Is it time to call time on the scleral buckle?
http://www.100md.com
《英国眼科学杂志》
One can’t help feeling that we are well past the beginning of the end
Keywords: scleral buckle; rhegmatogenous retinal detachment; pars plana vitrectomy; proliferative vitreoretinopathy
Overnight, an extensive pool of subretinal fluid (SRF) disappears following the judicious application of a small plomb over a peripheral U-tear, and one can only marvel at the underlying mechanisms at play. The scleral indentation has somehow overcome the dynamic tractional forces operating in the region of the break, allowing it to close; and the "pigment epithelial pump" has obligingly dried out the subretinal space despite our having decimated the outer retinal tissues around the break in order to induce break sealing. Once the chorioretinal adhesion is fully established, it usually ensures that the break remains closed and the retina reattached even if the indentation eventually recedes and vitreoretinal traction is reinstated.
The attractions of the "Custodis" procedure are plain to see, not least because the complications of SRF drainage (and other scleral transgressions) are potentially avoided.1,2 However, the majority of rhegmatogenous retinal detachments (RRDs) cannot be treated so simply and effectively using a segmental buckle, especially where the breaks are too many or too large to indent with any degree of subtlety, too small or too well hidden to be identified with certainty, too awkward in location to be easily reached, or subject to too much traction. Twenty years ago, these challenges could be met by one or more of a series of surgical adjuncts, including encirclement of the globe (to relieve anterior circumferential traction and to ensure the maintenance of the indentations created by segmental buckles)1–3; intravitreal injection of air or gases of low solubility (to temporarily sequester the breaks from fluid vitreous that might otherwise be recruited subretinally)4,5; or closed microsurgery combined with break tamponade (to eliminate tractional elements and opacities by vitrectomy while bringing air, gas, or silicone oil into direct relation with the breaks after hydraulic retinal reattachment).6–8 Which strategy a surgeon chose to pursue at that time depended more on the mesmeric influence of the grand masters of vitreoretinal surgery than on the outcomes of formal clinical trials.9 However, the benefits that vitrectomy brought to the management of particularly difficult detachments were so palpable that a randomised controlled trial could not have been ethically justified.10
Escoffery and his colleagues were among the first to have the temerity to undertake closed microsurgery without scleral buckling for RRDs in which the responsible breaks were regarded as being otherwise eminently "buckleable."11 Their thesis—that "vitrectomy and gas" might eventually supersede buckling for most types of primary RRD—acknowledged the fact that the more anteriorly a break is located, the more difficult it is for traction to be relieved internally and for SRF to be evacuated via the break. This is the converse of non-drainage buckling where anterior breaks settle most readily onto a scleral indentation.1,2 Nevertheless, the first hesitant steps had been taken towards internal surgery without external buckling becoming the procedure of choice for most primary RRDs.
Variations on the surgical theme at about the same time as Escoffery’s publication were another reason for the faltering start to this evolutionary progression. In the United States, "pneumatic retinopexy" was popularised by Hilton and Grizzard.12 This depended on the capability of a relatively small intravitreal gas bubble to close breaks in the absence of scleral indentation, and it sometimes succeeded in cooperative patients with superior breaks despite the several inherent threats in using expanding gases in non-vitrectomised eyes. These included the potential for residual gel to compromise internal break tamponade, for continuing traction to militate against effective break sealing, and especially for new breaks to form.9,13 In the United Kingdom meanwhile, the D-ACE surgical sequence (that is, Drainage of SRF followed by Air injection, Cryopexy around the reattached breaks and placement of a low profile Explant) was introduced for bullous RRDs. This order of surgical steps was designed to reduce the risk of haemorrhage during SRF drainage and to facilitate retinopexy and accurate break indentation. The results from two case series, one from St Thomas’s Hospital and the other from Moorfields Eye Hospital, were published in the same issue of the BJO in 1985,14,15 and the procedure enjoyed a significant following for some time on this side of the Atlantic.
