Original Article

Scleral Buckling Versus Primary Vitrectomy in the Management of Retinal Detachment Associated with Mild Vitreous Hemorrhage

10.4274/tjo.18189

  • Tansu Erakgün
  • Serhad Nalçaci
  • Filiz Afrashi
  • Jale Mentes
  • Cezmi Akkin

Received Date: 13.08.2012 Accepted Date: 26.11.2013 Turk J Ophthalmol 2014;44(2):92-97

Objectives:

The aim of this study was to compare the surgical impact of scleral buckling (SB) and pars plana vitrectomy (PPV) on the anatomic results and visual recovery in cases of rhegmatogenous retinal detachment with mild vitreous hemorrhage.

Materials and Methods:

In this study, we included 78 eyes of 78 phakic patients who underwent primary surgery for rhegmatogenous retinal detachment complicated by mild vitreous hemorrhage (38 eyes by SB,40 eyes by PPV);all patients had been followed for longer than 6 months after surgery. The surgical outcome and the rate of complications were retrospectively compared. Mann-Whitney U-test and chi-square test were used for the statistical analysis.

Results:

The reattachment rate after the first surgery was 78.9% (30/38) in the SB group and 95% (38/40) in the PPV group. The difference between the groups was statistically significant (p=0.035). Unseen retinal breaks in 4 eyes, malpositioned buckle in 2 eyes and insufficient closure of the break in 2 eyes of SB cases, as well as reopening of the original break in 2 eyes of PPV cases were the causes of failure. Visual improvement in the PPV group was significantly better than in the SB group in the third and sixth postoperative months (p<0.001 and p=0.026, respectively). Peroperative and postoperative complications were as follows: in the SB group-subretinal hemorrhage (5.2%), PVR of grade B or worse (10.5%), macular pucker (7.8%), and postoperative cataract (13%); in the PPV group-iatrogenic breaks (10%), lens damage (5%), PVR of grade B or worse (5%),macular pucker (5%), and postoperative cataract (35%).

Conclusion:

Even though the high incidence of peroperative complications such as iatrogenic breaks or postoperative complications such as cataract formation was the major drawback, the results indicate that PPV performed to alleviate peripheral vitreoretinal tractions and clear vitreous hemorrhage offers better anatomic and functional results than SB in the management of rhegmatogenous retinal detachment associated with mild vitreous hemorrhage.

Keywords: Retinal detachment, scleral buckling, vitrectomy, vitreous hemorrhage

Introduction

Scleral buckling (SB) is currently regarded by most vitreoretinal surgeons as the preferred method for treating most primary rhegmatogenous retinal detachments (RRD)1 despite inadvertent complications such as subretinal hemorrhage, extraocular muscle imbalance,2 corneal contour changes,3 chorioretinal circulatory disturbances,4 and limitations in achieving early functional recovery.5 A single operation is often successful in more than 90% of uncomplicated cases.6,7

On the other hand, pars plana vitrectomy (PPV) offers certain advantages over SB in that it affords a direct approach to vitreous traction especially in cases with large or deeply located tears. Kloeti8 in 1983 and Escoffery et al.9 in 1985 reported an excellent outcome of PPV in eyes with RRD caused by deeply located tears without proliferative vitreoretinopathy (PVR).

Although preoperative vitreous hemorrhage is assumed a PVR-inducing factor in RRD10 and PPV today is certainly more widely accepted, some surgeons still remain conservative especially in RRD with vitreous hemorrhage that the retinal break (s) are visible. A number of reports have been published about surgical outcomes of primary PPV in RRD complicated with vitreous hemorrhage; however, to the best of our knowledge, there is no study comparing the surgical outcomes of the two approaches in the ophthalmic literature.

The goal of this study was to compare the surgical impact of SB and PPV on the anatomic results and visual recovery in cases of RRDs associated with mild vitreous hemorrhage.


Materials and Methods

Seventy-eight eyes of 78 patients underwent SB or PPV as the initial surgery for primary RRD associated with mild vitreous hemorrhage. All surgeries were performed between January 2006 and December 2009. All patients provided informed consent before surgery.

