PRIMARY RETINAL DETACHMENT REPAIR

Comparison of 1-Year Outcomes of Four Surgical Techniques

SHLOMIT SCHAAL, MD, PHD, MARK P. SHERMAN, MD, CHARLES C. BARR, MD, HENRY J. KAPLAN, MD
Purpose: To compare functional and anatomical outcomes of modern methods of repair of primary retinal detachment. Materials and Methods: Retrospective interventional comparative case series. A total of 1,226 patients with primary retinal detachment were included in the study. All patients completed 1-year follow-up and were divided into 4 groups: 322 patients underwent scleral buckling surgery, 442 patients underwent pars plana vitrectomy surgery, 316 patients underwent a combination of scleral buckling and vitrectomy surgery, and 56 patients underwent pneumatic retinopexy surgery for the primary repair of retinal detachment. Reattachment success rates, pre- and postoperative visual acuity, complications, and change in refractive error were reviewed. Results: Initial success rate for retinal reattachment was 86% for scleral buckling only, 90% for vitrectomy only, 94% for the combination of scleral buckling and vitrectomy, and 63% for pneumatic retinopexy surgery. Although patients undergoing pneumatic retinopexy had a lower initial success rate, there was no statistically signicant difference in initial reattachment rates between the other three groups. There was no statistically signicant difference in nal visual acuity between the four groups. Complication rates varied among the techniques used. Conclusion: Postoperative visual acuity at 1 year did not differ among the various techniques used to repair primary rhegmatogenous retinal detachments. However, scleral buckling, vitrectomy, or a combination of both resulted in an initially better anatomical success rate and fewer operative procedures than pneumatic retinopexy. RETINA 31:15001504, 2011

reatment of primary retinal detachment continues to be of interest to vitreoretinal surgeons, comprising about half of all surgical cases in vitreoretinal surgery departments and practices.1 Modern techniques of primary retinal detachment repair allow most detachments to be repaired successfully. Although most techniques show high
From the Department of Ophthalmology and Visual Sciences, University of Louisville, Louisville, Kentucky. Presented in part at the Retina Society Meeting, Scottsdale, AZ, 2008. Supported in part by an unrestricted grant from Research to Prevent Blindness, Inc, New York, NY. Reprint requests: Shlomit Schaal, MD, PhD, Department of Ophthalmology and Visual Sciences, University of Louisville, 301 East Muhammad Ali Boulevard, Louisville, KY 40202; e-mail: s.schaal@louisville.edu

reattachment rates, each procedure has its own drawbacks and complications. About 40% of patients may not achieve reading ability, 10% to 40% may need .1 surgical procedure, and approximately 5% of eyes will have permanent anatomical and functional failure.2 Few studies39 to date have compared modern techniques of primary retinal detachment repair in regard to their technical and visual success rates and incidence of complications at 1 year. Primary retinal detachment repair techniques include scleral buckling alone, pars plana vitrectomy alone, a combination of scleral buckling and pars plana vitrectomy, and pneumatic retinopexy. We conducted the current retrospective case series involving four different surgeons to compare the results using different modern primary
1500

PRIMARY RETINAL DETACHMENT REPAIR  SCHAAL ET AL

1501

retinal detachment repair techniques. While retinal fellows participated to some extent in each case, an attending physician was present during the entire procedure. Methods Inclusion Criteria The medical records of 1,226 patients with retinal detachment who underwent primary repair at the University of Louisville, Louisville, KY, between the years 1996 and 2009 were reviewed. Patients who had diabetic retinopathy, proliferative vitreoretinopathy, giant retinal tears, combined retinal and choroidal detachments, or previous ocular trauma were excluded from the present study. The remaining patients were divided into 4 groups: 356 patients underwent scleral buckling surgery, 479 patients underwent vitrectomy surgery, 331 patients underwent a combination of scleral buckling and 20-gauge vitrectomy surgery, and 60 patients underwent pneumatic retinopexy surgery for the primary repair of their retinal detachment. There were no differences in the baseline clinical and anatomical characteristics of the patients in the four groups, although those undergoing pneumatic retinopexy had superior retinal breaks. The average visual acuity at presentation and the percentage of maculaoff presentations, which ranged between 54% and 58% and the extent of detachment, did not differ among the 4 groups (no statistical signicance). The choice on which primary procedure to perform was dependent on the surgeons preference. This study was approved by the institutional review board of the University of Louisville. Interventions For scleral buckling, varying silicone sponges and/ or encircling bands were used according to the surgeons preferences. Retinopexy of breaks was performed either with cryopexy or with indirect laser retinopexy. Optional surgical steps included intraocular injection of balanced saline solution, air, or sulfur hexauoride, drainage of subretinal uid, and anterior chamber paracentesis. Pars plana vitrectomy using Alcon Accurus vitrectomy system was performed either as standard 3-port sclerotomies or as transconjunctival 25-gauge surgery involving the insertion of transscleral cannulas using a beveled trocar after displacement of the conjunctiva. All eyes underwent core vitrectomy followed by elevation and removal of the posterior hyaloid membrane, if necessary. Subretinal uid was drained

