PARS PLANA VITRECTOMY COMPARED WITH PARS PLANA VITRECTOMY COMBINED WITH SCLERAL BUCKLE IN THE PRIMARY MANAGEMENT OF NONCOMPLEX RHEGMATOGENOUS RETINAL DETACHMENT ANTON ORLIN, MD,* NINA J. HEWING, MD,† MICHAEL NISSEN, MD,* SANGWOO LEE, MD,* SZILARD KISS, MD,* DONALD J. D’AMICO, MD,* R. V. PAUL CHAN, MD* Purpose: To compare pars plana vitrectomy (PPV) with PPV combined with scleral buckle (PPV/SB) in the treatment of primary, noncomplex rhegmatogenous retinal detachment in an academic setting. Methods: Retrospective review of 74 consecutive cases that underwent either PPV or PPV/SB for primary rhegmatogenous retinal detachment at New York Presbyterian Hospital, Weill Cornell Medical College. Fifty-two eyes underwent PPV alone while 22 eyes had PPV combined with SB. All eyes had a minimum of 2 months of follow-up. The main outcome measure was single surgery anatomical success. Results: Patients in the PPV/SB group were less likely to be phakic (P = 0.05) and more likely to have an inferior retinal break (P = 0.001) when compared with the PPV group. Between groups, there was no difference in eyes with peripheral retinal lattice degeneration (P = 0.929), multiple breaks (P = 0.801), breaks seen preoperatively (P = 0.095), or those presenting with the macula off retinal detachment (P = 0.548). The majority of patients in both groups underwent small-gauge surgery (23 G or 25 G) (P = 0.65). Attachment of the retina was obtained in 100% of the patients in both groups at most recent follow-up. Single surgery anatomical success was similar between groups (83% PPV vs. 86% PPV/SB; P = 0.695). Mean best-corrected Snellen visual acuity improved in both groups (P = 0.75), with a final best-corrected Snellen visual acuity of 0.418 logMAR in the PPV group and 0.479 logMAR in the PPV/SB group (P = 0.61). When comparing PPV with PPV/SB, no difference in single surgery anatomical success existed after evaluating eyes with inferior breaks (P = 0.68), pseudophakia (P = 0.75), or when small-gauge surgery was performed (P = 0.76). Conclusion: We did not find significant differences in single surgery anatomical success, final anatomical success, or change in visual acuity when comparing PPV with PPV/SB in the repair of primary noncomplex rhegmatogenous retinal detachment in an academic setting where vitreoretinal fellows participate in key aspects of all cases. RETINA 34:1069–1075, 2014
R
PPV remains controversial2 and offers potential advantages and disadvantages to PPV alone. Some surgeons propose that the addition of a SB results in improved success rate, particularly when the retinal detachment occurs in pseudophakic eyes,5 involves unseen breaks,6 or involves inferior pathology.7 It is also possible that an encircling band may improve success in phakic patients by relieving anterior traction, because of the difficulty of performing adequate
hegmatogenous retinal detachment (RRD) remains a significant cause of visual morbidity.1 The most common surgical techniques for the repair of retinal detachment include scleral buckling (SB) and pars plana vitrectomy (PPV). The recent advances in vitrectomy instrumentation and technique have influenced many retinal surgeons to prefer the use of PPV for the repair of RRD in select cases.2–4 The adjuvant use of an encircling scleral buckle (SB) to 1069
1070 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
peripheral vitrectomy in these eyes. Certain disadvantages include longer operation and anesthesia times, globe perforation, and the risk of developing postoperative problems, such as diplopia,8 buckle extrusion, or infection. The purpose of this study is to 1) determine whether there is a difference in success rates between PPV and PPV combined with SB (PPV/SB) in the treatment of primary noncomplex RRD in an academic setting where vitreoretinal fellows participate in key aspects of all cases; and 2) investigate whether certain baseline characteristics of the RRD were associated with different outcomes between the two surgical techniques. In addition to the setting in which this study takes place, we also evaluate the use of smaller-gauge instrumentation in this analysis. Methods This retrospective review was approved by the Institutional Review Board of Weill Cornell Medical College (WCMC). The research adhered to the tenets of the Declaration of Helsinki and was conducted in accordance with the Health Insurance Portability and Accountability Act regulations. Charts of patients who were operated on at New York Presbyterian Hospital, WCMC were reviewed, and patients diagnosed with primary RRD who underwent PPV or PPV combined with SB between 2006 and August 2011 were identified. Patients were excluded if their RRD was repaired with SB or pneumatic retinopexy alone. Additional exclusion criteria was a documented follow-up ,2 months, previous vitreoretinal surgery, giant tears, retinal detachment in patients younger than 18 years, the presence of any type of proliferative vitreoretinopathy, traction retinal detachment resulting from proliferative diabetic retinopathy, or RRD in the setting of infectious retinitis. After reviewing the data, we found that of the noncomplex detachments repaired by current faculty, additional eyes were excluded because of the following: Roughly 11 eyes had primary buckles, 58 patients were classified as noncomplicated detachments (no proliferFrom the *Department of Ophthalmology, Weill Cornell Medical College, New York, New York; and †Department of Ophthalmology, Campus Benjamin Franklin, Charité - Universitaetsmedizin Berlin, Berlin, Germany. Paper presented at the Association for Research in Vision and Ophthalmology Annual Meeting, Fort Lauderdale, FL, May 5, 2011. Supported by the St. Giles Foundation (R.V.P.C.) and by the Research to Prevent Blindness (R.V.P.C., A.O., S.K., and D.J.D.). None of the authors have any financial/conflicting interests to disclose. Reprint requests: R. V. Paul Chan, MD, Department of Ophthalmology, Weill Cornell Medical College, 1305 York Avenue, 11th Floor, Room Y 11.34, New York, NY 10021; e-mail: roc9013@med. cornell.edu
2014 VOLUME 34 NUMBER 6
ative vitreoretinopathy) but had received previous laser or cryo or PPV surgery (elsewhere or at Weill Cornell), 12 patients had inadequate follow-up, 10 eyes had giant retinal tear (GRT) repair or previous trauma. All surgery was performed at New York Presbyterian Hospital by an attending vitreoretinal surgeon who is currently on staff at WCMC and directly involved with vitreoretinal fellowship training. Any surgery performed by a vitreoretinal surgeon who no longer is affiliated with WCMC or not involved with vitreoretinal fellowship training was excluded. Vitreoretinal fellows participated in key aspects of all cases. All vitreoretinal attending surgeons involved with the training of fellows at WCMC, New York Presbyterian Hospital were included in this study. Seventy-four consecutive eyes from 74 patients were included in this study. The main outcome measure was single surgery anatomical success (SSAS). Bestcorrected Snellen visual acuity (BCVA) was recorded and converted to logarithm of the minimal angle of resolution (logMAR). Preoperative data was recorded, such as the lens status, location of retinal tear whether a tear was visualized and if multiple tears were present, macula status, presence of peripheral lattice retinal degeneration, and BCVA. Data from the most recent follow-up visit was recorded including the duration of follow-up, final BCVA, SSAS, final anatomical success, and whether cataract extraction was performed. Data collected from patients who redetached included the number of redetachments and subsequent procedures performed. Visual acuity corresponding to counting fingers, hand motion, light perception, and no light perception corresponded to 1.7, 2.0, 2.5, and 3.0 logMAR, respectively.2,9
Surgical Technique All patients underwent standard 3-port 20 G, 23 G, or 25 G PPV using a wide-angle contact viewing system. Vitrectomy was characterized as being either large-gauge (20 G) or small-gauge (23 G or 25 G) surgery. The 23 G and 25 G surgery was combined to characterize small-gauge surgery secondary to the few eyes undergoing each separately when compared with the traditionally larger 20 G surgery. Endolaser photocoagulation was applied to completely surround all retinal breaks. All 20 G sclerotomy sites were sutured with 7-0 vicryl suture. Any 23 G or 25 G sclerotomy sites that were found to be leaking at the end of the surgery were sutured with 7-0 vicryl suture. A 360° encircling band (Band 41, Watzke Sleeve 270) was used in the PPV/SB group. Intraocular tamponade was achieved with nonexpansile gas (16% C3F8 or 20% SF6) or with 1,000 centistokes of silicone oil.