Other developments then came along that were to reinvigorate the onward march of the vitrectomy and gas approach, not least the widespread adoption of phacoemulsification and wide angle fundus viewing. With the exception of some diabetic patients and patients under age 50,16 vitrectomy invariably leads to the development of nuclear lens sclerosis (to which type of cataract eyes with RRD harbour a distinct susceptibility anyhow because of pre-existing vitreous syneresis).17 This represented a significant disadvantage until small incision phacosurgical techniques reached full maturity,9 but intraoperative or post-vitrectomy lens removal can now be carried out routinely with minimal risk of compromising the posterior segment surgery. Similarly, although internal searching for retinal breaks had been shown to be very effective (especially when combined with deep kinetic scleral indentation),18 new systems for wide angle viewing not only improved fundus visualisation and identification of peripheral retinal breaks (even through a small pupil), but their optics also made internal drainage of SRF and fluid/gas exchange in phakic eyes that much more straightforward.19 The use of heavy liquids to reattach the retina intraoperatively by squeezing SRF back out through the breaks,20 and close "shaving" of vitreous from the retina,21 have added to the sense of control over intraoperative events for the surgeon and to a perceived reduction in his/her need to employ additional external means of combating traction.
Once segmental buckles to indent vitrectomised breaks in the superior fundus had been abandoned, routine encirclement of the globe quickly fell from favour as well. Traditionally, encirclement as an adjunct to vitrectomy had been credited with counteracting residual vitreoretinal traction and with closing unseen breaks in the vitreous base region. These presumed attributes weren’t subjected to any formal study, however, and in many centres encirclement was discarded without so much as a murmur. In others, several rows of 360 degree laser retinopexy replaced encirclement in fulfilling the surgeon’s desire to tackle problems suspected to be lurking peripherally as, for example, in pseudophakic eyes.22 The encircling band having been dispatched to oblivion, this trend has latterly extended "full circle" to the recommended withdrawal of external buckling even for detachments from inferior tears (that is, those located between 4 o’clock clockwise to 8 o’clock near the equator or more anteriorly).23 These RRDs were formerly thought to require scleral indentation in order to achieve secure closure and sealing of the tears in the context of disadvantageous posturing and the risk of future break reopening from proliferative vitreoretinopathy (PVR), a process that particularly afflicts the inferior retina.
There is already justification for recommending avoidance of a long, wide buckle during primary surgery, with its attendant risk of vortex vein compression
As with the D-ACE sequence, independent case series from St Thomas’s and Moorfields hospitals are published simultaneously in this issue of the BJO (pp 1372 and 1376) to provide mutual corroboration of the feasibility of this latest act of vitreoretinal surgical omission. Eighty one per cent and 89% respectively of previously unoperated detachments with inferior tears were reattached at the first attempt using vitrectomy and gas without buckling. These success rates are similar to those reported for mixed primary detachments, whether buckled or not, in a national audit.24 Apparently, PVR wasn’t a major problem in the Moorfields study, but the extent to which its malign influence caused or complicated the failure of break closure in one fifth of the St Thomas’s cases, and then compromised the outcome of secondary surgery, isn’t revealed in the bald statistics presented.
To further strengthen their argument, each group of authors wrapped up their case series in a retrospective study, the flaws in which are acknowledged. The St Thomas’s group compared the outcomes of vitrectomy and gas for RRDs with inferior breaks with those for RRDs without inferior breaks that had been matched for patient age and sex (Wickham et al, this issue), though neither of these are important confounding variables in determining the results of RRD management. The Moorfields study compared treatment outcomes for RRDs with inferior breaks that had either been buckled or were non-buckled (Sharma et al, this issue) though the decision to buckle would inevitably have been biased by (more or less) tangible manifestations of early PVR. However, both centres emphasised the need to eliminate tractional elements from the vicinity of the breaks and for them to have achieved this objective with so little instrumental lens trauma is remarkable. The implication to be drawn is that excessively demanding postoperative posturing regimes need no longer be regarded as crucial to the successful management of unbuckled inferior breaks provided traction is thus minimised. Another factor may be the "enforced" reattachment of the retina and the use of a large gas bubble, measures that are not so critical in other circumstances.18
Does this mean, therefore, that the scleral buckle is now effectively dead and buried so far as the treatment of fresh RRDs from tractional breaks is concerned? Possibly. Further guidance may be forthcoming as and when a randomised trial of "segmental buckling versus vitrectomy and gas" reports.25 However, encirclement is permitted in both arms of this particular study, and it remains to be seen whether and how associated quality of life and economic issues will be addressed. For example, is the burden of posturing comparable to the ciliary neuralgia from ocular indentation in the immediate postoperative period, and how will fiscal matters surrounding the relative costs of consumables and disposables, and the prohibition of air travel, be analysed?