To evaluate the impact of these two surgical techniques on anatomic and visual recovery, only patients who met the following criteria were included in the analysis: the presence of RRD with retinal breaks associated with vitreous hemorrhage resulting from rupture of a peripheral retinal blood vessel and not obscuring the view of the fundus periphery (especially the retinal break region), optic disc and vascular structures (grade1- mild vitreous hemorrhage with visible fundus details).

The degree of intravitreal hemorrhage was scaled according to the diabetic retinopathy vitrectomy study grading system (Table 1).11 Excluding criteria were presence of vitreous hemorrhage obscuring the evaluation of the preoperative fundus (grade 2 or 3), retinal detachment due to giant retinal tears or trauma, aphakia, and PVR of grade C or greater. Finally, a total of 78 cases (38 SB and 40 PPV) met the criteria for data analysis.

Surgical Technique

Informed consent was obtained from all patients before surgery. The choice of one of the two procedures was based on the surgeon’s preference. All of the patients were operated under peribulbar anesthesia. When SB was performed, preoperative bed rest and elevation of the patient’s head was applied for a couple of days. During the surgery, chorioretinal adhesions were created with cryopexy around the retinal breaks. Segmental silicone sponge in combination with an encircling silicone band was used to support the peripheral retinal breaks in all cases. External drainage of subretinal fluid was performed in most eyes.

In PPV cases, a standard three-port PPV under binocular indirect ophthalmomicroscope (BIOM- Oculus Germany) was performed. Vitrectomy involved releasing vitreous traction at the region of flap tears following core vitrectomy, and an attempt was made to remove as much of the peripheral vitreous as possible in combination with scleral indentation. Perfluorocarbon liquid injection and fluid-air exchange were performed, and subretinal fluid was drained from the original tears. All retinal tears were treated by either endophotocoagulation or transscleral cryotherapy. The crystalline lens was removed only when it was cataracteous. Long-acting gas (C3F8) of 16% concentration was used for endotamponade in all cases.

Statistical analysis of visual functional outcome and retinal reattachment status was performed with data obtained only from patients with third- and sixth-month follow-up visits.

Visual acuity was measured with decimal visual acuity system. For statistical comparison, visual acuity was expressed as a logarithm of the minimum angle of resolution (logMAR) equivalents. The data were analyzed with Mann-Whitney U-test and chi-square test. A p-value of <0.05 was considered to be statistically significant.


Results

SB or PPV was performed in 78 eyes of 78 patients; 38 patients (18 right, 20 left eyes) underwent SB and 40 patients (18 right, 22 left eyes) - primary PPV. Of the 78 patients, 59% (46 of 78) were men and 41% (32 of 78) were women. The ages of the patients ranged from 17 to 71 years (mean±SD, 50.5±12.9 years). A retinal detachment that involved the macula was found in 27 eyes (71%) of the SB group and in 27 eyes (68%) of the PPV group. Multiple tears were noted in 6 eyes (15.7%) in the SB group and in 10 eyes (25%) in the PPV group. The tears were located anteriorly in all cases. Superior location of the breaks was found in 87% (33 of 38) of the SB group and in 85% (34 of 40) in the PPV group. The mean interval from the onset of symptoms to the initial surgery was 9.1±7.5 days in the SB group and 10.3±6.2 days in the PPV group.

The postoperative follow-up period ranged from 8 to 24 months (mean±SD, 15.5±5.2 months) in the SB group and from 6 to 15 months (mean±SD, 10.9±3.5 months) in the PPV group. There were no statistically significant differences in the baseline characteristics between the two groups except for the duration of the follow-up. The baseline characteristics of the 78 patients are summarized in Table 2.

In the PPV group, vitrectomy combined with phacoemulsification followed by intraocular lens implantation was performed in 8 eyes (20%). Long-acting gas (16% C3F8) was used in all patients.

The retina was reattached after the initial surgery in 78.9% of the SB group (30 of 38 cases) and in 95% of the PPV group (38 of 40 cases). The difference was statistically significant (p=0.035). In 8 eyes of the SB group in which the first operation failed in reattachment, the cause of the initial failure was unseen retinal breaks in 4 eyes, malpositioned buckle in 2 eyes, and insufficient closure of the break in 2 eyes. In 2 eyes of the PPV group in which the first operation failed in reattachment, the cause of the initial failure was reopening of the original break. At the final follow-up examination, all eyes (78 eyes) in both groups achieved retinal reattachment after one or more operations.