1071
TREATMENT OF RETINAL DETACHMENT ORLIN ET AL Table 1. Baseline Characteristics of Patients Undergoing PPV Alone and PPV Combined With SB PPV Alone (n = 52) Mean Age, years Women, n (%) Phakic, n (%) Right eye, n (%) Macula off, n (%) Presence of lattice degeneration, n (%) Retinal break visualized, n (%) Multiple breaks present, n (%) Inferior break present, n (%) Preoperative mean visual acuity, logMAR (Snellen)
26 34 27 31 16 49 24 9 0.903
Statistical Analysis All statistical analyses were performed using SPSS (IBM Corp, New York, NY) and Excel (Microsoft Inc, Redmond, WA). Independent samples t-test and chi-square test were used as appropriate. P , 0.05 was considered statistically significant. Results Preoperative and Intraoperative Data Seventy-four consecutive eyes from 74 patients were included in this study. Fifty-two eyes underwent PPV alone while 22 eyes had PPV combined with SB to repair the primary RRD. Preoperative characteristics are described in Table 1. Patients in the PPV/SB group were less likely to be phakic (41 vs. 65%; P = 0.05) and more likely to have an inferior retinal break (54 vs. 17%; P = 0.001) when compared with the PPV alone group. The mean age at presentation was 58.48 years in the PPV group (range, 27–80 years) versus 55.68 years in the PPV/SB group (range, 19–79 years) (P = 0.447). Between the two groups, there was no difference in the percent of patients with peripheral retinal lattice degeneration (P = 0.929), multiple breaks (P = 0.801), or those presenting with the macula off retinal Table 2. Intraoperative Characteristics in Patients of Patients Undergoing PPV Alone and PPV Combined With SB
Gauge surgery Large gauge 20 G, n (%) Small gauge Total, n (%) 23 G, n (%) 25 G, n (%) Gas tamponade used, n (%)
PPV Alone
PPV/SB
P
6 (13)
7 (32)
0.054
42 32 10 48
(88) (67) (21) (92)
15 15 0 19
(68) (68) (0) (86)
0.054 0.900 0.021 0.43
54.49 (50) (65) (52) (61) (31) (94) (46) (17) (20/159.83)
PPV with SB (n = 22) 8 9 9 15 7 18 10 12 0.890
55.68 (36) (41) (41) (68) (32) (82) (45) (54) (20/155.29)
P 0.45 0.28 0.05 0.39 0.55 0.93 0.10 0.80 0.001 0.943
detachment (P = 0.548). Although not reaching statistical significance, there was a trend for examiners to visualize a retinal break preoperatively in the PPV alone group (94 vs. 82%; P = 0.095). The intraoperative data is summarized in Table 2. The majority of patients in both PPV alone and SB/PPV groups received gas tamponade (P = 0.43). Also, the majority of eyes underwent small-gauge surgery (23 G or 25 G) with no statistically significant difference between the groups (P = 0.065) regarding the selection of large-gauge versus small-gauge instrumentation. Long-term mean follow-up was obtained in both groups with a mean of 406.73 days in the PPV group (range, 57– 1,226 days) and a mean of 502.14 days in the PPV/SB group (range, 83–1874 days). Each patient was followed for a minimum of 2 months. Postoperative characteristics at most recent follow-up are summarized in Table 3. Visual Acuity There was no difference in baseline BCVA between the groups. The mean preoperative BCVA was 0.903 logMAR (20/159.83 Snellen BCVA) and 0.890 logMAR (20/155.29 Snellen BCVA) in the PPV and PPV/SB groups, respectively (P = 0.943). Mean BCVA improved in both groups (P = 0.75), with a final BCVA of 0.418 logMAR (20/52.39 Snellen BCVA) in the PPV group and 0.479 logMAR (20/60.37 Snellen BCVA) in the PPV/SB group (P = 0.61). The degree of visual acuity improvement was further characterized in Table 3 and there was no difference between the groups in the percent of patients improving vision (P = 0.23) or in patients improving vision by at least 3 Snellen chart lines (P = 0.58). Anatomical Success Full attachment of the retina was obtained in 100% of the patients in both groups at most recent follow-up. Single surgery anatomical success rates were similar
1072 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 6
Table 3. Characteristics at Most Recent Follow-up of Patients Undergoing PPV Alone and PPV Combined With SB
Mean follow-up, days BCVA, logMAR (Snellen) Change in VA, logMAR No. patients having improved or stable vision, n (%) No. patients improving by $3 lines Snellen VA, n (%) SSAS, n (%) Final anatomical success, n (%) No. retinal detachment reoperations Phakic patients undergoing cataract extraction, n (%)
PPV Alone
PPV/SB
P
406.73 0.418 (20/52.39) 0.467 45 (88)
502.14 0.479 (20/60.37) 0.411 17 (77)
0.31 0.61 0.75 0.23
29 (57) 43 (83) 52 (100) 13 for 9 eyes, 1 surgery for 5 eyes, 2 surgeries for 4 eyes 12 (35)
11 (50)
0.58
19 (86) 0.70 22 (100) 1.00 3 for 3 eyes, 1 surgery for 0.695 3 eyes 9 (100) ,0.001
VA, visual acuity.
between the groups (83% PPV vs. 86% PPV/SB; P = 0.695). Among the eyes that had a redetachment, the majority of patients required only one subsequent surgery to obtain anatomical success (five patients in the PPV group and all three patients in the PPV/SB group). Four eyes in the PPV group that had a redetachment required two subsequent surgeries to achieve retinal attachment. Single Surgery Anatomical Success Based on Tear Location Overall, patients with inferior retinal breaks underwent more PPV/SB surgery compared with PPV alone (P = 0.001). Surgeries involving inferior retinal breaks had a trend to have a lower SSAS than those without inferior breaks (71 vs. 89%, P = 0.07), although not statistically significant. Amongst surgeries involving inferior breaks, SSAS was achieved in 67% of PPV eyes compared with 75% of PPV/SB eyes (P = 0.68). Preoperatively, a retinal break was not visualized in 9% of all surgeries, with no difference between surgical groups (P = 0.10). Amongst these patients, a 100% SSAS was found in both groups (n = 3 in PPV group, n = 4 in PPV/SB group). Single Surgery Anatomical Success Based on Lens Status Pseudophakic patients were more likely to undergo PPV/SB surgery (P = 0.05), and SSAS was higher for pseudophakic patients when compared with phakic ones (94 vs. 78%, P = 0.05). However, there was no difference in SSAS amongst pseudophakic patients between the PPV and PPV/SB groups (89% PPV vs. 92% PPV/SB; P = 0.75). Similarly, no difference in SSAS was demonstrated amongst phakic patients (79% PPV vs. 78% PPV/SB, P = 0.92).