While we await this report, however, the possibility exists that the next generation of vitreoretinal surgeons, weaned on the sutureless incisions and fluidics of phacoemulsification, will be slower to acquire the necessary skills in external buckling anyway. Therefore, should the anatomical outcomes prove to be similar (as is likely), the balance of judgment will increasingly favour a procedure that expressly addresses vitreoretinal traction and vitreous opacities, and whose principal complication (namely, cataract) is eminently treatable, against one that doesn’t and that has a galaxy of potential drawbacks whether refractive, oculomotor, circulatory, or directly explant related.9 The possibility that vital staining of the retina might improve intraoperative break detection in future,26 and emerging evidence that reattachment of the foveal retina after vitrectomy is more definitive than that following scleral buckling,27 can only add to the appeal of the internal approach. Indeed, given these latest reports from Moorfields and St Thomas’s, there is already justification for recommending avoidance of a long, wide buckle during primary surgery for, say, an RRD from a tractional tear posterior to an inferior equatorial lattice patch, with its attendant risk of vortex vein compression.
What about the management of detachments resulting from multiple, inferiorly located, atrophic retinal holes and those caused by large oral dialyses? For the present, vitrectomy remains contraindicated for RRDs with no posterior vitreous detachment, but it shouldn’t be too long before the vitreous gel can be safely separated intraoperatively from detached retina, using a biochemically targeted intervention, as a prelude to complete vitrectomy and gas tamponade.28 Notwithstanding the continuing allure of the non-drainage approach for single, small, anteriorly located breaks, therefore, one can’t help feeling that we are well past the beginning of the end for the scleral buckle, particularly in primary RRDs uncomplicated by PVR.
REFERENCES
Lincoff H , Baras I, McLean J. Modifications to the Custodis procedure for retinal detachment. Arch Ophthalmol 1965;73:160–3.
Michels RG. Scleral buckling methods for rhegmatogenous retinal detachment. Retina 1986;6:1–49.
Schepens CL, Okamura ID, Brockhurst RJ, et al. Scleral buckling procedures. V. Synthetic sutures and silicone implants. Arch Ophthalmol 1960;64:868–81.
Rosengren B . Results of treatment of detachments of the retina with diathermy and injection of air into the vitreous. Acta Ophthalmol 1938;16:573–9.
Norton EWD. Intraocular gas in the management of selected retinal detachments. Trans Am Acad Ophthalmol Otolaryngol 1973;77:OP85–98.
Machemer R , Buettner H, Norton EWD, et al. Vitrectomy. A pars plana approach. Trans Am Acad Ophthalmol Otolaryngol 1971;75:813–20.
Charles S . Vitreous microsurgery. Baltimore: Williams and Wilkins 1981.
Leaver PK, Cooling RJ, Feretis EB, et al. Vitrectomy and fluid/silicone-oil exchange for giant retinal tears: results at six months. Br J Ophthalmol 1984;68:432–8.
Wilkinson CP. What is the ‘best’ way to fix a routine retinal detachment? Chapter 9,:85–102. In: Lewis H, Ryan SJ, eds. Medical and surgical retina. St Louis: Mosby, 1994.
Freedman B . Equipoise and the ethics of clinical research. N Engl J Med 1987;317:141–5.
Escoffery RF, Olk RJ, Grand MG, et al. Vitrectomy without scleral buckling for primary rhegmatogenous retinal detachment. Am J Ophthalmol 1985;99:275–81.
Hilton GF, Grizzard WS. Pneumatic retinopexy. A two-step outpatient operation without conjunctival incision. Ophthalmology 1986;93:626–41.
Grizzard WS, Hilton GF, Hammer ME, et al. Pneumatic retinopexy failures. Cause, prevention, timing, and management. Ophthalmology 1995;102:929–36.
Stanford MR, Chignell AH. Surgical treatment of superior bullous rhegmatogenous retinal detachments. Br J Ophthalmol 1985;69:729–32.
Gilbert C , McLeod D. D-ACE surgical sequence for selected bullous retinal detachments. Br J Ophthalmol 1985;69:733–6.
Thompson JT. The role of patient age and intraocular gas use in cataract progression after vitrectomy for macular holes and epiretinal membranes. Am J Ophthalmol 2004;137:250–7.