In the SB group, the incidence of intra- and postoperative complications was as follows: In 2 eyes (5.2%), subretinal hemorrhage occurred at the time of subretinal fluid drainage; postoperative visual acuity was not compromised. PVR of grade B or worse developed in 4 eyes (10.5%) within 2 weeks postoperatively. Macular pucker was noted in 3 eyes (7.8%); no further surgery was applied because the visual acuity was not impaired. Postoperative cataract developed in 5 eyes (13%) during the study period; 2 of 5 eyes underwent subsequent cataract surgery to treat the secondary visual impairment.

In 4 PPV cases (10%), iatrogenic breaks occurred during peripheral manipulation and were treated with endolaser photocoagulation. No retinal detachment attributable to these retinal breaks occurred. PVR of grade B or worse developed in 2 eyes (5%) within 2 weeks postoperatively. Macular pucker was noted in 2 eyes (5%); no further surgery was applied because of the minimal disturbance on visual acuity. Ocular hypertension was noted in 1 eye, but the IOP was controlled by a topical antiglaucoma drug in the postoperative period. Postoperative cataract developed in 14 eyes (35%) during the study period; 8 of 14 eyes required subsequent cataract surgery to treat the secondary visual impairment. Peroperative and postoperative complications are shown in Table 3.

In the SB group, at 3 months postoperatively, best-corrected visual acuity (BCVA) improved in 32 eyes (84%) and did not change in 6 eyes (16%). Visual acuity of 0.1 or better was achieved in 20 eyes (52%). Visual acuity of 0.4 or better was achieved in 4 eyes (10.5%).

In the PPV group, at 3 months postoperatively, BCVA improved in 38 eyes (95%) and did not changed in 2 eyes (5%). Visual acuity of 0.1 or better was achieved in 32 eyes (80%). Visual acuity of 0.4 or better was achieved in 14 eyes (35%).

Mean visual acuity at 3 months postoperatively was 1.09±0.47 logMAR (range, 0.2 and 2.1) in the SB group and 0.83±0.54 logMAR (range, 0.2 and 2.1) in the PPV group. The difference was statistically significant (p=0.01).

Mean visual acuity at 6 months postoperatively was 0.74±0.53 logMAR (range, 0.2 and 1.8) in the SB group and 0.71±0.44 logMAR (range, 0.2 and 2.1) in the PPV group. The difference was not statistically significant (p=0.93). Postoperative visual acuity of the study eyes is summarized in Table 4.

Mean visual acuity improvement at 3 months postoperatively was -0.29±0.53 logMAR (range, -1.10 and 0.90) in the SB group and -0.82±0.53 logMAR (range, -1.88 and 0.0) in the PPV group. The difference was statistically significant (p<0.001).

Mean visual acuity improvement at 6 months postoperatively was -0.64±0.67 logMAR (range, -1.80 and 1.40) in the SB group and -0.93±0.53 logMAR (range, -1.88 and 0.0) in the PPV group. The difference was statistically significant (p=0.026). Visual acuity improvement at the third and sixth months are summarized in Table 5.


Discussion

When vitreous hemorrhage is present in the setting of an acute PVD, the incidence of retinal tears is reported between 20 and 30%.12,13 Vitreous hemorrhage due to rupture of a blood vessel at the site of a retinal break is common in eyes with an acute RRD. Sarrafizadeh et al. reported that surgery (vitrectomy with/or without SB) was required for RRD in 78% of their patients presenting with acute, spontaneous, nontraumatic posterior vitreous separation with dense vitreous hemorrhage.14

Mild preoperative vitreous hemorrhage was not reported as an independent risk factor for postoperative PVR in primary RRDs by some authors,15 however, some others reported that the incidence of preoperative or postoperative PVR increases in the presence of preoperative vitreous hemorrhage in these eyes.10,16,17 Therefore, RRD with preoperative vitreous hemorrhage may be assumed one of the complicated types of RRD.

In some cases, the vitreous hemorrhage is severe and prohibits effective visualization of the fundus. Bed rest, elevation of the patient’s head, and bilateral patches often result in settling of the hemorrhage into the inferior vitreous cavity during a 1- to 5-day period.18 If a retinal break that accounts for the configuration of the detachment can be seen, conventional retinal reattachment surgery is performed. If visualization is not adequate despite a trial of bed rest and patching, a PPV is performed combined with retinal reattachment surgery.