Single Surgery Anatomical Success Based on Gauge Surgery The majority of patients in both groups underwent 23 G or 25 G vitrectomy surgery. Results were similar whether 20 G or smaller-gauge surgery was performed, with a SSAS of 82% and 81%, respectively (P = 0.971). Of the 42 patients in the PPV group undergoing 23 G or 25 G vitrectomy, 83% had a SSAS rate compared with 87% of the 15 patients in the PPV/ SB group (P = 0.76). Of the patients undergoing 20 G surgery, 83% in the PPV group and 86% in the PPV/ SB group had a SSAS (P = 0.91). Complications No visually threatening complications occurred in this study. Zero eyes had globe perforation during the SB procedure and no eye in either group had complications, such as endophthalmitis, choroidal hemorrhage, or elevated intraocular pressure requiring filtering surgery. Thirty-four patients in the PPV group and 9 patients in the PPV/SB group were phakic during the initial retinal detachment repair. Of the phakic patients, cataract surgery was performed during the follow-up period on 12 (35%) eyes in the PPV group and 9 (100%) eyes in the PPV/SB group (P , 0.001). The mean time to cataract surgery was 159 days in the PPV group and 156 days in the PPV/ SB group (P = 0.527). Thirty-five of the 43 patients who were phakic during the initial retinal detachment repair received gas tamponade. Of these, 8 of 23 patients (35%) who received SF6 and 8 of 12 (67%) who received C3F8 had subsequent cataract surgery (P = 0.072). Discussion Many factors, including surgeon experience and patient preoperative characteristics, influence the surgical
TREATMENT OF RETINAL DETACHMENT ORLIN ET AL
technique in RRD repair. Improving the success of RRDs remains an active area of research and the adjuvant use of an encircling SB to PPV remains controversial, offering potential advantages and disadvantages to PPV alone. In our academic institute where vitreoretinal fellows participate in key aspects of all cases, we found no difference in SSAS or final anatomical success between PPV alone when compared with PPV combined with an encircling sclera buckle in the repair of primary noncomplex RRD. In addition, there was no difference in the change in visual acuity when comparing PPV with PPV/SB in the repair of primary noncomplex RRD. Our single surgical anatomical success rate was 83% in the PPV group and 86% in the PPV/SB group while all patients (100%) were attached at the most recent follow-up (mean: 406.73 days in PPV group, 502.14 days in PPV/SB group). Of the eyes that redetached, all three in the PPV/SB obtained anatomical success with one additional operation at the most recent follow-up. Five eyes in the PPV group required one additional surgery for final reattachment while four eyes underwent two additional operations to achieve anatomical success at the most recent follow-up. All four of these eyes required subsequent silicone oil placement and all three of the phakic eyes underwent lensectomy. Our study included 5 eyes with a duration ,3 months and the range was 83 days to 1,974 days in the PPV/SB group, and 57 days to 1,226 days in the PPV group. Mansouri et al1 required a minimum follow-up of 1 month and found that the average time to redetachment was 45 days in the PPV/SB group and 35 days in the PPV group. Additionally, Kinori et al2 defined primary failure as redetachment observed within 8 weeks from the surgical procedure. Our minimum follow-up coincides with these studies and our mean follow-up is comparable to other studies.1,2,4,6,7 The baseline visual acuity was similar in the PPV and PPV/SB groups (P = 0.61). At the most recent follow-up, the mean Snellen visual acuity was 20/52.39 (0.418 logMAR) and 20/60.73 (0.479 logMAR) in the PPV and PPV/SB groups, respectively (P = 0.61). The majority of patients improved their vision at follow-up in both groups, and the mean change in vision was similar between both groups (0.467 logMAR in the PPV group and 0.411 logMAR in the PPV/SB group, P = 0.75). There was also no difference in the percentage of eyes improving by 3 or more lines of Snellen visual acuity between the PPV and PPV/SB groups (57 vs. 50%, P = 0.58). Seventeen percent of eyes in the PPV group had inferior retinal tears while 54% of eyes in the PPV/SB group had inferior breaks (P = 0.001). This discrep-
1073
ancy is indicative of the nonrandomization in our study and is related to the potential bias that internal tamponade alone may be inadequate to treat inferior breaks. In our patient population as a whole, there was a trend for surgeries involving inferior breaks to have a lower SSAS compared with those without inferior breaks (71 vs. 89%), although this did not reach statistical significance (P = 0.07). There were few total eyes (n = 21) containing inferior breaks in our study, and each redetachment influences the results dramatically. It is possible with a larger sample containing more eyes with inferior breaks, that the success rates in this subgroup would have been different. When evaluated further, there was no difference in SSAS amongst patients with inferior retinal breaks between the PPV and PPV/SB groups (67 vs. 75%, P = 0.68). In our sample, it did not seem that the addition of an encircling band improved SSAS in eyes with inferior retinal breaks. Alexander et al7 proposed that the supplementary buckling enhances primary success rates in inferior break detachments and obtained a 95% SSAS at 3 months. In their study, however, there was no control group with PPV alone, and the results were compared with previously published literature. In comparing the results with previously published literature, potential flaws occur including varying baseline patient characteristics, surgical technique, and surgeon experience between the studies. In addition, 3 months is a relatively short follow-up when compared with other studies including ours, and it is possible that late redetachment would have influenced their results differently. Other authors have questioned the need for combined surgery for RRD involving inferior breaks,2,10 and there does not seem to be clear evidence that the addition of an encircling buckle improves the success rates, which is in accordance with our results. Future, larger comparative studies should attempt to resolve this area of uncertainty and the results may also be applicable in choosing a preferred technique for patients who may have poor compliance with postoperative positioning. In our cohort, 65% of eyes in the PPV group and 41% of eyes in the PPV/SB group were phakic (P = 0.05). It is likely that a larger portion of pseudophakic and aphakic eyes underwent combined PPV/SB surgery secondary to the concern that smaller peripheral breaks would be present and not always easily visualized in these eyes. Ninety-four percent of pseudophakic/aphakic eyes and 78% of phakic eyes had SSAS (P = 0.053). This increased redetachment rate in phakic eyes is possibly the result of inadequate peripheral vitrectomy performed. It is also possible that an encircling band could relieve anterior traction not easily removed with vitrectomy in phakic eyes,
1074 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