Harocopos GJ, Shui Y-B, McKinnon M, et al. Importance of vitreous liquefaction in age-related cataract. Invest Ophthalmol Vis Sci 2004;45:77–85.
Rosen PH, Wong HC, McLeod D. Indentation microsurgery: internal searching for retinal breaks. Eye 1989;3:277–81.
Spitznas M . A binocular indirect ophthalmomicroscope (BIOM) for non-contact wide-angle vitreous surgery. Graefes Arch Clin Exp Ophthalmol 1987;225:13–5.
Chang S . Low viscosity liquid fluorochemicals in vitreous surgery. Am J Ophthalmol 1987;103:38–43.
Oyagi T , Emi K. Vitrectomy without scleral buckling for proliferative vitreoretinopathy. Retina 2004;24:215–18.
Campo RV, Sipperley JO, Sneed SR, et al. Pars plana vitrectomy without scleral buckle for pseudophakic retinal detachments. Ophthalmology 1999;106:1811–16.
Tanner V , Minihan M, Williamson TH. Management of inferior breaks during pars plana vitrectomy for retinal detachment. Br J Ophthalmol 2001;85:480–2.
Thompson JA, Snead MP, Billington BM, et al. National audit of the outcome of primary surgery for rhegmatogenous retinal detachment. II. Clinical outcomes. Eye 2002;16:771–7.
Heimann H , Hellmich M, Bornfeld N, et al. Scleral buckling versus primary vitrectomy in rhegmatogenous retinal detachment (SPR Study): design issues and implications. SPR Study report no 1. Graefes Arch Clin Exp Ophthalmol 2001;239:567–74.
Jackson TL, Marshall J. Fluorophore-assisted retinal break detection using antibodies to glial fibrillary acidic protein. Invest Ophthalmol Vis Sci 2004;45:993–1001.
Wolfensberger TJ. Foveal reattachment after macula-off retinal detachment occurs faster after vitrectomy than after buckle surgery. Ophthalmology 2004;111:1340–3.
Ramesh S , Bonshek RE, Bishop PN. Immunolocalisation of opticin in the human eye. Br J Ophthalmol 2004;88:697–702.(D McLeod)
Keywords: scleral buckle; rhegmatogenous retinal detachment; pars plana vitrectomy; proliferative vitreoretinopathy
Overnight, an extensive pool of subretinal fluid (SRF) disappears following the judicious application of a small plomb over a peripheral U-tear, and one can only marvel at the underlying mechanisms at play. The scleral indentation has somehow overcome the dynamic tractional forces operating in the region of the break, allowing it to close; and the "pigment epithelial pump" has obligingly dried out the subretinal space despite our having decimated the outer retinal tissues around the break in order to induce break sealing. Once the chorioretinal adhesion is fully established, it usually ensures that the break remains closed and the retina reattached even if the indentation eventually recedes and vitreoretinal traction is reinstated.
The attractions of the "Custodis" procedure are plain to see, not least because the complications of SRF drainage (and other scleral transgressions) are potentially avoided.1,2 However, the majority of rhegmatogenous retinal detachments (RRDs) cannot be treated so simply and effectively using a segmental buckle, especially where the breaks are too many or too large to indent with any degree of subtlety, too small or too well hidden to be identified with certainty, too awkward in location to be easily reached, or subject to too much traction. Twenty years ago, these challenges could be met by one or more of a series of surgical adjuncts, including encirclement of the globe (to relieve anterior circumferential traction and to ensure the maintenance of the indentations created by segmental buckles)1–3; intravitreal injection of air or gases of low solubility (to temporarily sequester the breaks from fluid vitreous that might otherwise be recruited subretinally)4,5; or closed microsurgery combined with break tamponade (to eliminate tractional elements and opacities by vitrectomy while bringing air, gas, or silicone oil into direct relation with the breaks after hydraulic retinal reattachment).6–8 Which strategy a surgeon chose to pursue at that time depended more on the mesmeric influence of the grand masters of vitreoretinal surgery than on the outcomes of formal clinical trials.9 However, the benefits that vitrectomy brought to the management of particularly difficult detachments were so palpable that a randomised controlled trial could not have been ethically justified.10
Escoffery and his colleagues were among the first to have the temerity to undertake closed microsurgery without scleral buckling for RRDs in which the responsible breaks were regarded as being otherwise eminently "buckleable."11 Their thesis—that "vitrectomy and gas" might eventually supersede buckling for most types of primary RRD—acknowledged the fact that the more anteriorly a break is located, the more difficult it is for traction to be relieved internally and for SRF to be evacuated via the break. This is the converse of non-drainage buckling where anterior breaks settle most readily onto a scleral indentation.1,2 Nevertheless, the first hesitant steps had been taken towards internal surgery without external buckling becoming the procedure of choice for most primary RRDs.
