Anatomic and Visual Outcomes of SB VS Primary PPV in Pseudophakic and Aphakic RD

8/7/2019 Anatomic and Visual Outcomes of SB VS Primary PPV in Pseudophakic and Aphakic RD

1/10

Anatomic and Visual Outcomes of ScleralBuckling versus Primary Vitrectomy inPseudophakic and Aphakic Retinal DetachmentSix-Month Follow-up Results of a Single OperationReport No. 1Hamid Ahmadieh, MD,1 Siamak Moradian, MD,1 Hooshang Faghihi, MD,2 Mohammad M. Parvaresh, MD,3

Heshmatollah Ghanbari, MD,4 Morsal Mehryar, MD,5 Ebadollah Heidari, MD,6 Hasan Behboudi, MD,7

Touka Banaee, MD, 8 Banafsheh Golestan, PhD,1 for the Pseudophakic and Aphakic Retinal Detachment (PARD) StudyGroup*

Purpose: To compare the anatomic and visual results and complications of conventional scleral bucklingversus primary vitrectomy for management of pseudophakic and aphakic retinal detachment.

Design: Prospective, randomized, multicenter clinical trial. Participants: Two hundred twenty-ve eyes of 225 patients with pseudophakic or aphakic retinal detachment.

Intervention: Eligible eyes were assigned randomly either to conventional scleral buckling or primaryvitrectomy without any buckle. Main Outcome Measures: Visual results, retinal reattachment rate, proliferative vitreoretinopathy, macular

pucker, cystoid macular edema, choroidal detachment, intraocular pressure, extraocular muscle dysfunction, andanisometropia.

Results: There were no statistically signicant differences between the 2 treatment groups regarding thesingle-operation retinal reattachment rate at the 1-, 2-, 4-, and 6-month follow-up examinations. Patients in thebuckle group had 28% greater likelihood of anatomic success compared with those in the vitrectomy group(odds ratio, 1.28; 95% condence interval, 0.732.24), indicating no statistically signicant difference. Prolifer-ative vitreoretinopathy was the main cause of anatomic failure in both groups and occurred independent of thesurgical technique used. Best-corrected visual acuity at the 1-, 2-, 4-, and 6-month postoperative follow-upexaminations showed no statistically signicant difference between the 2 groups. Six months after surgery,12.8% of eyes in the buckle group and 11.3% of eyes in the vitrectomy group achieved visual acuity of 20/40 orbetter. The difference between the 2 groups was not statistically signicant. Corresponding gures were 66.3%and 64.5% for visual acuity of 20/200 or better in the buckle and vitrectomy groups, respectively, again with nostatistically signicant difference. There were no statistically signicant differences in rates of complications.

Conclusions: Scleral buckling and primary vitrectomy without an encircling band have comparable results inpseudophakic and aphakic retinal detachment. The choice of surgical technique depends on various factors,including patient compliance, cost of surgery, experience and capability of surgeons, and availability of appropriateinstrumentation. Ophthalmology 2005;112:14211429 2005 by the American Academy of Ophthalmology.

Approximately 30% to 40% of rhegmatogenous retinal de-tachments (RDs) occur after cataract surgery. 1,2 Scleral

buckling is a standard surgical method to manage RD aftercataract surgery. Nevertheless, the outcomes of scleral

buckling in aphakic and pseudophakic eyes generally areOriginally received: September 7, 2004.Accepted: February 11, 2005. Manuscript no. 2004-83.1 Ophthalmic Research Center, Labbanejad Medical Center, Tehran, Iran.2 Farabi Eye Hospital, Tehran, Iran.3 Rasoul Akram Hospital, Tehran, Iran.4 Feiz Hospital, Esfahan, Iran.5 Khalili Eye Hospital, Shiraz, Iran.6 Nikoukari Hospital, Tabriz, Iran.7 Amiralmomenin Hospital, Rasht, Iran.8 Emam Reza Hospital, Mashad, Iran.

Presented at: American Academy of Ophthalmology Annual Meeting,November, 2003; Anaheim, California.Supported by the National Research Center of Medical Sciences, Tehran,Iran, and the Ophthalmic Research Center of Shaheed Beheshti Universityof Medical Sciences, Tehran, Iran.None of the authors have a nancial interest in the subject matter of the article.Correspondence to Hamid Ahmadieh, MD, Ophthalmic Research Center,Labbanejad Medical Center, Pasdaran Ave. Boostan 9 St., Tehran 16666,Iran. E-mail: [email protected].*For Study Group membership, see Appendix.

1421.e1 2005 by the American Academy of Ophthalmology ISSN 0161-6420/05/$see front matterPublished by Elsevier Inc. doi:10.1016/j.ophtha.2005.02.018


2/10

less favorable than in phakic eyes. 3,4 The range of anatomicsuccess varies from 61.5% to 80% in cases of pseudophakicand aphakic RD managed by scleral buckling. 35 Undiag-nosed retinal breaks are an important cause of failure inthese cases resulting from the smaller size and anteriorlocation of retinal breaks as well as incomplete fundus viewbecause of anterior or posterior capsular brosis, corticalremnants, poor pupillary dilation, vitreous opacities, andoptical aberrations secondary to the intraocular lens (IOL)itself. 6 In addition to unseen retinal breaks, proliferativevitreoretinopathy (PVR) has been shown to occur morefrequently in pseudophakic and aphakic RD. 3,7

Recent advances in vitrectomy technique have encour-aged vitreoretinal surgeons to expand the role of primarypars plana vitrectomy in the management of uncomplicatedRDs. 814 Primary vitrectomy especially has been consid-ered as rst-line s urgic al treatment in cases of pseudophakicand aphakic RD. 1420 The rationale for such an approachincludes the ability to remove retained lens material, vitre-ous opacities, and retinal pigment epithelial cells, whileallowing controlled drainage of subretinal uid. Other po-

tential advantages of primary vitrectomy in these cases maybe the ability to visualize small retinal breaks with orwithout simultaneous scleral depression and appropriateapplication of retinopexy. The absence of risk of cataractformation is another reason for surgeons tendency to per-form primary vitrectomy in aphakic and pseudophakic eyes.

