Primary Retinectomy in Proliferative Vitreoretinopathy


To describe the functional and anatomic results of retinectomy without scleral buckling for anterior proliferative vitreoretinopathy in eyes that did not undergo previous buckling surgery.


Retrospective, nonrandomized, interventional case series.


We reviewed the results of 123 consecutive cases of retinectomy for rhegmatogenous retinal detachment complicated by anterior proliferative vitreoretinopathy. The primary outcome measure was anatomic success, defined as complete retinal reattachment. Secondary outcome measures were final anatomic success, final visual acuity outcome, number of operations required for retinal reattachment, baseline extent of proliferative vitreoretinopathy, ocular pressure at final follow-up, extent of retinectomy, and occurrence of complications.


The anatomic success rate was 77.2%. Final attachment rate was 95.9%, reached after 1 rhegmatogenous retinal detachment reoperation in 21 cases and after 2 rhegmatogenous retinal detachment reoperations in 4 cases. We had a low rate of postoperative hypotony (4.1%). Visual acuity was significantly improved from 2.10 to 1.44 logarithm of the minimal angle of resolution units ( P < .001). Improvement was related significantly to retinectomy extent and primary anatomic success.


Our results show that primary retinectomy without scleral buckling has good anatomic and functional results.

Proliferative vitreoretinopathy (PVR) is a serious complication of rhegmatogenous retinal detachment (RRD). The development of PVR is a cell-mediated event. Retinal pigment epithelium cells, liberated into the vitreous cavity through retinal breaks, proliferate along the glial cells and form epiretinal membranes. These membranes contract and shorten the tangential, effective surface of the retina. Detachment of the retina also can cause contraction of intrinsic neuronal elements located within the retina. With detachment, ischemia and atrophy of intrinsic retinal elements also occurs. The combined result of these processes is foreshortening of the retina, making it physically impossible for the retina to line the interior of the globe. It is particularly prominent in anterior PVR. Studies have shown that eyes with anterior PVR have a much worse prognosis than those with posterior PVR.

To a certain degree, scleral buckling and membrane peeling can release tangential traction, but in the management of advanced anterior PVR, reattachment cannot be achieved by these means alone. A retinectomy often is required. Early published series on retinectomy surgery resulted in reasonable rates of anatomic attachment (35% to 80%), but low rates of visual recovery (6.7% to 55%). Postoperative complications were common, including reproliferation leading to recurrent rhegmatogenous retinal detachment and hypotony. Some reports documented a direct association between the extent of the retinectomy and a worse outcome. After such publications, a tradition developed through which retinectomies were reserved only for complex retinal detachments with PVR when other procedures, including membrane peeling and scleral buckling, were ineffective or inadequate.

More recent studies reported better result with retinectomy, in which 68% to 93% final anatomic success was achieved in addition to better visual results. In one study, final reattachment had a significant association with fewer previous operations, suggesting that earlier intervention may improve surgical outcome.

There is a general trend toward primary vitrectomy in the management of RRD. To date, no data exist on results of nonbuckle repair of PVR. The aim of our present study was to report on the outcome of retinectomy surgery without additional external buckling.


We reviewed the clinical charts of all consecutive patients that met the following inclusion criteria: an operation was performed between August 1, 2006, and August 1, 2008, for a RRD with any anterior PVR of grade C or more and no prior scleral buckling operation was performed. Eyes with a history of penetrating trauma were excluded, as were eyes with vasoproliferative disease.

All operations were performed with the Alcon Accurus (Accurus 600 DS; Alcon Laboratories, Fort Worth, Texas, USA) and the A BIOM wide-angle viewing system (Oculus, Inc, Wetzlar, Germany). A standard 3-port 20-gauge pars plana vitrectomy approach was used in all cases. The extent and severity of PVR was assessed during surgery. No additional buckling was performed in any of our cases.

Membranes were removed by peeling with a membrane pick or an intraocular forceps, usually after staining with Membrane Blue (DORC, Zuidland, The Netherlands). A retinectomy was carried out in the area of anterior proliferation when it caused persistent foreshortening of the retina despite removal of visible preretinal membranes.

Retinectomy implies the complete excision of the entire complex formed by the vitreous base, anterior membranes, and shortened retina. Much attention was paid to the anatomy surrounding old sclerotomy sites: often anterior traction was present at these sites, and if so, these areas were included in the retinectomy. The circumferential extent of anterior PVR dictated the minimal circumferential extent of the retinectomy, but often a larger circumferential extent of the retinectomy was carried out. We believe that a retinectomy of at least 6 clock hours, preferably covering the area between the 3- and 6-o’clock meridian, has the greatest chance of relieving circumferential traction.