Variations on the surgical theme at about the same time as Escoffery’s publication were another reason for the faltering start to this evolutionary progression. In the United States, "pneumatic retinopexy" was popularised by Hilton and Grizzard.12 This depended on the capability of a relatively small intravitreal gas bubble to close breaks in the absence of scleral indentation, and it sometimes succeeded in cooperative patients with superior breaks despite the several inherent threats in using expanding gases in non-vitrectomised eyes. These included the potential for residual gel to compromise internal break tamponade, for continuing traction to militate against effective break sealing, and especially for new breaks to form.9,13 In the United Kingdom meanwhile, the D-ACE surgical sequence (that is, Drainage of SRF followed by Air injection, Cryopexy around the reattached breaks and placement of a low profile Explant) was introduced for bullous RRDs. This order of surgical steps was designed to reduce the risk of haemorrhage during SRF drainage and to facilitate retinopexy and accurate break indentation. The results from two case series, one from St Thomas’s Hospital and the other from Moorfields Eye Hospital, were published in the same issue of the BJO in 1985,14,15 and the procedure enjoyed a significant following for some time on this side of the Atlantic.
Other developments then came along that were to reinvigorate the onward march of the vitrectomy and gas approach, not least the widespread adoption of phacoemulsification and wide angle fundus viewing. With the exception of some diabetic patients and patients under age 50,16 vitrectomy invariably leads to the development of nuclear lens sclerosis (to which type of cataract eyes with RRD harbour a distinct susceptibility anyhow because of pre-existing vitreous syneresis).17 This represented a significant disadvantage until small incision phacosurgical techniques reached full maturity,9 but intraoperative or post-vitrectomy lens removal can now be carried out routinely with minimal risk of compromising the posterior segment surgery. Similarly, although internal searching for retinal breaks had been shown to be very effective (especially when combined with deep kinetic scleral indentation),18 new systems for wide angle viewing not only improved fundus visualisation and identification of peripheral retinal breaks (even through a small pupil), but their optics also made internal drainage of SRF and fluid/gas exchange in phakic eyes that much more straightforward.19 The use of heavy liquids to reattach the retina intraoperatively by squeezing SRF back out through the breaks,20 and close "shaving" of vitreous from the retina,21 have added to the sense of control over intraoperative events for the surgeon and to a perceived reduction in his/her need to employ additional external means of combating traction.
Once segmental buckles to indent vitrectomised breaks in the superior fundus had been abandoned, routine encirclement of the globe quickly fell from favour as well. Traditionally, encirclement as an adjunct to vitrectomy had been credited with counteracting residual vitreoretinal traction and with closing unseen breaks in the vitreous base region. These presumed attributes weren’t subjected to any formal study, however, and in many centres encirclement was discarded without so much as a murmur. In others, several rows of 360 degree laser retinopexy replaced encirclement in fulfilling the surgeon’s desire to tackle problems suspected to be lurking peripherally as, for example, in pseudophakic eyes.22 The encircling band having been dispatched to oblivion, this trend has latterly extended "full circle" to the recommended withdrawal of external buckling even for detachments from inferior tears (that is, those located between 4 o’clock clockwise to 8 o’clock near the equator or more anteriorly).23 These RRDs were formerly thought to require scleral indentation in order to achieve secure closure and sealing of the tears in the context of disadvantageous posturing and the risk of future break reopening from proliferative vitreoretinopathy (PVR), a process that particularly afflicts the inferior retina.