through existing retinal breaks or retinotomies. A 20% sulfur hexauorideair mixture was used as endotamponade, and endolaser was used for retinopexy. Use of heavy liquids was optional. A combination of scleral buckling and vitrectomy included the suturing of an encircling buckle of varying width followed by a 20-gauge 3-port pars plana vitrectomy, internal drainage of subretinal uid, endolaser, and a 20% sulfur hexauorideair mixture for endotamponade as described above. The use of a fourth port intraocular chandelier lighting was optional. Pneumatic retinopexy patients were chosen according to the presence of retinal breaks within 1 clock hour of retinal arc in superior quadrants without other retinal breaks. These patients were treated in an ofce setting as described by Hilton and Tornambe5 and others3,4 using a subconjunctival or retrobulbar anesthesia supplemented by topical anesthesia. An injection of 0.3 cc of 100% C3F8 gas was used as endotamponade. Either laser retinopexy or cryopexy was used to treat retinal breaks, and patients were instructed to assume the appropriate head position. Outcome Measures Of the initial cohort of 1,226 patients, 1,136 completed 1-year of follow-up and were the subjects of this analysis. Retinal reattachment and nal visual acuity were the primary outcome measures. Complications that were recorded included increased intraocular pressure .21 mmHg that required treatment after 2 weeks, change in manifest refraction, cystoid macular edema, or epiretinal membrane formation as determined by clinical examination, uorescein angiography, or optical coherence tomography, and development of cataract that was clinically signicant or required cataract extraction within the 1-year follow-up period. Statistics Statistical analysis was performed using SPSS software (Version 11.0) for Microsoft Windows. Differences between the groups were calculated using Friedman test. Statistical signicance was accepted if P , 0.05. Results The results of our primary outcome measures are shown in Table 1. A total of 1,136 patients completed at least 1 year of follow-up, while 90 patients (7%)

1502 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES

2011  VOLUME 31  NUMBER 8

Table 1. Characteristics of Patients Undergoing Different Surgical Procedures Scleral Buckle Number of patients who completed 1-year follow-up Number of patients lost for follow-up Number phakic Number pseudophakic Anatomical success rate Average initial visual acuity Average nal visual acuity Increased IOP Change in refraction CME Cataract ERM Surgery time 322 34 (9%) 175 (54%) 147 (46%) 86% 20/400 20/50 11% 33% 16% 8% 5% 64 minutes Vitrectomy 442 37 (7%) 204 (46%) 238 (54%) 90% 20/400 20/40 6% 2% 4% 21% 10% 68 minutes Scleral Buckle and Vitrectomy 316 15 (4%) 120 (38%) 196 (62%) 94% 20/400 20/50 5% 14% 29% 28% 18% 120 minutes Pneumatic Retinopexy 56 4 (5%) 41 (73%) 15 (27%) 63% 20/80 20/30 6% 0% 0% 0% 0% NA

IOP, intraocular pressure; CME, cystoid macular edema; ERM, epiretinal membrane; NA, not applicable.

were lost to follow-up. Of the 1,136 patients, 540 were phakic and 596 were pseudophakic. As shown in Table 1, success rates for scleral buckling, vitrectomy and combined vitrectomy and scleral buckling procedures ranged from 86% to 94%. However, there was no statistically signicant difference between the anatomical success rates of these three procedures. The anatomical success rate for pneumatic retinopexy was signicantly loweronly 63% were anatomically attached after this procedure (P , 0.05). There was no statistically signicant difference between the average initial visual acuity measured for the different groups, except for the pneumatic retinopexy group which had signicantly better starting vision. All groups had improved nal visual acuity after 1 year of follow-up, with no statistically signicant difference in nal average vision between the 4 groups. This study has not compared the visual acuity between the macula-on cases, and the macula-off cases. There was no statistically signicant difference in intraocular pressure measurements between the 4 groups throughout the 1 year of follow-up. There was no difference in results depending on the primary surgeon (data not shown). Scleral buckling with or without vitrectomy had a statistically signicant increase in postoperative refractive change and cystoid macular edema when compared with vitrectomy alone (P , 0.05). By contrast, 1 of 4 phakic patients (25%, 79 patients) developed cataract after 1 of the vitrectomy procedures, while only 8% (14 patients) had cataract progression after a scleral buckle (P , 0.05) Surgery