and can help support small breaks not easily visualized in pseudophakic/aphakic eyes. However, in our cohort, there was no difference in SSAS amongst phakic patients between the PPV and PPV/SB groups (79 vs. 78%, P = 0.92). Similarly, we found no difference in SSAS amongst pseudophakic or aphakic patients between the PPV and PPV/SB groups (89 vs. 92%, P = 0.75). In our study, the adjuvant of an encircling band did not increase SSAS in the pseudophakic/ aphakic or phakic eye groups. Kinori et al2 did not obtain superior results with PPV/SB when compared with PPV alone based on lens status, and they also concluded that the presence of a crystalline lens is not an indication for adding a SB to PPV. Weichel et al4 similarly found no difference in SSAS between PPV and PPV/SB in repairing pseudophakic RRD. Very few eyes in our study had retinal detachments in which no clear retinal break was visualized. In our PPV and PPV/SB groups, 6% (n = 3) and 18% (n = 4) respectively, did not have a retinal break visualized (n = 0.10). Seven of these 7 eyes (100%) obtained SSAS and no difference was obtained between the PPV and PPV/ SB groups. Given the few eyes, it is not possible for us to draw adequate conclusions on whether an encircling band should be added to PPV to treat eyes with RRD and no visible break. In a pilot study, Desai and Strassman6 demonstrated good anatomical results when performing PPV/SB on 10 eyes (100% SSAS). No PPV control group was analyzed in their small sample, and as we illustrated, 100% of our eyes (when no tear was visualized) obtained SSAS regardless of whether an encircling band was used. Other authors also found no difference in SSAS based on whether a tear was visualized or not.2 Vitrectomy was characterized as being either largegauge (20 G) or small-gauge (23 G or 25 G) surgery. The 23 G and 25 G surgery was combined to characterize small-gauge surgery secondary to the few eyes undergoing each separately. The majority of eyes in both PPV and PPV/SB groups underwent small-gauge surgery (P = 0.065). We found similar SSAS between small-gauge (84%) and large-gauge (85%) surgery (P = 0.93). There was no difference in SSAS when further characterized based on whether PPV or PPV/SB was performed. The evolution of smaller-gauge vitrectomy instruments has been a technological advance in vitrectomy surgery offering advantages, such as improved fluidics, cut rates, smaller cutter diameters, faster wound healing, diminished conjunctival scarring, and improved patient comfort.11 A potential disadvantage includes wound leak resulting in poor gas or oil tamponade fill and lower SSAS. We along with others11,12 found similar success rates between small-gauge and largegauge surgery. In addition, there was no difference in
2014 VOLUME 34 NUMBER 6
SSAS in the PPV or PPV/SB groups based on whether small-gauge or large-gauge vitrectomy was performed. Four eyes in the PPV group and four in the PPV/SB group underwent surgery with oil tamponade. Of these, two eyes had oil placement because the patient required air travel soon after the surgery. Two additional eyes had oil placed because, preoperatively, the patients were felt to be poor positioners and their detachments contained multiple tears including inferior ones. One eye had oil placed because it was highly myopic with an inferior tear, and the patient was amblyopic in the contralateral eye. Three eyes had oil placed because they contained multiple tears including inferior ones, and had significant areas of peripheral lattice and scarring. Seven of the eight eyes subsequently had oil removal and all had a SSAS. We would expect redetachment after oil removal if the primary surgery was not successful just as would be seen after the disappearance of the gas bubble. Additionally, we found no difference between the PPV and PPV/SB groups in eyes undergoing oil placement (P = 0.43). Therefore, we do not think our results are drastically impacted in that oil was used in certain patients. We did not encounter any serious visually threatening complications during or after scleral buckling, such as globe perforation, endophthalmitis, choroidal hemorrhage, or intraocular pressure elevation requiring filtration surgery in any of our patients. Of the phakic eyes, 35% (n = 12) in the PPV group and 100% (n = 9) in the PPV/SB group had cataract extraction during follow-up (P , 0.001). Cataract progression is a common sequelae of vitrectomy surgery,2 and the high rates of cataract extraction in our series is not unexpected. It is unclear why those in the PPV/SB group underwent more cataract extraction. Although not statistically significant, the PPV/SB group was followed for a longer period of time (502.14 days) when compared with the PPV group (406.73 days). It is possible that in time, a larger number of eyes in the PPV group would require cataract extraction. In addition, a larger sample in our cohort may also lead to similar cataract extraction rate between the groups. There was a trend for phakic eyes receiving C3F8 tamponade to undergo subsequent cataract extraction more often when compared with phakic eyes receiving SF6 as a tamponade (67 vs. 35%, P = 0.072). C3F8 lasts longer than SF6 which likely results in more progression of cataract formation. There are several limitations to this study. Performing a large, randomized prospective surgical trial is difficult, and ours is retrospective and nonrandomized. We did, however, attempt to identify any baseline
TREATMENT OF RETINAL DETACHMENT ORLIN ET AL
differences between the PPV and PPV/SB groups and further analyze their impact on surgical outcomes which is detailed above. Another limitation is that multiple surgeons (five) participated in this study. Surgeon variability may, however, be clinically applicable to other academic departments that use multiple surgeons and the key participation of vitreoretinal fellows. It is also possible with longer follow-up that further differences in outcomes between the PPV and PPV/SB groups would have been seen. Our follow-up duration, however, is comparable to other studies addressing the subject matter. To conclude, we found no difference in SSAS, final anatomical success, or change in visual acuity when comparing PPV with PPV/SB in the repair of primary noncomplex RRD in an academic setting where vitreoretinal fellows participate in key aspects of all cases. Fellow participation in key aspects of the cases resulted in excellent outcomes in both groups. In addition to the academic setting in which this trial takes place, it is the first time to our knowledge that new smaller-gauge instrumentation is also evaluated in the analysis when comparing PPV with PPV/SB. Although scleral buckling was a safe adjunct in our series, no added benefit could be demonstrated in our cohort as a whole, or when characterizing the eyes based on lens status, tear location, gauge of instrumentation used, or whether a break was visualized preoperatively. Certain disadvantages described elsewhere of scleral buckling include longer operation and anesthesia times, globe perforation, and the risk of developing postoperative problems, such as diplopia,8 buckle extrusion, or infection. Key words: retinal detachment, pars plana vitrectomy, scleral buckle, retinal detachment surgery.