Recently, there have been retrospective and prospectivecase series showing promising results of primary vitrectomyin the management of pseudophakic and aphakic RD. 1520

Most of these studies however, suffer from selection bias,and the necessity of conducting a controll ed randomizedtrial has been emphasized by many authors. 1315,18,19,21,22

Our study was designed as a multicenter controlled clin-

ical trial to compare the anatomic and visual outcomes of scleral buckling with primary vitrectomy alone in pseu-dophakic and aphakic RD. Herein, we report the primaryoutcomes of a single operation. These are the results of a6-month follow-up study.

Patients and Methods

Clinical Examination

All patients underwent an interview for comprehensive history,including onset of visual symptoms resulting from RD, date of cataract surgery, history of previous ocular operations, other oculardisorders including glaucoma and diabetic retinopathy, oculartrauma, and history of RD in the fellow eye. Clinical examinationincluded determination of best-corrected visual acuity (BCVA);relative afferent pupillary defect in the affected eye and refractionof the fellow eye; slit-lamp examination, including assessment of the anterior segment, type, and position of IOL and integrity of theposterior capsule; and intraocular pressure (IOP) measurement.Funduscopic examination was performed to evaluate the extent of RD, the presence of any predisposing pathologic features in theperipheral retina, PVR grading, signs of myopic degeneration andto nd retinal breaks and to determine their location, type, andnumber. Findings were recorded in relevant information sheets.

Exclusion CriteriaPatients with history of RD surgery in the affected eye, oculartrauma, diabetic retinopathy, glaucoma, uveitis, macular hole, wettype age-related macular degeneration and macular geographicatrophy, giant retinal tear, PVR greater than grade B, 1-eyedpatients, patients younger than 15 years of age, patients with densevitreous hemorrhage obscuring fundus view, and those with alocalized detachment extending less than 1 quadrant with a denite

retinal break were excluded from the study.After recording data in information sheets and applying theinclusion criteria, patients were randomly allocated into one of thefollowing treatment groups: (1) scleral buckling or (2) vitrectomywithout an encircling band or any buckle. Simple random a ssign-ment was performed using SPSS software (version 10.0). 23 Allpatients were operated either by an attending vitreoretinal surgeonor by vitreoretinal fellows with at least 6 months of training.

The clinical trial was approved by the Review Board/ EthicsCommittee of the Ophthalmic Research Center.

Surgical Technique

Scleral Buckling. Conventional techniques of scleral bucklingwere used for the buckle group: after a 360 limbal peritomy andpassing traction sutures under the rectus muscles, retinal breakswere localized (if possible) and sutures were placed in the sclerasuch that the buckle indented the site of the break and 1 clock houron either side, 4 mm posterior and 2 to 3 mm anterior to it. If shmouth phenomenon was a possibility because of the type of thebreak or if circumferential scleral buckling was not feasible, ameridional sponge (no. 505 or 507) was placed in addition to anencircling band no. 240. There were cases in which no deniteretinal break could be found, either before or during the operation.In these cases, if the RD was incomplete, a circumferential buckle(silicone tire no. 276) was used in the area of RD, but if there wastotal RD, the same buckle was used for 360. In all cases, anencircling band no. 240 was also placed to produce a moderatebuckle height. Cryotherapy was used to induce a chorioretinal scar

when the retinal break had been localized and scleral sutures hadbeen placed. Cryospots were applied at the edges of the breaks andnot on their beds. Cryotherapy was not used in cases of unseenretinal breaks; in such cases, 360 laser treatment was applied onthe encircling buckle within 1 week after surgery.

For subretinal uid (SRF) drainage, a 2-mm sclerotomy wascreated approximately 1 clock hour adjacent to horizontal rectimuscles or beneath the vertical recti muscles, in loci in which thegreatest amount of SRF existed. After cauterization or diathermyof the choroidal bed at the sclerotomy site, SRF was drained andthe sclerotomy was closed with 8-0 silk suture. If severe ocularhypotony occurred after SRF drainage, saline solution was injectedintracamerally (in aphakic eyes) or intravitreally (in pseudophakiceyes). In cases of shallow RD or little SRF, the stage of subretinaluid drainage was omitted.

Primary Vitrectomy. Three-port pars plana vitrectomy wasused for patients in the vitrectomy group. Sclerotomies werecreated 3 mm from the limbus, and a standard pars plana deepvitrectomy was performed, avoiding debulking of the vitreousbase. All vitreous attachments to the edge of retinal breaks wereremoved, as were attachments to the IOL, iris, or wound. Subreti-nal uid was drained using peruorocarbon liquid. Peruorocar-bon liquid was injected on the posterior pole through a 20-gaugeblunt cannula as a single large bubble. This resulted in subretinaluid egressing from the preexisting peripheral retinal break(s).Endolaser was used to create chorioretinal adhesions. If retinalbreaks could not be found, laser was used to create 2 to 3 rows of burns posterior to the entire vitreous base. Finally, uidair ex-

Ophthalmology Volume 112, Number 8, August 2005

1421.e2


3/10

change was followed by sulfur hexauoride injection (20% non-expansile concentration), using the ush technique. If some sub-retinal uid remained at the end of the surgery, no attempt wasmade to drain it by retinotomy. Patients were strictly advised toremain in prone position for 5 days; after that, the appropriateposition was advised, depending on the location of the retinalbreak.

Intraoperative ndings related to surgical technique in bothgroups, as well as early postoperative complications including IOPrise, choroidal detachment, severe uveitis, and failure to achieveretinal reattachment, were recorded in relevant information sheets.Patients were examined 1, 2, 4, and 6 months after the operation.The following data were recorded in the information sheets: re-fractive error, BCVA, retinal reattachment, indications for a sec-ond operation, ocular deviation or impaired ocular movements,severity of PVR resulting in retinal redetachment, choroidal de-tachment, cystoid macular edema (CME) and macular puckerformation. Similar settings were provided to measure BCVA be-fore and after the operation, and the optometrists were maskedabout the status of the patients. In all cases of capsular opacity(more than 2 ) obscuring the surgeons view during the operation,a posterior capsulectomy with a diameter of 5 mm was created(2 opacity was dened as an opacity that impeded visualizationof the second branching of retinal vessels).