time was signicantly increased (43 minutes longer in average) for the combined vitrectomy and scleral buckling group (P , 0.05). Figure 1 summarizes the percentage of procedures performed during the study period. There was a continuous decline in the percentage of scleral buckling for the repair of primary retinal detachment from 52% of cases in 1996 to 11% in 2008, with a concomitant increase in the number of primary vitrectomies. The percentage of combined procedures or pneumatic retinopexies did not change. This study was not designed to determine what caused failure of initial retinal detachment repair. Not all patient records documented the reason for surgical failure, although new retinal tears and proliferative vitreoretinopathy accounted for most of the documented reasons for failure. Most initial failures were repaired after 1 additional surgery (85%), while 12% needed 2 additional procedures (12%), and 3% needed $3 procedures.

Discussion Randomized trials comparing scleral buckling, vitrectomy, and pneumatic retinopexy as treatment for primary retinal detachment are infrequent.6,9 In previous studies, one procedure could not be demonstrated as clearly superior to another. Therefore, the selection of scleral buckling, vitrectomy, or pneumatic retinopexy for primary retinal detachment remains a subjective decision made by the surgeon, weighing the variables of each case.

PRIMARY RETINAL DETACHMENT REPAIR  SCHAAL ET AL

1503

In this study, eyes with noncomplex rhegmatogenous retinal detachments underwent a primary repair by one of four modern surgical techniques. The success rate for retinal reattachment was not statistically different for vitrectomy (both 20 gauge and 25 gauge), scleral buckling or combined vitrectomy and scleral buckling (ranging from 86% to 94%). The success rate for pneumatic retinopexy was signicantly lower (63%, P , 0.05). In their summary of 1,274 eyes in 26 published reports on pneumatic retinopexy performed between 1986 and 1989, Hilton et al5 found single-operation success for pneumatic retinopexy to range from 53% to 100%. While our primary success rate with pneumatic retinopexy was inferior to those reported in the pneumatic retinopexy trial, they are in keeping with those reported in the Southeast Wisconsin study9 and the study of Ross and Lavina,7 who reported a 51% success rate with pneumatic retinopexy. Although our study shows that our pneumatic retinopexy patients having similar nal visual acuities as the other procedures, this is partly because of the required use of additional procedures to achieve a desirable outcome for many patients. Approximately 40 of our 1,136 patients underwent 25-gauge vitrectomy to repair retinal detachment. The technical and visual results in this group did not differ from the 20-gauge group, which is in keeping with recently published studies by Miller et al10 and by Lai et al.11 This report is also the rst to document the incidence of complications of retinal reattachment surgery in the modern era. Signicant differences were found in incidence of and type of postoperative complications (Table 1). We found that the scleral

Fig. 1. Percentage of different retinal reattachment procedures performed during the years under study. SB, scleral buckle; Vit, vitrectomy; Comb, combined vitrectomy and scleral buckle; PR, pneumatic retinopexy.

buckling procedure resulted in refractive change in one third of the cases. Vitrectomy, however, had a minimal impact on patients refraction. The scleral buckling procedure, whether alone or in combination with vitrectomy had signicantly higher incidence of cystoid macular edema. We can only speculate as to the cause of increased cystoid macular edema in patients with scleral buckle. Possible factors include an increased amount of inammatory mediators in the vitreous cavity and the increased tissue manipulation that is part of the scleral buckling procedure. The vitrectomy procedure, with or without scleral buckling had a higher incidence of cataract formation. Most patients (142 patients of 210) in our series who developed visually signicant cataract underwent cataract surgery, and nal visual results were good. Epiretinal membranes have been found histologically in 60% to 75.5% of successful retina reattachment eyes postmortem.12,13 We found clinically apparent epiretinal membranes in up to 18% of successfully reattached eyes and were surprised by the relatively high number of eyes developing this complication after otherwise successful vitrectomy. Final visual acuity was rarely affected; however; only 11 patients required vitrectomy and removal of these epiretinal membranes. Our report also shows a declining trend in the use of scleral buckle for primary retinal detachment repair with an increased use of vitrectomy (Figure 1). This nding can be attributed to the fact that scleral buckling requires more tissue manipulation and results in greater refractive changes than vitrectomy. Also, the instrumentation used in vitrectomy surgery has improved over time. The introduction of 25-gauge vitrectomy has decreased the necessity of suturing and may have increased surgeons preference of vitrectomy over scleral buckling. At our institution, all patients receive the same care in the same hospital regardless of ability to pay. Nonetheless, in our geographic area, vitrectomy and vitrectomy combined with scleral buckling are reimbursed at a higher rate than scleral buckling alone. That this and the patients ability to pay may have played an unconscious role in the surgeons choice of procedure cannot be discounted. As vitreoretinal specialists, we are aware that vitrectomy procedures are inherently more expensive than scleral buckling and pneumatic retinopexy because of equipment and necessary supplies. Although vitrectomy is becoming a more common procedure for uncomplicated primary retinal detachment repair, scleral buckle is still favored in certain cases, such as inferior retinal tears. When discussing these procedures one has to keep in mind that higher single-operation success rate may not indicate