1075
References 1. Mansouri A, Almony A, Shah GK, et al. Recurrent retinal detachment: does initial treatment matter? Br J Ophthalmol 2010;94:1344–1347. 2. Kinori M, Moisseiev E, Shoshany N, et al. Comparison of pars plana vitrectomy with and without scleral buckle for the repair of primary rhegmatogenous retinal detachment. Am J Ophthalmol 2011;152:291–297.e2. 3. SPR Study Group. View 2: the case for primary vitrectomy. Br J Ophthalmol 2003;87:784–787. 4. Weichel ED, Martidis A, Fineman MS. Pars plana vitrectomy versus combined pars plana vitrectomy-scleral buckle for primary repair of pseudophakic retinal detachment. Ophthalmology 2006;113:2033–2040. 5. Devenyi RG, de Carvalho Nakamura H. Combined scleral buckle and pars plana vitrectomy as a primary procedure for pseudophakic retinal detachments. Ophthalmic Surg Lasers 1999;30:615–618. 6. Desai UR, Strassman IB. Combined pars plana vitrectomy and scleral buckling for pseudophakic and aphakic retinal detachments in which a break is not seen preoperatively. Ophthalmic Surg Lasers 1997;28:718–722. 7. Alexander P, Ang A, Poulson A, et al. Scleral buckling combined with vitrectomy for the management of rhegmatogenous retinal detachment associated with inferior retinal breaks. Eye (Lond) 2008;22:200–203. 8. Cooper LL, Harrison S, Rosenbaum AL. Ocular torsion as a complication of scleral buckle procedures for retinal detachments. J AAPOS 1998;2:279–284. 9. 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;114:2142–2154. 10. Mokete B, Williamson TH. Scleral buckling combined with vitrectomy for the management of rhegmatogenous retinal detachment associated with inferior retinal breaks. Eye (Lond) 2009;23:1233; author reply 1233–1234. 11. Misra A, Ho-Yen G, Burton RL. 23-gauge sutureless vitrectomy and 20 gauge vitrectomy: a case series comparison. Eye (Lond) 2009;23:1187–1191. 12. Miller DM, Riemann CD, Foster RE, et al. Primary repair of retinal detachment with 25-gauge pars plana vitrectomy. Retina 2008;28:931–936.
R
PPV remains controversial2 and offers potential advantages and disadvantages to PPV alone. Some surgeons propose that the addition of a SB results in improved success rate, particularly when the retinal detachment occurs in pseudophakic eyes,5 involves unseen breaks,6 or involves inferior pathology.7 It is also possible that an encircling band may improve success in phakic patients by relieving anterior traction, because of the difficulty of performing adequate
hegmatogenous retinal detachment (RRD) remains a significant cause of visual morbidity.1 The most common surgical techniques for the repair of retinal detachment include scleral buckling (SB) and pars plana vitrectomy (PPV). The recent advances in vitrectomy instrumentation and technique have influenced many retinal surgeons to prefer the use of PPV for the repair of RRD in select cases.2–4 The adjuvant use of an encircling scleral buckle (SB) to 1069
1070 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
peripheral vitrectomy in these eyes. Certain disadvantages include longer operation and anesthesia times, globe perforation, and the risk of developing postoperative problems, such as diplopia,8 buckle extrusion, or infection. The purpose of this study is to 1) determine whether there is a difference in success rates between PPV and PPV combined with SB (PPV/SB) in the treatment of primary noncomplex RRD in an academic setting where vitreoretinal fellows participate in key aspects of all cases; and 2) investigate whether certain baseline characteristics of the RRD were associated with different outcomes between the two surgical techniques. In addition to the setting in which this study takes place, we also evaluate the use of smaller-gauge instrumentation in this analysis. Methods This retrospective review was approved by the Institutional Review Board of Weill Cornell Medical College (WCMC). The research adhered to the tenets of the Declaration of Helsinki and was conducted in accordance with the Health Insurance Portability and Accountability Act regulations. Charts of patients who were operated on at New York Presbyterian Hospital, WCMC were reviewed, and patients diagnosed with primary RRD who underwent PPV or PPV combined with SB between 2006 and August 2011 were identified. Patients were excluded if their RRD was repaired with SB or pneumatic retinopexy alone. Additional exclusion criteria was a documented follow-up ,2 months, previous vitreoretinal surgery, giant tears, retinal detachment in patients younger than 18 years, the presence of any type of proliferative vitreoretinopathy, traction retinal detachment resulting from proliferative diabetic retinopathy, or RRD in the setting of infectious retinitis. After reviewing the data, we found that of the noncomplex detachments repaired by current faculty, additional eyes were excluded because of the following: Roughly 11 eyes had primary buckles, 58 patients were classified as noncomplicated detachments (no proliferFrom the *Department of Ophthalmology, Weill Cornell Medical College, New York, New York; and †Department of Ophthalmology, Campus Benjamin Franklin, Charité - Universitaetsmedizin Berlin, Berlin, Germany. Paper presented at the Association for Research in Vision and Ophthalmology Annual Meeting, Fort Lauderdale, FL, May 5, 2011. Supported by the St. Giles Foundation (R.V.P.C.) and by the Research to Prevent Blindness (R.V.P.C., A.O., S.K., and D.J.D.). None of the authors have any financial/conflicting interests to disclose. Reprint requests: R. V. Paul Chan, MD, Department of Ophthalmology, Weill Cornell Medical College, 1305 York Avenue, 11th Floor, Room Y 11.34, New York, NY 10021; e-mail: roc9013@med. cornell.edu
2014 VOLUME 34 NUMBER 6
ative vitreoretinopathy) but had received previous laser or cryo or PPV surgery (elsewhere or at Weill Cornell), 12 patients had inadequate follow-up, 10 eyes had giant retinal tear (GRT) repair or previous trauma. All surgery was performed at New York Presbyterian Hospital by an attending vitreoretinal surgeon who is currently on staff at WCMC and directly involved with vitreoretinal fellowship training. Any surgery performed by a vitreoretinal surgeon who no longer is affiliated with WCMC or not involved with vitreoretinal fellowship training was excluded. Vitreoretinal fellows participated in key aspects of all cases. All vitreoretinal attending surgeons involved with the training of fellows at WCMC, New York Presbyterian Hospital were included in this study. Seventy-four consecutive eyes from 74 patients were included in this study. The main outcome measure was single surgery anatomical success (SSAS). Bestcorrected Snellen visual acuity (BCVA) was recorded and converted to logarithm of the minimal angle of resolution (logMAR). Preoperative data was recorded, such as the lens status, location of retinal tear whether a tear was visualized and if multiple tears were present, macula status, presence of peripheral lattice retinal degeneration, and BCVA. Data from the most recent follow-up visit was recorded including the duration of follow-up, final BCVA, SSAS, final anatomical success, and whether cataract extraction was performed. Data collected from patients who redetached included the number of redetachments and subsequent procedures performed. Visual acuity corresponding to counting fingers, hand motion, light perception, and no light perception corresponded to 1.7, 2.0, 2.5, and 3.0 logMAR, respectively.2,9
Surgical Technique All patients underwent standard 3-port 20 G, 23 G, or 25 G PPV using a wide-angle contact viewing system. Vitrectomy was characterized as being either large-gauge (20 G) or small-gauge (23 G or 25 G) surgery. The 23 G and 25 G surgery was combined to characterize small-gauge surgery secondary to the few eyes undergoing each separately when compared with the traditionally larger 20 G surgery. Endolaser photocoagulation was applied to completely surround all retinal breaks. All 20 G sclerotomy sites were sutured with 7-0 vicryl suture. Any 23 G or 25 G sclerotomy sites that were found to be leaking at the end of the surgery were sutured with 7-0 vicryl suture. A 360° encircling band (Band 41, Watzke Sleeve 270) was used in the PPV/SB group. Intraocular tamponade was achieved with nonexpansile gas (16% C3F8 or 20% SF6) or with 1,000 centistokes of silicone oil.