Statistical AnalysisData were analyzed in 2 parts, descriptive and analytic. Descrip-tive statistics were given by means ( standard deviation) forcontinuous variables and counts (%) for categorical variables.Further analysis was carried out rst by univariate analysis,namely t test (or MannWhitney U test when necessary) forcomparing means between groups and the chi-square test (orFisher exact test) for comparing categorical variables betweengroups.

At the second step, means or counts of outcome variables werecompared between groups adjusting for confounders. This wascarried out by analysis of covariance and the MantelHaenzel test.A 2-way repeated measure analysis of variance was used to com-pare the changes of BCVA between the 2 groups over timeintervals.

Finally, multiple regression and multiple logistic regressionanalysis were used for taking into account the effects of covariateson the outcome variables. Statistical level of signicance waspreset at 0.05. Data were analyzed using SPSS software version10.0. 23

Sample Size CalculationThe minimum sample size required to detect a 20% increase inretinal reattachment rate with primary vitrectomy compared withscleral buckling at 95% condence level with the power of 0.8 was

obtained using the following formula:

n(u v)2 p(1 p) p0(1 p0)

2

( p p0)2

P Retinal reattachment rate in vitrectomy groupP0 Retinal reattachment rate in scleral buckling group (as-sumed to be 70%)u 0.95v 0.80

A total sample size of 150 eyes (at least 75 eyes in each treatmentgroup) was calculated.

Results

Two hundred twenty-ve eyes of 225 patients (142 men) wereincluded in this study. The mean age was 62.5 11.2 years. Of 225eyes, 144 (64%) were pseudophakic and the remaining were apha-kic. Retinal detachment had a mean extension of 3.08 quadrantswith involvement of the macula in all except 5 eyes. Beforesurgery, PVR grade B was present in all eyes. Of 225 eyes, 190(84.4%) had visual acuity of light perception or hand movementsbefore surgery. No retinal breaks were detected in 67 eyes (29.8%)before surgery. One retinal break was seen in 127 eyes (56.4%). In26 eyes (11.6%), there were 2 retinal breaks, and in 5 eyes (2.2%),3 retinal breaks were found. Overall, of the retinal breaks beingnoted before surgery, 152 (78.4%) were equatorial, 36 (18.6%)were located anterior to the equator, and 6 (3%) were in both loci.Regarding type, 77.4% of the retinal breaks were ap tears, and theremaining were round holes.

Baseline CharacteristicsOf 225 eligible eyes, 126 and 99 eyes underwent scleral bucklingand vitrectomy, respectively. The 2 groups were matched forbaseline characteristics except the following variables: patientsage, preoperative visual acuity, family history of RD, RD exten-sion, vitreous incarceration into cataract wou nd or adhesion to iris,and history of secon dary IOL implantation ( Table 1 [available athttp://aaojournal.org ]).

Signs of myopic degeneration were present in 17.5% and28.3% of the buckle and vitrectomy gro ups, respectively, whichwas of borderline statistical signicance ( Table 1 ).

Anatomic ResultsThe single-operation anatomic success rate at the 1-, 2-, 4-, and6-month follow-up examinations was 82.5%, 73.8%, 72.2%, and68.2% in the buckle group and was 75.7%, 67.7%, 62.6%, and62.6% in the vitrectomy group, respectively. The difference be-tween the anatomic results in the 2 groups was not statistically

signicant by the chi-square test ( Table 2 ). After further opera-tions, the nal reattachment rate was 85% in the buckle group and92% in the vitrectomy group.

In the buckle group, 40 of 126 eyes (31.8%) and 37 of 99 eyes(37.4%) in the vitrectomy group developed retinal redetachmentwithin 6 months after surgery. The vast majority of redetachmentsoccurred within 2 months (82.5% and 86.5% in the buckle andvitrectomy groups, respectively; Fig 1). The most common causeof retinal redetachment was PVR grade C (buckle group, 92.5%;vitrectomy group, 94.6%). Proliferative vitreoretinopathy grade Cwas of both anterior and posterior types. Other causes of anatomicfailure within 6 months included missed and new retinal breaks:5% and 2.5% in the buckle group and 2.7% and 2.7% in thevitrectomy group, respectively ( Fig 2).

The observed difference in retinal reattachment rate betweenthe 2 groups at 6 months was 5.6% (68.2% in the buckle group vs.62.6% in the vitrectomy group; Table 2 ), which was not statisti-cally signicant. Although patients in the buckle group had a 28%more likelihood of anatomic success compared with those in thevitrectomy group (odds ratio, 1.28; 95% condence interval, 0.732.24) at 6 months, there was no statistically signicant differencebetween the 2 groups.

Six-month postoperative anatomic success rates of myopic eyesin buckle and vitrectomy groups were 70.6% and 55%, respec-tively, with no statistically signicant difference. The odds of anatomic success in the buckle group compared with the vitrec-tomy group adjusted for myopic degeneration was approximately1.27 (95% condence interval, 0.57.69), conrming that stratify-

Ahmadieh et al Scleral Buckling versus Primary Vitrectomy

1421.e3
http://www.aaojournal.org/http://www.aaojournal.org/http://www.aaojournal.org/http://www.aaojournal.org/


4/10

ing by myopic degeneration did not appreciably change the oddsratio. Nevertheless, in comparing the cases of retinal reattachmentwith cases of redetached retina after 6 months, myopic degenera-tion was found to have a statistically signicant effect on theanatomic outcome in both groups ( P 0.04). The anatomicsuccess rate in eyes with vitreous incarceration at the 6-month

follow-up examination in the buckle and vitrectomy groups were57.1% and 66.7%, respectively, which did not show any statisti-cally signicant difference ( Fig 3).