A few number of studies that report only the surgical outcomes of primary PPV or SB to treat RRD with vitreous hemorrhage have been published in the ophthalmic literature. In the presence of vitreous hemorrhage, a final anatomic success rate of 86.5% for primary PPV in RRD has been reported by Hoerauf et al.19 Bonnet estimated a postoperative PVR rate of 30% after SB for RRD with vitreous hemorrhage.10 However, to the best of our knowledge, to date, there is no study comparing the surgical outcomes of the two approaches in the ophthalmic literature. In this retrospective study including cases of RRD with mild vitreous hemorrhage, we observed a single-procedure success rate of 78.9% in the SB group and 95% in the PPV group. The difference was statistically significant (p=0.035). Following a second operation, our final reattachment rate was 100% in both treatment groups. Regarding complicated types of RRDs, in eyes with multiple tears, the initial reattachment rate was reported better in the PPV group (96.9%) than in the SB Group (69.9%).20 The primary reattachment rate in eyes with complicated RRDs was found high and similar with both procedures.21 Oshima et al. have reported a single-procedure success rate of 91% in both treatment groups in non-complicated RRDs.22 And finally, in the SPR Study in which SB was compared with primary PPV in RRDs of medium complexity, the difference in primary reattachment rates was not statistically significant in the phakic subtrial.23

In our series, visual acuity improved in both treatment groups at the third- and sixth-month examinations. The visual improvement was better in the PPV group than in the SB group at three and six months (p<0.001 and p=0.026, respectively). There was a statistically significant difference in the BCVA at the third postoperative month (p=0.01). However, there was no statistically significant difference in the mean visual acuity at the sixth postoperative month (p=0.93). In the phakic subgroup of the SPR study, the mean BCVA change was reported significantly greater in the scleral buckling group than in the vitrectomy group. It is difficult to compare the visual results of primary vitrectomy with those of a conventional procedure because many preoperative factors including surgical indications, and postoperative factors such as corneal refractive changes,3,24 choroidal circulation disturbances,4 and extraocular muscle imbalance2 may be quite different between these two techniques. However, in the present series, we found that primary PPV was more effective than SB for achieving early visual recovery. In cases treated with SB surgery, although retinal reattachment can be funduscopically observed after external subretinal fluid drainage, some subretinal fluid may still remain in the submacular space because of its high viscosity. During the delay in fluid absorption, photoreceptor degeneration or apoptosis at the macula may result in poor visual recovery. In addition, in the presence of vitreous hemorrhage, cleaning of the vitreous gel from the hemorrhage may take time with a delayed visual recovery in SB procedure when comparing with primary PPV. This may be confirmed as there was no statistically difference between the BCVA of the two groups at the sixth month examination.

In our series, the encircling buckle was not used in the PPV group. The use of an additional buckling procedure is considered to relieve traction caused by residual vitreous, which can cause the original breaks to reopen or create new breaks. But, even when encircling buckle is used, inadequate removal of the vitreous may cause residual vitreous contraction and reopening of the retinal breaks leading to retinal detachment. Vitreous hemorrhage may play a role as an additive causative factor in the anatomic failure by reopening the break with persistent traction.

No iatrogenic break occurred during the subretinal fluid drainage in the SB group. During PPV, iatrogenic tears occurred in 4 eyes (10%). The incidence of this complication was higher than that in previous studies in which the incidence of iatrogenic breaks ranged from 8% to 15%.19-21,25 The effort to perform as much as possible vitrectomy for the vitreous hemorrhage close to the detached retina to prevent postoperative PVR may increase this complication. Endolaser photocoagulation was applied around the breaks to prevent postoperative retinal detachment and they did not affect the initial success rate.