1071
TREATMENT OF RETINAL DETACHMENT ORLIN ET AL Table 1. Baseline Characteristics of Patients Undergoing PPV Alone and PPV Combined With SB PPV Alone (n = 52) Mean Age, years Women, n (%) Phakic, n (%) Right eye, n (%) Macula off, n (%) Presence of lattice degeneration, n (%) Retinal break visualized, n (%) Multiple breaks present, n (%) Inferior break present, n (%) Preoperative mean visual acuity, logMAR (Snellen)
26 34 27 31 16 49 24 9 0.903
Statistical Analysis All statistical analyses were performed using SPSS (IBM Corp, New York, NY) and Excel (Microsoft Inc, Redmond, WA). Independent samples t-test and chi-square test were used as appropriate. P , 0.05 was considered statistically significant. Results Preoperative and Intraoperative Data Seventy-four consecutive eyes from 74 patients were included in this study. Fifty-two eyes underwent PPV alone while 22 eyes had PPV combined with SB to repair the primary RRD. Preoperative characteristics are described in Table 1. Patients in the PPV/SB group were less likely to be phakic (41 vs. 65%; P = 0.05) and more likely to have an inferior retinal break (54 vs. 17%; P = 0.001) when compared with the PPV alone group. The mean age at presentation was 58.48 years in the PPV group (range, 27–80 years) versus 55.68 years in the PPV/SB group (range, 19–79 years) (P = 0.447). Between the two groups, there was no difference in the percent of patients with peripheral retinal lattice degeneration (P = 0.929), multiple breaks (P = 0.801), or those presenting with the macula off retinal Table 2. Intraoperative Characteristics in Patients of Patients Undergoing PPV Alone and PPV Combined With SB
Gauge surgery Large gauge 20 G, n (%) Small gauge Total, n (%) 23 G, n (%) 25 G, n (%) Gas tamponade used, n (%)
PPV Alone
PPV/SB
P
6 (13)
7 (32)
0.054
42 32 10 48
(88) (67) (21) (92)
15 15 0 19
(68) (68) (0) (86)
0.054 0.900 0.021 0.43
54.49 (50) (65) (52) (61) (31) (94) (46) (17) (20/159.83)
PPV with SB (n = 22) 8 9 9 15 7 18 10 12 0.890
55.68 (36) (41) (41) (68) (32) (82) (45) (54) (20/155.29)
P 0.45 0.28 0.05 0.39 0.55 0.93 0.10 0.80 0.001 0.943
detachment (P = 0.548). Although not reaching statistical significance, there was a trend for examiners to visualize a retinal break preoperatively in the PPV alone group (94 vs. 82%; P = 0.095). The intraoperative data is summarized in Table 2. The majority of patients in both PPV alone and SB/PPV groups received gas tamponade (P = 0.43). Also, the majority of eyes underwent small-gauge surgery (23 G or 25 G) with no statistically significant difference between the groups (P = 0.065) regarding the selection of large-gauge versus small-gauge instrumentation. Long-term mean follow-up was obtained in both groups with a mean of 406.73 days in the PPV group (range, 57– 1,226 days) and a mean of 502.14 days in the PPV/SB group (range, 83–1874 days). Each patient was followed for a minimum of 2 months. Postoperative characteristics at most recent follow-up are summarized in Table 3. Visual Acuity There was no difference in baseline BCVA between the groups. The mean preoperative BCVA was 0.903 logMAR (20/159.83 Snellen BCVA) and 0.890 logMAR (20/155.29 Snellen BCVA) in the PPV and PPV/SB groups, respectively (P = 0.943). Mean BCVA improved in both groups (P = 0.75), with a final BCVA of 0.418 logMAR (20/52.39 Snellen BCVA) in the PPV group and 0.479 logMAR (20/60.37 Snellen BCVA) in the PPV/SB group (P = 0.61). The degree of visual acuity improvement was further characterized in Table 3 and there was no difference between the groups in the percent of patients improving vision (P = 0.23) or in patients improving vision by at least 3 Snellen chart lines (P = 0.58). Anatomical Success Full attachment of the retina was obtained in 100% of the patients in both groups at most recent follow-up. Single surgery anatomical success rates were similar
1072 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 6
Table 3. Characteristics at Most Recent Follow-up of Patients Undergoing PPV Alone and PPV Combined With SB
Mean follow-up, days BCVA, logMAR (Snellen) Change in VA, logMAR No. patients having improved or stable vision, n (%) No. patients improving by $3 lines Snellen VA, n (%) SSAS, n (%) Final anatomical success, n (%) No. retinal detachment reoperations Phakic patients undergoing cataract extraction, n (%)
PPV Alone
PPV/SB
P
406.73 0.418 (20/52.39) 0.467 45 (88)
502.14 0.479 (20/60.37) 0.411 17 (77)
0.31 0.61 0.75 0.23
29 (57) 43 (83) 52 (100) 13 for 9 eyes, 1 surgery for 5 eyes, 2 surgeries for 4 eyes 12 (35)
11 (50)
0.58
19 (86) 0.70 22 (100) 1.00 3 for 3 eyes, 1 surgery for 0.695 3 eyes 9 (100) ,0.001
VA, visual acuity.
between the groups (83% PPV vs. 86% PPV/SB; P = 0.695). Among the eyes that had a redetachment, the majority of patients required only one subsequent surgery to obtain anatomical success (five patients in the PPV group and all three patients in the PPV/SB group). Four eyes in the PPV group that had a redetachment required two subsequent surgeries to achieve retinal attachment. Single Surgery Anatomical Success Based on Tear Location Overall, patients with inferior retinal breaks underwent more PPV/SB surgery compared with PPV alone (P = 0.001). Surgeries involving inferior retinal breaks had a trend to have a lower SSAS than those without inferior breaks (71 vs. 89%, P = 0.07), although not statistically significant. Amongst surgeries involving inferior breaks, SSAS was achieved in 67% of PPV eyes compared with 75% of PPV/SB eyes (P = 0.68). Preoperatively, a retinal break was not visualized in 9% of all surgeries, with no difference between surgical groups (P = 0.10). Amongst these patients, a 100% SSAS was found in both groups (n = 3 in PPV group, n = 4 in PPV/SB group). Single Surgery Anatomical Success Based on Lens Status Pseudophakic patients were more likely to undergo PPV/SB surgery (P = 0.05), and SSAS was higher for pseudophakic patients when compared with phakic ones (94 vs. 78%, P = 0.05). However, there was no difference in SSAS amongst pseudophakic patients between the PPV and PPV/SB groups (89% PPV vs. 92% PPV/SB; P = 0.75). Similarly, no difference in SSAS was demonstrated amongst phakic patients (79% PPV vs. 78% PPV/SB, P = 0.92).