There is already justification for recommending avoidance of a long, wide buckle during primary surgery, with its attendant risk of vortex vein compression
As with the D-ACE sequence, independent case series from St Thomas’s and Moorfields hospitals are published simultaneously in this issue of the BJO (pp 1372 and 1376) to provide mutual corroboration of the feasibility of this latest act of vitreoretinal surgical omission. Eighty one per cent and 89% respectively of previously unoperated detachments with inferior tears were reattached at the first attempt using vitrectomy and gas without buckling. These success rates are similar to those reported for mixed primary detachments, whether buckled or not, in a national audit.24 Apparently, PVR wasn’t a major problem in the Moorfields study, but the extent to which its malign influence caused or complicated the failure of break closure in one fifth of the St Thomas’s cases, and then compromised the outcome of secondary surgery, isn’t revealed in the bald statistics presented.
To further strengthen their argument, each group of authors wrapped up their case series in a retrospective study, the flaws in which are acknowledged. The St Thomas’s group compared the outcomes of vitrectomy and gas for RRDs with inferior breaks with those for RRDs without inferior breaks that had been matched for patient age and sex (Wickham et al, this issue), though neither of these are important confounding variables in determining the results of RRD management. The Moorfields study compared treatment outcomes for RRDs with inferior breaks that had either been buckled or were non-buckled (Sharma et al, this issue) though the decision to buckle would inevitably have been biased by (more or less) tangible manifestations of early PVR. However, both centres emphasised the need to eliminate tractional elements from the vicinity of the breaks and for them to have achieved this objective with so little instrumental lens trauma is remarkable. The implication to be drawn is that excessively demanding postoperative posturing regimes need no longer be regarded as crucial to the successful management of unbuckled inferior breaks provided traction is thus minimised. Another factor may be the "enforced" reattachment of the retina and the use of a large gas bubble, measures that are not so critical in other circumstances.18
Does this mean, therefore, that the scleral buckle is now effectively dead and buried so far as the treatment of fresh RRDs from tractional breaks is concerned? Possibly. Further guidance may be forthcoming as and when a randomised trial of "segmental buckling versus vitrectomy and gas" reports.25 However, encirclement is permitted in both arms of this particular study, and it remains to be seen whether and how associated quality of life and economic issues will be addressed. For example, is the burden of posturing comparable to the ciliary neuralgia from ocular indentation in the immediate postoperative period, and how will fiscal matters surrounding the relative costs of consumables and disposables, and the prohibition of air travel, be analysed?
While we await this report, however, the possibility exists that the next generation of vitreoretinal surgeons, weaned on the sutureless incisions and fluidics of phacoemulsification, will be slower to acquire the necessary skills in external buckling anyway. Therefore, should the anatomical outcomes prove to be similar (as is likely), the balance of judgment will increasingly favour a procedure that expressly addresses vitreoretinal traction and vitreous opacities, and whose principal complication (namely, cataract) is eminently treatable, against one that doesn’t and that has a galaxy of potential drawbacks whether refractive, oculomotor, circulatory, or directly explant related.9 The possibility that vital staining of the retina might improve intraoperative break detection in future,26 and emerging evidence that reattachment of the foveal retina after vitrectomy is more definitive than that following scleral buckling,27 can only add to the appeal of the internal approach. Indeed, given these latest reports from Moorfields and St Thomas’s, there is already justification for recommending avoidance of a long, wide buckle during primary surgery for, say, an RRD from a tractional tear posterior to an inferior equatorial lattice patch, with its attendant risk of vortex vein compression.
What about the management of detachments resulting from multiple, inferiorly located, atrophic retinal holes and those caused by large oral dialyses? For the present, vitrectomy remains contraindicated for RRDs with no posterior vitreous detachment, but it shouldn’t be too long before the vitreous gel can be safely separated intraoperatively from detached retina, using a biochemically targeted intervention, as a prelude to complete vitrectomy and gas tamponade.28 Notwithstanding the continuing allure of the non-drainage approach for single, small, anteriorly located breaks, therefore, one can’t help feeling that we are well past the beginning of the end for the scleral buckle, particularly in primary RRDs uncomplicated by PVR.
REFERENCES
Lincoff H , Baras I, McLean J. Modifications to the Custodis procedure for retinal detachment. Arch Ophthalmol 1965;73:160–3.
Michels RG. Scleral buckling methods for rhegmatogenous retinal detachment. Retina 1986;6:1–49.
Schepens CL, Okamura ID, Brockhurst RJ, et al. Scleral buckling procedures. V. Synthetic sutures and silicone implants. Arch Ophthalmol 1960;64:868–81.