In a logistic regression model, none of the preoperative vari-ables (patient age, preoperative visual acuity, family history of RD,RD extension, vitreous incarceration into cataract wound, historyof secondary IOL implantation) nor type of RD surgery (scleralbuckling vs. primary vitrectomy) had signicant effect on ana-tomic success.

Visual ResultsOf 86 eyes with reattached retinas after 6 months, 11 eyes (12.8%)gained visual acuity of 20/40 or better in the buckle group. In thevitrectomy group, 7 of 62 eyes with reattached retinas (11.3%)gained this level of visual acuity after 6 months ( P 0.78). Of 86eyes, 57 eyes (66.3%) and 40 of 62 eyes (64.5%) achieved visualacuity of 20/200 or better in the buckle and vitrectomy groups,respectively. Again, there was no statistically signicant differencebetween the 2 groups ( P 0.82). The distribution of visual acuity6 months after surgery in each treatment group according to 3

visual categories (20/40 or better, 20/200 to 20/50, less than20/200) is shown in Figure 4 . The difference between the 2 groupswas not statistically signicant ( P 0.95).

The BCVA (logarithm of the minimum angle of resolution[logMAR]) at the 1-, 2-, 4-, and 6-month postoperative follow-upexaminations were 1.25 0.67, 1.08 0.65, 0.98 0.65, and

0.96 0.68 in the buckle group and 1.24 0.68, 1.16 0.70,1.01 0.63, and 0.96 0.62 in the vitrectomy group, respectively.There was no statistically signicant difference between the 2groups (Table 2 ). Based on a general linear model (2-way repeatedmeasure), time had a signicant effect on improving visual acuity.Visual acuity signicantly improved after surgery compared withpreoperative visual acuity in both groups ( P 0.0001). This im-provement continued until 2 months after surgery in both groups(P 0.007), but there was no signicant improvement after 4months ( Figs 5, 6).

Multiple regression analysis showed that preoperative visualacuity was the only variable with a signicant effect on postoper-ative visual acuity. The nal model was: 0.345 0.277 logMARof preoperative BCVA, indicating that being in either surgicaltreatment group had no effect on postoperative visual acuity.

ComplicationsThe rates of macular pucker at the 1-, 2-, 4-, and 6-monthpostoperative examinations were 15%, 15.9%, 22.2%, and22.2% in the buckle group, respectively, and were 12.1%,

0

5

10

15

20

25

30

1st month 2nd month 4th month 6th month

Buckle

Vitrectomy

%

R e t

i n a l r e

d e t a c

h m e n

t r a

t e

Figure 1. Rate of retinal redetachment at follow-up examinations in each treatment group.

Table 2. Comparison of Surgical Results in the 2 Treatment Groups

Follow-upPeriod (mos) Buckle Group Vitrectomy Group P Value

logMAR (BCVA; mean [SD]) 1 1.25 (0.67) 1.24 (0.68) 0.922 1.08 (.65) 1.16 (0.70) 0.474 0.98 (.65) 1.01 (0.63) 0.846 0.96 (.68) 0.96 (.62) 1

Retinal reattachment (no. [%]) 1 104 (82.5) 75 (75.7) 0.26

2 93 (73.8) 67 (67.7) 0.774 91 (72.2) 62 (62.6) 0.326 86 (68.2) 62 (62.6) 0.24

BCVA best-corrected visual acuity; logMAR logarithm of the minimum angle of resolution; SD standarddeviation.


1421.e4


5/10

13.1%, 14.1%, and 22.2%, respectively, in the vitrectomygroup. The rates of clinical CME were 11.1%, 9.5%, 8.7%, and6.3% in the buckle group, respectively, and were 16.2%, 13.1%,

9.1%, and 9.1%, respectively, in the vitrectomy group. Therates of postoperative IOP more than 20 mmHg with medica-tions were 11.9%, 8.7%, 7.1%, and 6.3% in the buckle group,respectively, and were 10.1%, 9.1%, 7.1%, and 6.1%, respec-tively, in the vitrectomy group. The rates of early postoperativeIOP rise (IOP 20 mmHg) within 1 week after RD surgery inthe buckle and vitrectomy groups were 24.6% and 26.3%,respectively. The rates of choroidal detachment were 6.3%,3.2%, 1.6%, and 0%, respectively, in the buckle group and 2.0%at month 1 and 0% at month 2 and thereafter, in the vitrectomygroup. The rates of extraocular muscle dysfunction in thebuckle and vitrectomy groups at 6 months were 4.0% and 0%,respectively. None of these complications showed any statisti-cally signicant difference between the 2 groups ( Table 3 ).

Discussion

We conducted a randomized clinical trial to compare theanatomic and visual outcomes of the 2 techniques (conven-tional scleral buckling vs. primary vitrectomy withoutscleral buckle) in the management of pseudophakic and

aphakic RD. The 2 treatment groups were matched for mostof the preoperative characteristics (sex, interval of cataractsurgery and RD, history of RD in the fellow eye, history of neodymium:yttriumaluminumgarnet laser capsulotomy,macular status, IOL position, type of cataract surgery, IOP,vitreous hemorrhage, relative afferent pupillary defect, ex-traocular muscle dysfunction, and interval of RD symptomsto RD surgery). There were, however, statistically signi-cant differences in some of the preoperative characteristicsin the 2 treatment groups. These variables included age,BCVA (logMAR), family history of RD, RD extension,vitreous incarceration into the cataract wound, and rate of secondary IOL implantation. There was a borderline differ-ence between the 2 groups related to the prevalence of myopic degeneration. The 2 groups were adjusted for thesevariables.

There are no studies demonstrating a consistent relationbetween patients age and anatomic success after RD sur-gery. Most patients in both groups were cases with a historyof age-related (senile) cataract surgery. The statisticallysignicant difference between the mean age of patients inthe 2 treatment groups did not seem to be of clinical sig-nicance. In addition, after adjusting for preoperative pa-

2.5 5

92.5 94.6

2.72.7

01020304050

60708090

100

New break Missed Break PVRC

Buckle

Vitrectomy

C a u s e s o f r e

t i n a l r e

d e t a c

h m e n

t

%

Figure 2. Causes of retinal redetachment in each treatment group. PVRC proliferative vitreoretinopathy grade C.

