71% (95% CI 65.9-76.0). Allowing for case-mix, there was a significant difference between specialists and non-specialists for grade 2 detachments of 87% and 70% respectively (P Ͻ 0.0001). Analysing detachments by break type, the largest difference between specialists and non-specialists was observed for retinal detachments secondary to horseshoe tears, 80% and 68% respectively (P Ͻ 0.003). Specialists met the standards set for primary reattachment rates, while non-specialists did not. Over a third of patients had at least one complication reported at some point during the audit period. Conclusions Significant differences were seen in reattachment rates between specialists and non-specialists, overall and for specific subgroups of patients. This study provides relevant, robust and valid standards to enable all surgeons to audit their own surgical outcomes for primary retinal detachment repair in rhegmatogenous retinal detachments, identify common categories of failure and aim to improve results.

outcome of primary surgery for rhegmatogenous retinal detachment. I. Sample and methods AbstractPurpose This national study was designed to audit anatomical outcome and complications relating to primary surgery for rhegmatogenous retinal detachments. This paper presents survey methods, characteristics of participating consultants and the demographic and clinical characteristics of the patient sample. Methods Two surveys were undertaken. The first identified consultants who at the time performed retinal detachment surgery in the National Health Service. These surgeons formed the sampling frame for a nationwide cross-sectional clinical study that audited the outcomes of primary surgery for rhegmatogenous retinal detachments. Consultants selected patients according to the study eligibility criteria and data were collected by self-administered postal questionnaires. A validation exercise was carried out to examine selection bias and reporting accuracy. Results Only 256/671 (38%) of UK consultants, who responded to the first survey, indicated that they performed retinal detachment surgery on NHS patients. Annual activity varied between 0 and 400 primary procedures for rhegmatogenous retinal detachments. Seven hundred and sixty-eight eligible patients from 167 consultants were recruited for the clinical study. Twenty per cent of patients had a single retinal break with less than one quadrant of associated detachment and 45% had single or multiple breaks within the same quadrant and/or less than two quadrants of associated retinal detachment. Over 50% patients had single or multiple horseshoe tears. Validation studies suggested that there was no significant bias from the selection of patients or inaccuracy in reporting outcomes. Conclusions This large unselected group of primary rhegmatogenous retinal detachments provides a representative sample for considering variations in re-attachment rates.

have proposed on cytological and biochemical evidence that subretinal fluid is derived from retinal elements, from vitreous humour, and from serum leaking from choroidal and retinal vessels.8 If vitreous humour and subretinal fluid were derived from multiple sources and no exchange occurred across the retinal hole, the two fluids might be expected to differ (Fig 1A). Although most evidence is compatible with a net movement of fluid through the retinal hole in a direction from the vitreous humour to the subretinal space (Fig IB), there remains doubt as to the degree of mixing of the two fluids by this route. In a study of monkey eyes with longstanding rhegmatogenous retinal detachment, fluorescein isothiocyanate-dextran (molecular weight 70000 daltons) was shown to leave the vitreous cavity by bulk flow into the subretinal space,8 though egress from the latter space was very slow, tracer molecules still being present after six weeks.If vitreous humour and subretinal fluid reach chemical equilibrium, either by diffusion of solutes (Fig IC) or by a bulk, two-directional exchange of fluid (albeit with unequal flux in either direction; Fig 1D), then the two fluids should display biochemical similarity.In order to assess the degree of identity or dissimilarity of the two fluids, we present the first reported comparison of paired specimens of vitreous and subretinal fluid obtained during pars plana vitrectomy for rhegmatogenous retinal detachment. (D) Bidirectional bulkflow and diffusion; twofluids are not significantly different.