The anteroposterior extent of the retinectomy was dictated by the individual pathologic features. The lower edge of the retinectomy was aimed to lie between the equator and the lower vascular arcade, preferably closer to the equator than to the vascular arcade. After completion of the retinectomy and coagulation of cut vessels, we used perfluorocarbon liquid to flatten the entire retina. Laser retinopexy was applied in 4 rows along the entire edge of the retinectomy. The spots were placed in a confluent pattern leading to a total width of the barrage of 1500 to 2000 μm. We consider the starting and ending points of the retinectomy incision as crucial and most vulnerable, and therefore often secured these locations with 5 rows of laser coagulates. We did not apply 360-degree laser as standard procedure. At the end, silicone oil was infused after air–fluid exchange or during direct perfluorocarbon liquid–silicone oil exchange. In aphakic cases, a peripheral iridectomy was made with the vitrectomy probe in the lower quadrant.

Pars plana lensectomy was carried out with endocapsular fragmentation of the crystalline lens. We attempted to preserve an intact anterior capsule. Epithelium was removed from the posterior surface of the anterior capsule as much as possible to prevent postoperative capsular opacification.

Patients were instructed to posture for 5 days after surgery. During waking hours, patients were told to sit straight but to keep their gaze angled slightly downward, at approximately 30 degrees from the horizon. At night, patients were asked to incline the upper half of their beds to attain a 45-degree inclination of the head and to avoid lying nose-up.

Preoperative data collected from the charts included patient gender and age, preoperative visual acuity, number and type of previous operations, and original diagnosis. Data obtained from the operative charts included the PVR grade as defined by the classification system developed by the Retina Society, extent of the retinectomy, and peeling of macular membranes. Information gathered in the postoperative period included length of follow-up, retinal attachment duration after the initial retinectomy procedure, final attachment rate, type and number of successive operations, occurrence of postoperative complications as hypotony, glaucoma, keratopathy, and final visual acuity. The intraocular pressure was considered elevated when it was more than 25 mm Hg, and the eye was considered hypotonus when the intraocular pressure was 5 mm Hg or lower. The primary outcome was anatomic success, defined as complete retinal reattachment.

Statistical analysis was performed using SPSS software for Windows (version 16.0; SPSS, Inc, Chicago, Illinois, USA) for Pearson chi-square and Mann–Whitney U test analysis. Averaged visual acuity data are expressed in logarithm of the minimal angle of resolution units. All statistical analyses on visual acuity data were performed using distribution-free ranking tests.


A total of 123 eyes from 122 patients were included. All cases had a follow-up of at least 6 months. Baseline ocular and patient characteristics are summarized in Table 1 . The median age was 59 years (range, 19 to 87 years). Mean follow-up was 15.7 months (range, 6 to 44 months). Twenty-one percent of eyes had a visual acuity of better than 0.1. Twenty-nine percent of cases were primary RRD procedures and 70.7% of cases were reoperations. Several patients had 1 or more high-risk factors, including: chronic retinal detachments in 5 cases (3 cases of subretinal proliferation), 2 cases of subretinal and choroidal hemorrhage, 6 cases of infectious disease, 3 cases of preceding blunt trauma, 2 cases of myopic macular holes, 2 cases of mentally disability. Two eyes, in retrospect, had inoperable pathologic features, with a fixed funnel and extensive central fibrosis. We did not exclude these cases, and they were considered failures.


Retinectomy in Proliferative Vitreoretinopathy: Baseline Characteristics (n = 123)

Median age, yrs (range) 59 (19 to 87)
Follow-up, mos (range) 15.7 (6 to 44)
Male 69 (56.1%)
Female 54 (43.9%)
Operated eye
Right 52 (42.3%)
Left 71 (57.7%)
Macula status
On 19 (15.4%)
Off 101 (82.1%)
Not documented 3
Baseline visual acuity
>0.1 26 (21.1%)
CF to 0.1 42 (34.1%)
<CF 55 (44.7%)
No. of previous RRD-related operations
0 36 (29.3%)
1 60 (48.8%)
2 24 (19.5%)
3 2 (1.6%)
4 1 (0.8%)
Lens status
Phakic 58 (47.2%)
Pseudophakic 57 (46.3%)
Aphakic 8 (6.5%)

CF = counting fingers; RRD = rhegmatogenous retinal detachment.

Table 2 shows intraoperative characteristics. Local anesthesia with a retrobulbar or sub-Tenon block was used in 75.6%. Total duration of surgery was on average 82.0 minutes for vitrectomy alone and 91.4 minutes when combined with lens extraction. Posterior PVR was present in 39.0%, defined as fixed starfolds. The extent of anterior PVR is reported in Table 2 . Intraoperative dissection of macular membranes was deemed necessary in 41 cases. Retinectomy was tailored to the extent of anterior PVR and involvement of old sclerotomy sites. All but one case underwent tamponade with silicone oil, and in 1 case, SF 6 gas was chosen as tamponading agent because of a relatively small retinectomy superiorly.