010203040506070

8090

with vitreous incarceration without vitreousincarceration

Buckle

Vitrectomy

A n a t o m

i c s u c c e s s r a

t e

%

Figure 3. Comparison of anatomic success in eyes with and without vitreous incarceration in each treatment group.


1421.e5


6/10

tient characteristics, age showed no signicant effect on theoutcomes of vitrectomy versus scleral buckling.

Review of previous reports showed no association be-tween a positive family history of RD and a lower rate of anatomic success. There was no case of hereditary vitreo-retinal degeneration among our patients.

The relationship between myopic RD and surgical suc-cess has been in dispute. Burton and Lambert 24 reported asomewhat lower success rate in eyes with more than 4 D of myopia. Eyes with high myopia may have a slightly lowersuccess rate when managed by scleral buckling. This in partresults from an increased risk of complications duringplacement of scleral sutures because of marked scleral thin-ning and an increased risk of choroidal hemorrhage duringdrainage of subretinal uid. However, high myopia has beenmentioned as a risk factor f or suprachoroidal hemorrhageduring pars plana vitrectomy. 25 In our study, comparison of 6-month postoperative anatomic success rates of myopiceyes in the buckle and vitrectomy groups did not show asignicant difference. However, when we compared casesof retinal redetachment with successful cases after 6months, high myopia was found to have a predictive role foranatomic failure in both groups.

There are no reports on the correlation between vitreousincarceration into the cataract wound and anatomic successrates of RD surgery. However, there are some clinicalfeatures in these eyes that may make the outcomes worse.

Chronic inammation and vitreous traction in these eyescan increase the risk of CME and PVR. Vitrectomy alreadyhas been shown to have a therapeutic role for CME in theseinstances. 26 In our study, anatomic success rate in these eyeswas slightly higher in the vitrectomy group, but the differ-ence between the 2 treatment groups was not statisticallysignicant.

After adjusting for preoperative patients characteristics,there was no statistically signicant difference between the2 treatment groups regarding the single-operation retinalreattachment rates at the 1-, 2-, 4-, and 6-month follow-upexaminations. The retinal reattachment rates were 82.5%and 75.7% in the buckle and vitrectomy groups, respec-tively, 1 month after surgery. Nevertheless, the retinal reat-tachment rates declined to 68.2% in the scleral bucklinggroup and 62.6% in the vitrectomy group after 6 months.Most cases of retinal redetachment occurred within 1 to 2months after the surgical procedure in both groups.

Missed retinal breaks have been reported to be a signif-icant cause of anatomic failure in cases of pseudophakic andaphakic RD. In our cases, the rate of unseen retinal breakswas higher in the buckle group. Nevertheless, PVR was themain cause of anatomic failure in both groups. Our ndingsare comparable with previous reports in the literature andconrm the high incidence of PVR in cases of aphakic andpseudophakic RD. 3,7 The results of this study showed thatprimary vitrectomy, in comparison with scleral bucklin g,did not reduce the rate of PVR. In a study by Cowley et al 27

to identify the clinical risk factors for PVR, the use of vitrectomy was revealed to be the strongest predictor. Manyauthors, however, hypothesized that removal of the vitreoustraction bands and RPE cells may decrease the rate of postoperative PVR. 1420 This has been one of the mainreasons for suggesting primary vitrectomy as a substitute for

scleral buckling in uncomplicated cases, especially for RDsafter cataract surgery. The ndings of this study, however,were not in line with this hypothesis. This study showed thatthe occurrence of PVR as the main cause of retinal rede-tachment was independent of the surgical technique used(primary vitrectomy vs. scleral buckling).

Isernhagen and Wilkinson 28 evaluated the effect of pre-operative factors on nal visual acuity after repair of pseu-dophakic RD with scleral buckling. Preoperative visual

0

10

20

30

40

50

60

20/40 orbetter

20/50 -20/200

Less than20/200

BuckleVitrectomy

P e r c e n

t

%

Figure 4. Distribution of visual acuity 6 months after surgery in eachtreatment group according to 3 visual categories.

0

0.5

1

1.5

2

2.5

preop VA postop VA

Buckle

vitrectomy

B e s

t c o r r e c

t e d v i s u a l a c u

i t y

Figure 5. Preoperative and 6 month postoperative best-corrected visual acuity (VA) in each treatment group according to logarithm of the minimumangle of resolution.


1421.e6


7/10

acuity was the most important variable. In our study, pre-operative visual acuity was also the single factor having the

highest correlation with visual outcome. There was no sta-tistically signicant difference between the 2 treatmentgroups regarding visual results. In other words, being ineither surgical technique group did not have a signicanteffect on the visual outcome. This shows that preoperativevisual acuity is the most inuential factor on nal vision notonly in scleral buckling, but also with newer methods of RDsurgery. Improvement in postoperative visual acuity wasobserved during the follow-up period, especially during therst 4 months after surgery.

Macular involvement is an important factor inuencing

vision after RD surgery. Duration of macular involvement isone of the most important variables associated with return

of central vision.28

There is a signicant decrease in visualrecovery with macula-off detachments lasting longer than 1week. 29 The duration of RD with macular involvement inour series was longer than in other reports. Higher patientage is associated with a trend toward reduced postoperativevisual acuity. 24,2830 Most of our patients were older, witha median age of 64 years. The extent of RD also has beenshown to have an inverse relationship with anatomic andvisual outcomes. 24,28,3032 Most RDs in both treatmentgroups in our study were total or nearly total. Only 2.4% of eyes in the scleral buckling group and 2.1% of eyes in the

0

0.5

1

1.5

2

2.5

Beforeoperation

1st month 2nd month 4th month 6th month

Buckle

Vitrectomy

B e s

t c o r r e c t e

d v i s u a l a c u

i t y

Figure 6. Improvement of best-corrected visual acuity (logarithm of the minimum angle of resolution) after retinal detachment surgery over time in eachtreatment group.