1504 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES

2011  VOLUME 31  NUMBER 8

superiority of a specic procedure, especially if the total cost of the surgery and reoperations are included. In the present study, postoperative morbidities such as diplopia, micropsia, or anesthesia-related morbidity were not assessed. These may be important in selecting the specic procedure to repair a retinal detachment in the individual patient. Cautious case selection and wise clinical judgment remain invaluable in tailoring a specic surgical technique to our patients. Although there was no statistically signicant difference in visual acuity or technical success, combined vitrectomy and scleral buckling had the highest initial technical success rate of 94%. As seen in Table 1, however, this procedure took almost twice as long as scleral buckling or vitrectomy alone and had much higher rates of complications. This is in keeping with the work of Weichel et al,14 who also found a higher complication rate in patients undergoing scleral buckle combined with vitrectomy. Although pars plana vitrectomy is currently the preferred surgical technique for primary retinal detachment repair at our institution, we continue to select the procedure that we believe meets the needs of the individual patient. We look forward to future investigations that will enable us to reattach retinas with higher success rates, better nal vision, and fewer complications. Acknowledgment The authors thank Tongalp H. Tezel, MD, whose patients were included in this study. References
1. Ah-Fat FG, Sharma MC, Majid MA, et al. Trends in vitreoretinal surgery at a tertiary referral centre: 1987 to 1996. Br J Ophthalmol 1999;83:396398.

2. Barrie T. Debate overview. Repair of a primary rhegmatogenous retinal detachment. Br J Ophthalmol 2003;87:790. 3. Kulkarni KM, Roth DB, Prenner JL. Current visual and anatomic outcomes of pneumatic retinopexy. Retina 2007;27: 10651070. 4. Chan CK, Lin SG, Nuthi AS, Salib DM. Pneumatic retinopexy for the repair of retinal detachments: a comprehensive review (19862007). Surv Ophthalmol 2008;53:443478. 5. Hilton GF, Tornambe PE. Pneumatic retinopexy. An analysis of intraoperative and postoperative complications. The Retinal Detachment Study Group. Retina 1991;11:285294. 6. Heimann H, Bartz-Schmidt KU, Bornfeld N, et al. Scleral buckling versus primary vitrectomy in rhegmatogenous retinal detachment: a prospective randomized multicenter clinical study. Ophthalmology 2007;114:21422154. 7. Ross WH, Lavina A. Pneumatic retinopexy, scleral buckling, and vitrectomy surgery in the management of pseudophakic retinal detachments. Can J Ophthalmol 2008; 43:6572. 8. Hoerauf H, Heimann H, Hansen L, Laqua H. Scleral buckling surgery and pneumatic retinopexy. Techniques, indications and results. Ophthalmologe 2008;105:718. 9. Han DP, Mohsin NC, Guse CE, et al. Comparison of pneumatic retinopexy and scleral buckling in the management of primary rhegmatogenous retinal detachment. Southern Wisconsin Pneumatic Retinopexy Study Group. Am J Ophthalmol 1998;126:658668. 10. Miller DM, Riemann CD, Foster RE, Petersen MR. Primary repair of retinal detachment with 25-gauge pars plana vitrectomy. Retina 2008;28:931936. 11. Lai MM, Ruby AJ, Sarrazadeh R, et al. Repair of primary rhegmatogenous retinal detachment using 25-gauge transconjunctival sutureless vitrectomy. Retina 2008;28: 729734. 12. Barr CC. The histopathology of successful retinal reattachment. Retina 1990;10:189194. 13. Wilson DJ, Green WR. Histopathologic study of the effect of retinal detachment surgery on 49 eyes obtained post mortem. Am J Ophthalmol 1987;103:167179. 14. Weichel ED, Martidis A, Fineman MS, et al. Pars plana vitrectomy versus combined pars plana vitrectomy-scleral buckle for primary repair of pseudophakic retinal detachment. Ophthalmology 2006;113:20332040.