Rosengren B . Results of treatment of detachments of the retina with diathermy and injection of air into the vitreous. Acta Ophthalmol 1938;16:573–9.
Norton EWD. Intraocular gas in the management of selected retinal detachments. Trans Am Acad Ophthalmol Otolaryngol 1973;77:OP85–98.
Machemer R , Buettner H, Norton EWD, et al. Vitrectomy. A pars plana approach. Trans Am Acad Ophthalmol Otolaryngol 1971;75:813–20.
Charles S . Vitreous microsurgery. Baltimore: Williams and Wilkins 1981.
Leaver PK, Cooling RJ, Feretis EB, et al. Vitrectomy and fluid/silicone-oil exchange for giant retinal tears: results at six months. Br J Ophthalmol 1984;68:432–8.
Wilkinson CP. What is the ‘best’ way to fix a routine retinal detachment? Chapter 9,:85–102. In: Lewis H, Ryan SJ, eds. Medical and surgical retina. St Louis: Mosby, 1994.
Freedman B . Equipoise and the ethics of clinical research. N Engl J Med 1987;317:141–5.
Escoffery RF, Olk RJ, Grand MG, et al. Vitrectomy without scleral buckling for primary rhegmatogenous retinal detachment. Am J Ophthalmol 1985;99:275–81.
Hilton GF, Grizzard WS. Pneumatic retinopexy. A two-step outpatient operation without conjunctival incision. Ophthalmology 1986;93:626–41.
Grizzard WS, Hilton GF, Hammer ME, et al. Pneumatic retinopexy failures. Cause, prevention, timing, and management. Ophthalmology 1995;102:929–36.
Stanford MR, Chignell AH. Surgical treatment of superior bullous rhegmatogenous retinal detachments. Br J Ophthalmol 1985;69:729–32.
Gilbert C , McLeod D. D-ACE surgical sequence for selected bullous retinal detachments. Br J Ophthalmol 1985;69:733–6.
Thompson JT. The role of patient age and intraocular gas use in cataract progression after vitrectomy for macular holes and epiretinal membranes. Am J Ophthalmol 2004;137:250–7.
Harocopos GJ, Shui Y-B, McKinnon M, et al. Importance of vitreous liquefaction in age-related cataract. Invest Ophthalmol Vis Sci 2004;45:77–85.
Rosen PH, Wong HC, McLeod D. Indentation microsurgery: internal searching for retinal breaks. Eye 1989;3:277–81.
Spitznas M . A binocular indirect ophthalmomicroscope (BIOM) for non-contact wide-angle vitreous surgery. Graefes Arch Clin Exp Ophthalmol 1987;225:13–5.
Chang S . Low viscosity liquid fluorochemicals in vitreous surgery. Am J Ophthalmol 1987;103:38–43.
Oyagi T , Emi K. Vitrectomy without scleral buckling for proliferative vitreoretinopathy. Retina 2004;24:215–18.
Campo RV, Sipperley JO, Sneed SR, et al. Pars plana vitrectomy without scleral buckle for pseudophakic retinal detachments. Ophthalmology 1999;106:1811–16.
Tanner V , Minihan M, Williamson TH. Management of inferior breaks during pars plana vitrectomy for retinal detachment. Br J Ophthalmol 2001;85:480–2.
Thompson JA, Snead MP, Billington BM, et al. National audit of the outcome of primary surgery for rhegmatogenous retinal detachment. II. Clinical outcomes. Eye 2002;16:771–7.
Heimann H , Hellmich M, Bornfeld N, et al. Scleral buckling versus primary vitrectomy in rhegmatogenous retinal detachment (SPR Study): design issues and implications. SPR Study report no 1. Graefes Arch Clin Exp Ophthalmol 2001;239:567–74.
Jackson TL, Marshall J. Fluorophore-assisted retinal break detection using antibodies to glial fibrillary acidic protein. Invest Ophthalmol Vis Sci 2004;45:993–1001.
Wolfensberger TJ. Foveal reattachment after macula-off retinal detachment occurs faster after vitrectomy than after buckle surgery. Ophthalmology 2004;111:1340–3.
Ramesh S , Bonshek RE, Bishop PN. Immunolocalisation of opticin in the human eye. Br J Ophthalmol 2004;88:697–702.(D McLeod)