Table 3. Surgical Complications in the 2 Treatment Groups

Follow-upPeriod (mos) Buckle Group Vitrectomy Group * P Value

EOM dysfunction (no. [%]) 1 4 (3.2) 0 (0.0) 0.122 5 (4.0) 1 (1.0) 0.224 5 (4.0) 0 (0.0) 0.076 5 (4.0) 0 (0.0) 0.07

Choroidal detachment (no. [%]) 1 8 (6.3) 2 (2.0) 0.102 4 (3.2) 0 (0.0) 0.134 2 (1.6) 0 (0.0) 0.506 0 (0.0) 0 (0.0)

Clinical CME (no. [%]) 1 14 (11.1) 16 (16.2) 0.412 12 (9.5) 13 (13.1) 0.514 11 (8.7) 9 (9.1) 1.06 8 (6.3) 9 (9.1) 0.43

Macular pucker (no. [%]) 1 19 (15.0) 12 (12.1) 0.402 20 (15.9) 13 (13.1) 0.694 28 (22.2) 14 (14.1) 0.136 28 (22.2) 22 (22.2) 1.0

Postoperative IOP 20 mmHg with medications(no. [%])

1 15 (11.9) 10 (10.1) 0.66

2 11 (8.7) 9 (9.1) 1.04 9 (7.1) 7 (7.1) 1.06 8 (6.3) 6 (6.1) 1.0

Anisometropia at sixth month (mean [SD]) Sphere 1.92 (2.18) 1.78 (2.2) 0.76Cylinder 0.44 (0.71) 0.26 (0.49) 0.20

CME cystoid macular edema; EOM extraocular mucle; IOP intraocular pressure; SD standard deviation.*Fisher exact probability.


1421.e7


8/10

vitrectomy group were macula-on. Proliferative vitreoreti-nopathy grade B already ha s been identied to be a risk factor for anatomic failure. 33 All patients in our seriesshowed the signs of PVR grade B before surgery. Trainingvitreoretinal fellows in educational centers performed someof the operations. This also may have had a negative effecton the results.

Macular pucker is a common nding after RD surgery. It

has been reported in 2% to 17% of detachment cases un-dergoing scleral buckling. 34 Extensive RD and involvementof the macula have been n oted to be associated with devel-opment of macular pucker. 34 Macular pucker also has beenreported as a frequent complication of primary vitrectomyfor pseudophakic and aphakic RD. 19 We observed a highrate of this complication in our study, and there was nostatistically signicant difference between the 2 treatmentgroups. The incidence of this complication increased grad-ually during the 6-month follow-up period.

Choroidal detachment has been reported as one of themost common complications of s cleral buckling for repairof pseudophakic and aphakic RD. 3,5 In the series reported

by Yoshida et al,5

it was the most common postoperativecomplication and was observed in 21.5% of the cases. In ourseries, this complication was more frequent in the scleralbuckling group compared with the vitrectomy group, but thedifference was not statistically signicant.

Cystoid macular edema is another frequent complicationafter retinal surgery. 34 It is more prevalent in aphakic thanin phakic eyes. We evaluated the incidence of clinical CMEin our patients. There was no signicant difference betweenthe 2 treatment groups regarding this complication.

Refractive error changes after R D surgery can result inanisometropia. In a study by Rubin, 35 changes in refractiveerror were related to the height of the scleral buckle andwere greater in phakic than in aphakic eyes. Among thepresumed advantages of primary vitrectomy without anencircling element over scle ral buckling is the possibility of less induced anisometropia. 19,20 In our study, postoperativeanisometropia was analyzed considering spherical and cy-lindrical refractive errors separately, which revealed nostatistically signicant difference between the 2 groups.

In most studies, early transient IOP rise was the mostcommon complication of primary vitrectomy combinedwith uidgas exchang e. In the studies by Bartz-Schmidt etal15 and Speicher et al, 20 this rate was reported to be 48%and 17.9%, respectively. In our study, early postoperativeIOP rises ( 20 mmHg with medication) in the buckle andvitrectomy groups were 24.6% and 26.3%, respectively,which did not show any signicant difference. Rates of IOPof more than 20 mmHg with medications again had nostatistically signicant difference between the 2 groups 6months after surgery.

Persistent extraocular muscle dysfunction after bucklesurgery has been reported in up to 4%, a nd most casesoccurred in eyes with an encircling element. 28 The rate of this complication after vitrectomy was not reported in otherstudies. In the current study, extraocular muscle dysfunctiondeveloped in 4% of cases in the buckle group and in nocases in the vitrectomy group 6 months after surgery.

At the end of the previous decade, with the invention of

a variety of surgical techniques as substitutes for scleralbuckling to repair uncomplicated RD, Wilkinson 21 wrote aneditorial entitled Wanted: Optimal Data Regarding Surgeryfor Retinal Detachment. He emphasized the importance of discovering the optimal method of repairing RDs and men-tioned that a large, prospective, randomized trial would berequired to answer this question. He assumed, however, thatno single technique would be considered optimal for all

RDs. Our study provides adequate information in this re-gard.

We found that primary vitrectomy without scleral bucklewas not more effective than conventional scleral buckling inthe management of uncomplicated RD in eyes with a historyof cataract extraction. Based on our results, we believe thatat present, no single surgical technique can be considered asoptimal and routine for all cases of pseudophakic and apha-kic RD. It is noteworthy to consider some of the followingfactors in selecting each of these options in this group of patients: the costs of these operations, experience and ca-pabilities of vitreoretinal surgeons, and the availability of appropriate instrumentation. The possible effect of an en-

circling band on the rate of retinal reattachment in cases of primary vitrectomy remains a question that needs to beanswered in another controlled study.

Acknowledgments. The PARD study group thanks ShahinYazdani, MD, Arash Anissian, MD, Zahra Rabbanikhah, MD,Nasrin Rafati, MD, Mr Ali A. Aghdaee, Ms Maryam Ghazaee, MsLeila Azadvari, and Mr Mohammad R. Farrokhian for their assis-tance during the course of the study.