Macular pucker was observed in 3 cases (7.8%) in the SB group and in 2 cases (5%) in the PPV group. Intraocular proliferative factors, such as retinal pigment epithelial cells and cytokines, can be directly eliminated during vitrectomy and additionally, laser photocoagulation is less invasive than cryoretinopexy to disperse viable retinal pigment epithelial cells. The difference however, was not statistically significant between the two groups (p=0.67). The incidence of this complication was comparable to that in previous studies.26-28

In our study, vitrectomy combined with phacoemulsification surgery was performed in 8 eyes (20%). Simultaneous hydrophobic acrylic IOL implantation was performed in all cases. In the follow-up period, postoperative cataract developed in 14 eyes (35%); 8 of 14 eyes required subsequent cataract surgery to treat the secondary visual impairment. When excluding 8 eyes in which cataract surgery was combined with vitrectomy, 44% of the eyes (14/32) developed postoperative cataract. Cataract progression is recognized as a major drawback of primary vitrectomy.29 The incidence of nuclear cataract following vitrectomy in patients older than 55 has been reported to be very high.23,30 For this reason, more surgeons today are performing simultaneous cataract surgery in patients over a certain age even though the patient may have a relatively transparent crystalline lens.

As one of the postoperative complications, PVR of grade B or worse developed in 4 eyes (10.5%) in the SB group and in 2 eyes (5%) in the PPV group. The difference was not statistically significant (p=0.425). The incidence of this complication in the SB group was comparable with previous reports including complicated RRDs10,31 and extremely low in the PPV group compared with previous reports.26,31 As the retinal detachment and proliferative changes occur together, PVR may be immediately induced especially in the presence of the vitreous hemorrhage. However, the development of the surgical instruments such as wide-field viewing systems and the increased use of cataract surgery combined with vitrectomy facilitate intraoperative viewing and safe manipulation at the vitreous base.

The limitation of the study is the small numbers in the study groups. A prospective trial with larger series is desirable to obtain an optimal information. Nevertheless, we think that the current study may give us some important information when comparing with the previous studies reporting only the surgical impact on the anatomic results and visual recovery of SB or vitrectomy procedures in RRDs complicated by mild vitreous hemorrhage.

In summary, the present series demonstrated that primary PPV offers better anatomic results and is more effective than SB in obtaining early visual recovery in RRD complicated by mild vitreous hemorrhage. Even though the high incidence of peroperative or postoperative complications such as iatrogenic breaks and postoperative cataract formation is the major drawback, PPV has some advantages over SB in eyes with RRD complicated by mild vitreous hemorrhage.