Retinectomy in Proliferative Vitreoretinopathy: Intraoperative Characteristics (n = 123)

Type of anesthesia
General 30 (24.4%)
Local 93 (75.6%)
Pars plana lensectomy 24 (19.5%)
Phacoemulsification 9 (7.3%)
Posterior PVR 48 (39.0%)
Anterior PVR extent (clock hours)
1 to 5 91 (74.0%)
6 to 12 32 (26.0%)
Peeling posterior PVR 41 (33.3%)
Tamponade type
Silicone oil 122 (99.2%)
Gas 1 (0.8%)
Retinectomy extent (clock hours)
1 to 5 17 (13.8%)
6 to 12 106 (86.2%)

PVR = proliferative vitreoretinopathy.

No unmanageable intraoperative or postoperative complications occurred in this series. In one case (0.8%), a choroidal hemorrhage was stabilized during surgery and resolved in the direct postoperative period. No cases of endophthalmitis or sympathic ophthalmia were encountered. An iatrogenic posterior break occurred in 5 cases (4.1%), and in 2 cases, droplets of silicone oil migrated to the anterior chamber (1.6%).

Table 3 reports the anatomic results of the initial retinectomy surgery, in relation to preoperative variables. Retinal reattachment after the initial retinectomy procedure was achieved in 95 (77.2%) of 123 eyes. In 20 of the 95 successful cases, silicone oil was left in the eye chronically. Of the 28 primary failures, 2 were detached from the day after surgery; 16 occurred under oil tamponade and 10 occurred after planned removal of silicone oil. Of the 26 redetachments, 24 were the result of reproliferation and were repaired by membrane dissection, extension of the retinectomy, or both. In 2 cases, redetachments were the result of a new or previously missed hole without reproliferation. Of the 28 primary failures, 3 had no further operations, 20 were reattached after 1 reoperation, and 3 required 2 further operations. Two were not successfully reattached. The decision not to operate further was reached after 1 and after 2 additional operations, respectively. Of all preoperative variables, only the extent of anterior PVR showed a statistically significant relation to anatomic outcome.


Retinectomy in Proliferative Vitreoretinopathy: Predicting Primary Anatomic Success

Success (n = 95) Failure (n = 28) P Value
Age (median) 59 57 .348 a
Gender .575 b
Male 52 (54.7%) 17 (60.7%)
Female 43 (45.3%) 11 (39.3%)
Operated eye .217 b
Right 43 (45.3%) 9 (32.1%)
Left 52 (54.7%) 19 (67.9%)
Macula status .445 b
On 16 (17.2%) 3 (11.1%)
Off 77 (82.8%) 24 (88.9%)
Lens status .125 b
Phakic 41 (43.2%) 17 (60.7%)
Pseudophakic 46 (48.4%) 11 (39.3%)
Aphakic 8 (8.4%) 0 (0%)
Anterior PVR extent (clock hours) .021 b
1 to 5 (n = 91) 75 (78.9%) 16 (57.1%)
6 to 12 (n = 32) 20 (21.1%) 12 (42.9%)
Posterior PVR .636 b
Present 36 (37.9%) 12 (42.9%)
Not present 59 (62.1%) 16 (57.1%)
Baseline VA .931 a
Median Snellen CF CF
Mean logMAR 2.11 2.09
Retinectomy extent (clock hours) .935 b
1 to 5 (n = 17) 13 (13.7%) 4 (14.3%)
6 to 12 (n = 106) 82 (86.3%) 24 (85.7%)

CF = counting fingers; logMAR = logarithm of the minimal angle of resolution; PVR = proliferative vitreoretinopathy; VA = visual acuity.

a Mann–Whitney U test.

b Pearson chi-square test.

Final retinal reattachment was achieved in 118 (95.9%) of 123 eyes as determined at the final follow-up visit ( Table 4 ). In 87 cases, the state of final reattachment was retained after silicone oil removal. Final visual acuity of counting fingers or better was reached in 82.9% of cases. Within the follow-up, postoperative macular pucker formation was common. In 38 (30.9%) of 123 eyes, a premacular membrane was removed surgically by peeling.


Retinectomy in Proliferative Vitreoretinopathy: Final Results (n = 123)

Final success 118 (95.9%)
With oil removed 87 (70.7%)
With chronic oil 30 (24.4%)
Gas 1 (0.8%)
Final failure 5 (4.1%)
Final visual acuity
>0.1 53 (43.1%)
CF to 0.1 49 (39.8%)
<CF 21 (17.1%)
No. macular peel operations 38 (30.9%)
Median postoperative IOP (mm Hg) 13
>25 2 (1.6%)
5 to 25 without medication 106 (86.2%)
5 to 25 with medication 10 (8.1%)
<5 5 (4.1%)
No. of reoperations
0 95 (77.2%)
1 21 (17.1%)
2 4 (3.3%)
Total no. of RRD operations
1 28 (22.8%)
2 51 (41.5%)
3 37 (30.1%)
4 6 (4.9%)
5 1 (0.8%)

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 17, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Primary Retinectomy in Proliferative Vitreoretinopathy

Full access? Get Clinical Tree

Get Clinical Tree app for offline access