References

1. Haimann MH, Burton TC, Brown CK. Epidemiology of reti-nal detachment. Arch Ophthalmol 1982;100:28992.

2. Ah-Fat FG, Sharma MC, Majid MA, et al. Trends in vitreo-retinal surgery at a tertiary referral centre: 1987 to 1996. Br JOphthalmol 1999;83:3968.

3. Ho PC, Tolentino FI. Pseudophakic retinal detachment. Sur-gical success rate with various types of IOLs. Ophthalmology1984;91:84752.

4. Cousins S, Boniuk I, Okun E, et al. Pseudophakic retinaldetachments in the presence of various IOL types. Ophthal-mology 1986;93:1198208.

5. Yoshida A, Ogasawara H, Jalkh AE, et al. Retinal detachmentafter cataract surgery. Surgical results. Ophthalmology 1992;99:4605.

6. Yoshida A, Ogasawara H, Jalkh AE, et al. Retinal detachmentafter cataract surgery. Predisposing factors. Ophthalmology1992;99:4539.

7. Wilkinson CP. Pseudophakic retinal detachments. Retina1985;5:14.

8. Escoffery RF, Olk RJ, Grand MG, Boniuk I. Vitrectomywithout scleral buckling for primary rhegmatogenous retinaldetachment. Am J Ophthalmol 1985;99:27581.

9. Gartry DS, Chignell AH, Franks WA, Wong D. Pars planavitrectomy for the treatment of rhegmatogenous retinal de-tachment uncomplicated by advanced proliferative vitreoreti-nopathy. Br J Ophthalmol 1993;77:199203.

10. Hakin KN, Lavin MJ, Leaver PK. Primary vitrectomy forrhegmatogenous retinal detachment. Graefes Arch Clin ExpOphthalmol 1993;231:3446.

11. Heimann H, Bornfeld N, Friedrichs W, et al. Primary vitrec-


1421.e8


9/10

tomy without scleral buckling for rhegmatogenous retinaldetachment. Graefes Arch Clin Exp Ophthalmol 1996;234:5618.

12. Oshima Y, Emi K, Motokura M, Yamanishi S. Survey of surgical indications and results of primary pars plana vitrec-tomy for rhegmatogenous retinal detachments. Jpn J Ophthal-mol 1999;43:1206.

13. Oshima Y, Yamanishi S, Sawa M, et al. Two-year follow-upstudy comparing primary vitrectomy with scleral buckling formacula-off rhegmatogenous retinal detachment. Jpn J Oph-thalmol 2000;44:53849.

14. Brazitikos PD. The expanding role of primary pars planavitrectomy in the treatment of rhegmatogenous noncompli-cated retinal detachment. Semin Ophthlmol 2000;15:6577.

15. Bartz-Schmidt KU, Kirchhof B, Heimann K. Primary vitrec-tomy for pseudophakic retinal detachment. Br J Ophthalmol1996;80:3469.

16. Desai UR, Strassman IB. Combined pars plana vitrectomy andscleral buckling for pseudophakic and aphakic retinal detach-ments in which a break is not seen preoperatively. OphthalmicSurg Lasers 1997;28:71822.

17. Devenyi RG, de Carvalho Nakamura H. Combined scleralbuckle and pars plana vitrectomy as a primary procedure for

pseudophakic retinal detachments. Ophthalmic Surg Lasers1999;30:6158.18. Brazitikos PD, DAmico DJ, Tsinopoulos IT, Stangos NT.

Primary vitrectomy with peruoro-n-octane use in the treat-ment of pseudophakic retinal detachment with undetectedretinal breaks. Retina 1999;19:1039.

19. Campo RV, Sipperley JO, Sneed SR, et al. Pars plana vitrec-tomy without scleral buckle for pseudophakic retinal detach-ments. Ophthalmology 1999;106:18115, discussion 1816.

20. Speicher MA, Fu AD, Martin JP, von Fricken MA. Primaryvitrectomy alone for repair of retinal detachments followingcataract surgery. Retina 2000;20:45964.

21. Wilkinson CP. Wanted: optimal data regarding surgery forretinal detachment. Retina 1998;18:199201.

22. Heimann H, Hellmich M, Bornfeld N, et al. Scleral bucklingversus primary vitrectomy in rhegmatogenous retinal detach-ment (SPR Study): design issues and implications. SPR Studyreport no. 1. Graefes Arch Clin Exp Ophthalmol 2001;239:56774.

23. SPSS [computer program]. Version 10.0. Chicago: SPSS Inc.;1999.

24. Burton TC, Lambert RW Jr. A predictive model for visualacuity recovery following retinal detachment surgery. Oph-thalmology 1978;85:61925.

25. Tabandeh H, Flynn HW Jr. Suprachoroidal hemorrhage duringpars plana vitrectomy. Curr Opin Ophthalmol 2001;12:17985.

26. Harbour JW, Smiddy WE, Rubsamen PE, et al. Pars planavitrectomy for chronic pseudophakic cystoid macular edema.Am J Ophthalmol 1995;120:3027.

27. Cowley M, Conway BP, Campochiaro PA, et al. Clinical risk factors for proliferative vitreoretinopathy. Arch Ophthalmol1989;107:114751.

28. Isernhagen RD, Wilkinson CP. Visual acuity after the repair of pseudophakic retinal detachments involving the macula. Ret-ina 1989;9:1521.

29. Ross WH, Kozy DW. Visual recovery in macula-off rhegmat-ogenous retinal detachments. Ophthalmology 1998;105:214953.

30. Ahmadieh H, Entezari M, Soheilian M, et al. Factors inu-encing anatomic and visual results in primary scleral buckling.Eur J Ophthalmol 2000;10:1539.

31. Tani P, Robertson DM, Langworthy A. Prognosis for central

vision and anatomic reattachment in rhegmatogenous retinaldetachment with macula detached. Am J Ophthalmol 1981;92:61120.

32. Greven CM, Sanders RJ, Brown GC, et al. Pseudophakicretinal detachments. Anatomic and visual results. Ophthalmol-ogy 1992;99:25762.

33. Bonnet M. The development of severe proliferative vitreoreti-nopathy after retinal detachment surgery. Grade B: a deter-mining risk factor. Graefes Arch Clin Exp Ophthalmol 1988;226:2015.