Single Surgery Anatomical Success Based on Gauge Surgery The majority of patients in both groups underwent 23 G or 25 G vitrectomy surgery. Results were similar whether 20 G or smaller-gauge surgery was performed, with a SSAS of 82% and 81%, respectively (P = 0.971). Of the 42 patients in the PPV group undergoing 23 G or 25 G vitrectomy, 83% had a SSAS rate compared with 87% of the 15 patients in the PPV/ SB group (P = 0.76). Of the patients undergoing 20 G surgery, 83% in the PPV group and 86% in the PPV/ SB group had a SSAS (P = 0.91). Complications No visually threatening complications occurred in this study. Zero eyes had globe perforation during the SB procedure and no eye in either group had complications, such as endophthalmitis, choroidal hemorrhage, or elevated intraocular pressure requiring filtering surgery. Thirty-four patients in the PPV group and 9 patients in the PPV/SB group were phakic during the initial retinal detachment repair. Of the phakic patients, cataract surgery was performed during the follow-up period on 12 (35%) eyes in the PPV group and 9 (100%) eyes in the PPV/SB group (P , 0.001). The mean time to cataract surgery was 159 days in the PPV group and 156 days in the PPV/ SB group (P = 0.527). Thirty-five of the 43 patients who were phakic during the initial retinal detachment repair received gas tamponade. Of these, 8 of 23 patients (35%) who received SF6 and 8 of 12 (67%) who received C3F8 had subsequent cataract surgery (P = 0.072). Discussion Many factors, including surgeon experience and patient preoperative characteristics, influence the surgical
TREATMENT OF RETINAL DETACHMENT ORLIN ET AL
technique in RRD repair. Improving the success of RRDs remains an active area of research and the adjuvant use of an encircling SB to PPV remains controversial, offering potential advantages and disadvantages to PPV alone. In our academic institute where vitreoretinal fellows participate in key aspects of all cases, we found no difference in SSAS or final anatomical success between PPV alone when compared with PPV combined with an encircling sclera buckle in the repair of primary noncomplex RRD. In addition, there was no difference in the change in visual acuity when comparing PPV with PPV/SB in the repair of primary noncomplex RRD. Our single surgical anatomical success rate was 83% in the PPV group and 86% in the PPV/SB group while all patients (100%) were attached at the most recent follow-up (mean: 406.73 days in PPV group, 502.14 days in PPV/SB group). Of the eyes that redetached, all three in the PPV/SB obtained anatomical success with one additional operation at the most recent follow-up. Five eyes in the PPV group required one additional surgery for final reattachment while four eyes underwent two additional operations to achieve anatomical success at the most recent follow-up. All four of these eyes required subsequent silicone oil placement and all three of the phakic eyes underwent lensectomy. Our study included 5 eyes with a duration ,3 months and the range was 83 days to 1,974 days in the PPV/SB group, and 57 days to 1,226 days in the PPV group. Mansouri et al1 required a minimum follow-up of 1 month and found that the average time to redetachment was 45 days in the PPV/SB group and 35 days in the PPV group. Additionally, Kinori et al2 defined primary failure as redetachment observed within 8 weeks from the surgical procedure. Our minimum follow-up coincides with these studies and our mean follow-up is comparable to other studies.1,2,4,6,7 The baseline visual acuity was similar in the PPV and PPV/SB groups (P = 0.61). At the most recent follow-up, the mean Snellen visual acuity was 20/52.39 (0.418 logMAR) and 20/60.73 (0.479 logMAR) in the PPV and PPV/SB groups, respectively (P = 0.61). The majority of patients improved their vision at follow-up in both groups, and the mean change in vision was similar between both groups (0.467 logMAR in the PPV group and 0.411 logMAR in the PPV/SB group, P = 0.75). There was also no difference in the percentage of eyes improving by 3 or more lines of Snellen visual acuity between the PPV and PPV/SB groups (57 vs. 50%, P = 0.58). Seventeen percent of eyes in the PPV group had inferior retinal tears while 54% of eyes in the PPV/SB group had inferior breaks (P = 0.001). This discrep-
1073
ancy is indicative of the nonrandomization in our study and is related to the potential bias that internal tamponade alone may be inadequate to treat inferior breaks. In our patient population as a whole, there was a trend for surgeries involving inferior breaks to have a lower SSAS compared with those without inferior breaks (71 vs. 89%), although this did not reach statistical significance (P = 0.07). There were few total eyes (n = 21) containing inferior breaks in our study, and each redetachment influences the results dramatically. It is possible with a larger sample containing more eyes with inferior breaks, that the success rates in this subgroup would have been different. When evaluated further, there was no difference in SSAS amongst patients with inferior retinal breaks between the PPV and PPV/SB groups (67 vs. 75%, P = 0.68). In our sample, it did not seem that the addition of an encircling band improved SSAS in eyes with inferior retinal breaks. Alexander et al7 proposed that the supplementary buckling enhances primary success rates in inferior break detachments and obtained a 95% SSAS at 3 months. In their study, however, there was no control group with PPV alone, and the results were compared with previously published literature. In comparing the results with previously published literature, potential flaws occur including varying baseline patient characteristics, surgical technique, and surgeon experience between the studies. In addition, 3 months is a relatively short follow-up when compared with other studies including ours, and it is possible that late redetachment would have influenced their results differently. Other authors have questioned the need for combined surgery for RRD involving inferior breaks,2,10 and there does not seem to be clear evidence that the addition of an encircling buckle improves the success rates, which is in accordance with our results. Future, larger comparative studies should attempt to resolve this area of uncertainty and the results may also be applicable in choosing a preferred technique for patients who may have poor compliance with postoperative positioning. In our cohort, 65% of eyes in the PPV group and 41% of eyes in the PPV/SB group were phakic (P = 0.05). It is likely that a larger portion of pseudophakic and aphakic eyes underwent combined PPV/SB surgery secondary to the concern that smaller peripheral breaks would be present and not always easily visualized in these eyes. Ninety-four percent of pseudophakic/aphakic eyes and 78% of phakic eyes had SSAS (P = 0.053). This increased redetachment rate in phakic eyes is possibly the result of inadequate peripheral vitrectomy performed. It is also possible that an encircling band could relieve anterior traction not easily removed with vitrectomy in phakic eyes,
1074 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