1. Kreissig I. Repair of primary retinal detachment: A review of its development and a comparison of the present. Turkiye Klinikleri J Med Sci. 2010;30:293-302.
2. Sewell JJ, Knobloch WH, Eifrig DE. Extraocular muscle imbalance after surgical treatment for retinal detachment. Am J Ophthalmol. 1974;78:321-3.
3. Smiddy WE, Loupe DN, Michels RG, Enger C, Glaser BM, deBustros S. Refractive changes after scleral buckling surgery. Arch Ophthalmol. 1989;107:1469-71.
4. Yoshida A, Feke GT, Green GJ, et al. Retinal circulatory changes after scleral buckling procedures. Am J Ophthalmol. 1983;95:182-8.
5. Bovey EH, Gonvers M. Retinal detachment and laser drainage. Klin Monbl Augenheilkd. 1995;206:332-5.
6. Wong D, Chignell AH, Inglesby DV, Little BC, Franks W. The treatment of bullous rhegmatogenous retinal detachment. Graefes Arch Clin Exp Ophthalmol. 1992;230:218-20.
7. The repair of rhegmatogenous retinal detachments. American Academy of Ophthalmology. Ophthalmology 1990;97:1562-72.
8. Kloeti R. Amotio-Chirurgie ohne Skleraindellung. Priare Vitrektomie. Klin Monbl Augenheilkd. 1983;182:474-8.
9. Escoffery RF, Olk RJ, Grand MG, Boniuk I. Vitrectomy without scleral buckling for primary rhegmatogenous retinal detachment. Am J Ophthalmol. 1985;99:275-81.
10. Bonnet M. The development of severe proliferative vitreoretinopathy after retinal detachment surgery. Grade B: a determining risk factor. Graefes Arch Clin Exp Ophthalmol. 1988;226:201-5.
11. 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.
12. Novak MA, Welch RB. Complications of acute symptomatic posterior vitreous detachment. Am J Ophthalmol. 1984;97:308-14.
13. Sharma S, Walker R, Brown GC, Cruess AF. The importance of qualitative vitreous examination in patients with acute posterior vitreous detachment. Arch Ophthalmol. 1999;117:343-6.
14. Sarrafizadeh R, Hassan TS, Ruby AJ, et al. Incidence of retinal detachment and visual outcome in eyes presenting with posterior vitreous separation and dense fundus-obscuring vitreous hemorrhage. Ophthalmology. 2001;108:2273-8.
15. Duquesne N, Bonnet M, Adeleine P. Preoperative vitreous hemorrhage associated with rhegmatogenous retinal detachment: a risk factor for postoperative proliferative vitreoretinopathy? Graefes Arch Clin Exp Ophthalmol. 1996;234:677-82.
16. Fournier P, Aracil P, Bonnet M. Rhegmatogenous retinal detachment and retinal tear with intravitreous hemorrhage. J Fr Ophtalmol. 1988;11:7-14.
17. Grizzard WS, Hilton GF, Hammer ME, Taren D. A multivariate analysis of anatomic success of retinal detachments treated with scleral buckling. Graefes Arch Clin Exp Ophthalmol. 1994;232:1-7.
18. Seelenfreund MH, Sternberg I, Hirsch I, Silverstone BZ. Retinal tears with total vitreous hemorrhage. Am J Ophthalmol. 1983;95:659-62.
19. Hoerauf H, Roider J, Herboth T, Hager A, Laqua H. Outcome after vitrectomy in rhegmatogenous retinal detachment and dense vitreous opacities. Klin Monbl Augenheilkd. 1997;211:369-74.
20. Miki D, Hida T, Hotta K, Shinoda K, Hirakata A. Comparison of scleral buckling and vitrectomy for retinal detachment resulting from flap tears in superior quadrants. Jpn J Ophthalmol. 2001;45:187-91.
21. Roider J, Hoerauf H, Hager A, Herboth T, Laqua H. Conventional ablation surgery or primary vitrectomy in complicated retinal holes. Ophthalmologe. 2001;98:887-91.
22. Oshima Y, Yamanishi S, Sawa M, Motokura M, Harino S, Emi K. Two-year follow-up study comparing primary vitrectomy with scleral buckling for macula-off rhegmatogenous retinal detachment. Jpn J Ophthalmol. 2000;44:538-49
23. Heimann H1, Bartz-Schmidt KU, Bornfeld N, Weiss C, Hilgers RD, Foerster MH; Scleral Buckling versus Primary Vitrectomy in Rhegmatogenous Retinal Detachment Study Group. Scleral buckling versus primary vitrectomy in rhegmatogenous retinal detachment: a prospective randomized multicenter clinical study. Ophthalmology. 2007;114:2142-54.
24. Hayashi H, Hayashi K, Nakao F, Hayashi F. Corneal shape changes after scleral buckling surgery. Ophthalmology. 1997;104:831-7.
25. Oshima Y, Emi K, Motokura M, Yamanishi S. Survey of surgical indications and results of primary pars plana vitrectomy for rhegmatogenous retinal detachments. Jpn J Ophthalmol. 1999;43:120-6.
26. Kertes PJ, Wafapoor H, Peyman GA, Calixto N Jr, Thompson H. The management of giant retinal tears using perfluoroperhydrophenanthrene. A multicenter case series. Vitreon Collaborative Study Group. Ophthalmology. 1997;104:1159-65.
27. Heimann H, Bornfeld N, Friedrichs W, et al. Primary vitrectomy without scleral buckling for rhegmatogenous retinal detachment. Graefes Arch Clin Exp Ophthalmol. 1996;234:561-8.
28. Lobes LA, Burton TC. The incidence of macular pucker after retinal detachment surgery. Am J Ophthalmol. 1978;85:72-7.
29. Kılıç R, Öner A. Regmatojen retina dekolmanı tedavisinde skleral çökertme, pars plana vitrektomi ve kombine skleral çökertme ve pars plana vitrektomi sonuçlarının değerlendirilmesi. Retina-Vitreus. 2011;19:180-4.
30. deBustros S, Thompson JT, Michels RG, Enger C, Rice TA, Glaser BM. Nuclear sclerosis after vitrectomy for idiopathic epiretinal membranes. Am J Ophthalmol. 1988;105:160-4.
31. Hakin KN, Lavin MJ, Leaver PK. Primary vitrectomy for rhegmatogenous retinal detachment. Graefes Arch Clin Exp Ophthalmol. 1993;231:344-6.