34. Wilkinson CP, Rice TA. Michels Retinal Detachment. 2nded. St. Louis: Mosby; 1997:103942.

35. Rubin ML. The induction of refractive errors by retinal de-tachment surgery. Trans Am Ophthalmol Soc 1975;73:45290.

Appendix: The Pseudophakic and AphakicRetinal Detachment (PARD) Study Group

Clinical Sites That Participated in This Study

Farabi Eye Hospital, Tehran. Principal investigator: Hoo-shang Faghihi, MD.

Coinvestigators: Ali Tabatabaei, MD, Ali R. Lasheie,MD, Reza Karkhaneh, MD, Mohammad Riazi, MD, AhmadMirshahi , MD, Zahra Alami Harandi, MD, Mohammad R.Mansouri, MD, Mehrdad Mehrazma, MD, Hamid Ghazi,MD, Fedra Hajizadeh, MD.

Labbanejad Medical Center, Tehran. Principal inves-tigator: Siamak Moradian, MD.

Coinvestigators: Hamid Ahmadieh, MD, Mohsen Az-armina, MD, Masoud Soheilian, MD, Mohammad H. Deh-ghan, MD, Mohsen Shahsavari, MD.

Rasool Akram Hospital, Tehran. Principal investiga-tor: Mohammad M. Parvaresh, MD.

Coinvestigators: Mehdi Modarreszadeh, MD, MasihHashemi, MD, Aminollah Nikeghbali, MD, Mohammad M.Noorani, MD.

Khalili Eye Hospital, Shiraz. Principal investigator:Morsal Mehryar, MD.

Coinvestigators: Morteza Mehdizadeh, MD, MansourRahimi, MD.

Feiz Hospital, Esfahan. Principal investigator: Heshma-tollah Ghanbari, MD.

Nikoukari Hospital, Tabriz. Principal investigator:Ebadollah Heidari, MD.

Amiralmomenin Hospital, Rasht. Principal investiga-tor: Hasan Behboudi, MD.

Emam Reza Hospital, Mashad. Principal investigator:Touka Banaei, MD.Emam Hosein Hospital, Tehran. Principal investiga-

tor: Morteza Entezari, MD.

Coordinating CenterOphthalmic Research Center, Shahid Beheshti University of Medical Sciences: Siamak Moradian, MD, Hamid Ah-madieh, MD, Arash Anissian, MD, Zahra Rabbanikhah,MD, Nasrin Rafati, MD, Farshid Siadat, MD, Hamid Soori,PhD, Banafsheh Golestan, PhD.


1421.e9


10/10

Table 1. Basic Characteristics of Patients in the 2 Treatment Groups at the Beginning of Study

Buckle Group Vitrectomy Group P Value

Sex (no [%])Male 76 (60.3) 66 (66.7) 0.33Female 50 (39.7) 33 (33.3)

Age (mean [ SD]) 64.23 (11.34) 60.63 (13.65) 0.03LogMAR (BCVA mean [ SD]) 2.21 (0.67) 2.37 (0.46) 0.04Interval of cataract surgery with RD (mos; mean [ SD]) 53 (66.8) 45 (64.2) 0.39

Family history of RD (no. [%]) 0 (.0) 4 (4.0) 0.02Myopic degeneration (no. [%]) 22 (17.5) 28 (28.3) 0.07History of RD in sound eye (no. [%]) 11 (8.7) 3 (3.0) 0.09RD extension (mean quadrant [SD]) 2.95 (0.92) 3.27 (3.02) 0.01Vitreous in wound (no [%]) 14 (11.1) 24 (24.2) 0.01Secondary IOL (no. [%]) 0 (0.0) 9 (9.1) 0.00History of YAG capsulotomy (no. [%]) 24 (19.0) 18 (18.2) 0.98Macula on (no. [%]) 3 (2.4) 2 (2.0) 0.86IOL status (no. [%])

Aphakia 47 (37.3) 34 (34.4)AC IOL 16 (12.7) 20 (20.2) 0.31PC IOL 63 (50.0) 45 (45.4)

Type of cataract surgery (no. [%])ICCE with vitreous loss AC IOL 23 (18.3) 14 (14.2)ICCE without vitreous loss AC IOL 7 (5.6) 6 (6.1)ECCE PCIOL with vitreous loss 10 (7.9) 8 (8.1)

ECCE PCIOL without vitreous loss 46 (36.5) 41 (41.4) 0.33ECCE with vitreous loss AC IOL 21 (16.7) 24 (24.2)ECCE without vitreous loss AC IOL 10 (7.9) 5 (5.0)PE PCIOL with vitreous loss 1 (0.8) 0 (.0)PE PCIOL without vitreous loss 8 (6.3) 1 (1.0)Vitreous loss 54 (42.8) 45 (45.4) 0.39IOP (mean [ SD]) 9.72 (4.0) 10.58 (3.73) 0.11Choroidal detachment (no. [%]) 3 (2.4) 2 (2.1) 0.86Vitreous hemorrhage (no. [%]) 9 (7.2) 9 (9.1) 0.62RAPD (mean [ SD]) 1.35 (0.9) 1.37 (0.93) 0.84Predisposing lesions of RD (no. [%]) 37 (30.1) 24 (24.5) 0.36Interval of RD Symptoms with RD surgery (days; mean

[ SD])20.7 (16.5) 22.3 (16.1) 0.56

AC IOL anterior chamber intraocular lens; BCVA best-corrected visual acuity; ECCE extracapsular cataract extraction; ICCE intracapsularcataract extraction; IOL intraocular lens; IOP intraocular pressure; LogMAR logarithm of minimum angle of resolution; PCIOL posteriorchamber intraocular lens; PE phacoemulsication; RAPD relative afferent pupillary defect; RD retinal detachment.


1421.e10

Anatomic and Visual Outcomes of SB VS Primary PPV in Pseudophakic and Aphakic RD

Documents

Transcript of Anatomic and Visual Outcomes of SB VS Primary PPV in Pseudophakic and Aphakic RD