and can help support small breaks not easily visualized in pseudophakic/aphakic eyes. However, in our cohort, there was no difference in SSAS amongst phakic patients between the PPV and PPV/SB groups (79 vs. 78%, P = 0.92). Similarly, we found no difference in SSAS amongst pseudophakic or aphakic patients between the PPV and PPV/SB groups (89 vs. 92%, P = 0.75). In our study, the adjuvant of an encircling band did not increase SSAS in the pseudophakic/ aphakic or phakic eye groups. Kinori et al2 did not obtain superior results with PPV/SB when compared with PPV alone based on lens status, and they also concluded that the presence of a crystalline lens is not an indication for adding a SB to PPV. Weichel et al4 similarly found no difference in SSAS between PPV and PPV/SB in repairing pseudophakic RRD. Very few eyes in our study had retinal detachments in which no clear retinal break was visualized. In our PPV and PPV/SB groups, 6% (n = 3) and 18% (n = 4) respectively, did not have a retinal break visualized (n = 0.10). Seven of these 7 eyes (100%) obtained SSAS and no difference was obtained between the PPV and PPV/ SB groups. Given the few eyes, it is not possible for us to draw adequate conclusions on whether an encircling band should be added to PPV to treat eyes with RRD and no visible break. In a pilot study, Desai and Strassman6 demonstrated good anatomical results when performing PPV/SB on 10 eyes (100% SSAS). No PPV control group was analyzed in their small sample, and as we illustrated, 100% of our eyes (when no tear was visualized) obtained SSAS regardless of whether an encircling band was used. Other authors also found no difference in SSAS based on whether a tear was visualized or not.2 Vitrectomy was characterized as being either largegauge (20 G) or small-gauge (23 G or 25 G) surgery. The 23 G and 25 G surgery was combined to characterize small-gauge surgery secondary to the few eyes undergoing each separately. The majority of eyes in both PPV and PPV/SB groups underwent small-gauge surgery (P = 0.065). We found similar SSAS between small-gauge (84%) and large-gauge (85%) surgery (P = 0.93). There was no difference in SSAS when further characterized based on whether PPV or PPV/SB was performed. The evolution of smaller-gauge vitrectomy instruments has been a technological advance in vitrectomy surgery offering advantages, such as improved fluidics, cut rates, smaller cutter diameters, faster wound healing, diminished conjunctival scarring, and improved patient comfort.11 A potential disadvantage includes wound leak resulting in poor gas or oil tamponade fill and lower SSAS. We along with others11,12 found similar success rates between small-gauge and largegauge surgery. In addition, there was no difference in
2014 VOLUME 34 NUMBER 6
SSAS in the PPV or PPV/SB groups based on whether small-gauge or large-gauge vitrectomy was performed. Four eyes in the PPV group and four in the PPV/SB group underwent surgery with oil tamponade. Of these, two eyes had oil placement because the patient required air travel soon after the surgery. Two additional eyes had oil placed because, preoperatively, the patients were felt to be poor positioners and their detachments contained multiple tears including inferior ones. One eye had oil placed because it was highly myopic with an inferior tear, and the patient was amblyopic in the contralateral eye. Three eyes had oil placed because they contained multiple tears including inferior ones, and had significant areas of peripheral lattice and scarring. Seven of the eight eyes subsequently had oil removal and all had a SSAS. We would expect redetachment after oil removal if the primary surgery was not successful just as would be seen after the disappearance of the gas bubble. Additionally, we found no difference between the PPV and PPV/SB groups in eyes undergoing oil placement (P = 0.43). Therefore, we do not think our results are drastically impacted in that oil was used in certain patients. We did not encounter any serious visually threatening complications during or after scleral buckling, such as globe perforation, endophthalmitis, choroidal hemorrhage, or intraocular pressure elevation requiring filtration surgery in any of our patients. Of the phakic eyes, 35% (n = 12) in the PPV group and 100% (n = 9) in the PPV/SB group had cataract extraction during follow-up (P , 0.001). Cataract progression is a common sequelae of vitrectomy surgery,2 and the high rates of cataract extraction in our series is not unexpected. It is unclear why those in the PPV/SB group underwent more cataract extraction. Although not statistically significant, the PPV/SB group was followed for a longer period of time (502.14 days) when compared with the PPV group (406.73 days). It is possible that in time, a larger number of eyes in the PPV group would require cataract extraction. In addition, a larger sample in our cohort may also lead to similar cataract extraction rate between the groups. There was a trend for phakic eyes receiving C3F8 tamponade to undergo subsequent cataract extraction more often when compared with phakic eyes receiving SF6 as a tamponade (67 vs. 35%, P = 0.072). C3F8 lasts longer than SF6 which likely results in more progression of cataract formation. There are several limitations to this study. Performing a large, randomized prospective surgical trial is difficult, and ours is retrospective and nonrandomized. We did, however, attempt to identify any baseline
TREATMENT OF RETINAL DETACHMENT ORLIN ET AL
differences between the PPV and PPV/SB groups and further analyze their impact on surgical outcomes which is detailed above. Another limitation is that multiple surgeons (five) participated in this study. Surgeon variability may, however, be clinically applicable to other academic departments that use multiple surgeons and the key participation of vitreoretinal fellows. It is also possible with longer follow-up that further differences in outcomes between the PPV and PPV/SB groups would have been seen. Our follow-up duration, however, is comparable to other studies addressing the subject matter. To conclude, we found no difference in SSAS, final anatomical success, or change in visual acuity when comparing PPV with PPV/SB in the repair of primary noncomplex RRD in an academic setting where vitreoretinal fellows participate in key aspects of all cases. Fellow participation in key aspects of the cases resulted in excellent outcomes in both groups. In addition to the academic setting in which this trial takes place, it is the first time to our knowledge that new smaller-gauge instrumentation is also evaluated in the analysis when comparing PPV with PPV/SB. Although scleral buckling was a safe adjunct in our series, no added benefit could be demonstrated in our cohort as a whole, or when characterizing the eyes based on lens status, tear location, gauge of instrumentation used, or whether a break was visualized preoperatively. Certain disadvantages described elsewhere of scleral buckling include longer operation and anesthesia times, globe perforation, and the risk of developing postoperative problems, such as diplopia,8 buckle extrusion, or infection. Key words: retinal detachment, pars plana vitrectomy, scleral buckle, retinal detachment surgery.
1075
References 1. Mansouri A, Almony A, Shah GK, et al. Recurrent retinal detachment: does initial treatment matter? Br J Ophthalmol 2010;94:1344–1347. 2. Kinori M, Moisseiev E, Shoshany N, et al. Comparison of pars plana vitrectomy with and without scleral buckle for the repair of primary rhegmatogenous retinal detachment. Am J Ophthalmol 2011;152:291–297.e2. 3. SPR Study Group. View 2: the case for primary vitrectomy. Br J Ophthalmol 2003;87:784–787. 4. Weichel ED, Martidis A, Fineman MS. Pars plana vitrectomy versus combined pars plana vitrectomy-scleral buckle for primary repair of pseudophakic retinal detachment. Ophthalmology 2006;113:2033–2040. 5. Devenyi RG, de Carvalho Nakamura H. Combined scleral buckle and pars plana vitrectomy as a primary procedure for pseudophakic retinal detachments. Ophthalmic Surg Lasers 1999;30:615–618. 6. Desai UR, Strassman IB. Combined pars plana vitrectomy and scleral buckling for pseudophakic and aphakic retinal detachments in which a break is not seen preoperatively. Ophthalmic Surg Lasers 1997;28:718–722. 7. Alexander P, Ang A, Poulson A, et al. Scleral buckling combined with vitrectomy for the management of rhegmatogenous retinal detachment associated with inferior retinal breaks. Eye (Lond) 2008;22:200–203. 8. Cooper LL, Harrison S, Rosenbaum AL. Ocular torsion as a complication of scleral buckle procedures for retinal detachments. J AAPOS 1998;2:279–284. 9. 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;114:2142–2154. 10. Mokete B, Williamson TH. Scleral buckling combined with vitrectomy for the management of rhegmatogenous retinal detachment associated with inferior retinal breaks. Eye (Lond) 2009;23:1233; author reply 1233–1234. 11. Misra A, Ho-Yen G, Burton RL. 23-gauge sutureless vitrectomy and 20 gauge vitrectomy: a case series comparison. Eye (Lond) 2009;23:1187–1191. 12. Miller DM, Riemann CD, Foster RE, et al. Primary repair of retinal detachment with 25-gauge pars plana vitrectomy. Retina 2008;28:931–936.
