Clinical Spectrum and Long-Term Anatomical and Functional Outcomes of Pars Plana Vitrectomy in Retinal Detachments Associated With Choroidal Coloboma

Abstract

Purpose:

Choroidal coloboma (CC) eyes are associated with a high risk of retinal detachment (RD), which are challenging to repair.

Methods:

We retrospectively evaluated medical records of 30 patients with CC associated with RD who underwent pars plana vitrectomy at our tertiary care center. Data recorded were age, CC and RD morphology, breaks, resurgeries performed, and final outcome.

Results:

Patients’ ages ranged from 2 to 49 years. CC was present in 53 of 60 eyes (88%), with 41 of 60 eyes (68.3%) associated with iris coloboma. RD was present in 36 of 53 CC eyes (67.9%), with 10 of 53 (18.9%) having bilateral RD. Thirty-one eyes with RD underwent 23-gauge pars plana vitrectomy. Colobomatous region breaks were detected in 5 of 31 eyes (16.1%), peripheral breaks only in 12 of 31 (38.7%), and both types in 3 of 31 eyes (9.7%). Silicone oil endotamponade was used in 29 of 31 eyes (93.5%) and perfluoropropane (C₃F₈) in 2 of 31 eyes (6.45%). Median postoperative follow-up duration was 31.5 months. Fourteen of 31 eyes (45.2%) developed redetachment requiring resurgery, with 50% developing within 6 months. Eighteen of 31 eyes (58.1%) maintained retinal attachment at last follow-up without endotamponade, whereas the rest were oil filled. Final visual outcome between macula-involved and macula-spared CC eyes was not significantly different; however, best visual acuity achieved any time during follow-up was significantly better in macula-spared eyes.

Conclusions:

RD surgery in CC eyes has a risk of frequent redetachments and subnormal gain of vision over the long term, especially in colobomas involving the macula. Endolaser in 360° peripheries in addition to barrage of the margin of the coloboma may be considered to help maintain the attached retina.

Introduction

Choroidal coloboma (CC) occurs because of an aberration in the closure of the embryonic fissure during development of the fetal eye. The colobomatous area is histologically devoid of retinal pigment epithelium (RPE), Bruch membrane, choriocapillaris, and choroid, and the normal neuroretina is replaced by a rudimentary thin structure called the intercalary membrane (ICM), which covers the bare staphylomatous sclera. These eyes are at risk of retinal detachment (RD) mainly because of breaks that form in the ICM and patent communication between the subretinal and sub-ICM spaces. ¹

The incidence of RD in coloboma eyes has been reported to range from 2.4% to 42%. ^{2 -5} RD associated with colobomas has been historically difficult to manage even with different surgical techniques. Scleral buckling has been tried with varied anatomical success rates ranging from 35% to 57%. ^6,7 Postoperative photocoagulation at the margins of the coloboma to create chorioretinal adhesions may be tried but is difficult because a large number of coloboma patients are young or have nystagmus or microphthalmos. Pars plana vitrectomy (PPV) offers benefits over buckling for its effective release of vitreous traction, endophotocoagulation along the margin of the coloboma, and endotamponade with gas or silicone oil. ¹ Moreover, identification of breaks during PPV is more accurate, and direct closure of breaks with cyanoacrylate glue, cryotherapy, and endophotocoagulation has been reported. ^8,9 Despite advances in surgical techniques and improvement in anatomical outcomes of coloboma eyes with RD, functional prognosis remains poor mainly because of the occurrence of redetachments and complications, namely intractable glaucoma. In this article we present the outcomes of 23-gauge PPV with endotamponade in colobomatous eyes with RD with a long duration of follow-up.

Methods

Thirty patients with CC who presented to our tertiary care center between 2009 and 2018 were included in this retrospective study. The study adhered to the tenets of the Declaration of Helsinki and was started after obtaining ethical clearance from the institutional ethics review board (RET201600193). Patient records were included for data analysis if they had an RD that was associated with CC in at least 1 eye and had undergone 1 or more surgeries for the same. Past medical records of all patients were meticulously evaluated to document demographic and clinical data, which included age at presentation, sex, date of RD diagnosis, best-corrected visual acuity (BCVA), intraocular pressure (IOP), results from anterior segment examination, type of CC, type and extent of RD, location and number of retinal breaks detected, nature of initial surgery performed, follow-up status of the patient, and number of resurgeries performed along with reasons and clinical status at final follow-up.

Visual acuity (VA) was measured using the Snellen VA chart and converted to logarithm of the minimal angle of resolution (logMAR) scale. Counting fingers VA was considered equivalent to Snellen 6/600 (2.0 logMAR), hand movements VA to Snellen 6/1200 (2.3 logMAR), perception of light vision to 6/3000 (2.7 logMAR), and no perception of light vision to 6/6000 (3.0 logMAR). ¹⁰ Complications arising from surgery or the natural history of the disease process were also noted. At each follow-up, BCVA, IOP, and color fundus diagrams were documented.

Surgical Protocol

All patients underwent 23-gauge PPV (with an encircling No. 42 band in select cases) along with pars plana lensectomy in cases having cataractous changes in the lens. Induction of posterior vitreous detachment (PVD) was performed beginning near the optic disc, assisted by injecting triamcinolone acetate into the vitreous cavity. Membrane peeling was performed in cases with advanced proliferative vitreoretinopathy (PVR) changes. Anterior vitreous base dissection was performed in all cases. ICM and coloboma margin were carefully inspected for localization of breaks, assisted by Schlieren imaging.

Fluid air exchange helped identify whether all parts of the retina had settled, especially over the colobomatous region. In cases with PVR changes in which the ICM or the posterior pole would not settle easily, perfluorocarbon liquid was used to reattach the posterior pole, which also allowed effective endophotocoagulation of the coloboma margins. Endotamponade in the form of silicone oil (1000 or 5000 cSt) and, in select cases, perfluoropropane (C₃F₈) was used. One-eyed patients, eyes with large colobomas involving the macula and disc, and eyes undergoing recurrent redetachments were considered for injection of 5000 cSt oil. Gas tamponade was considered according to the clinician’s discretion in rare cases with isolated localized superior breaks with limited RD, no ICM break, and no PVR changes.

Statistical Analysis

Data were entered in a Microsoft Excel spreadsheet and analyzed using SPSS for Windows software (version 20.0, IBM Corp). Data normality was checked using histograms. Mean, median, and SD of each variable were recorded and the statistical differences among groups were tested using the independent sample t test for parametric data and Mann-Whitney test for nonparametric data. Difference was considered significant at a 2-tailed P value of .05.

Results

Demographic Characteristics

Between January 2009 and December 2018, 30 patients with CC presented to our center with RD in at least 1 eye. The mean age of the patients at presentation was 18 ± 11.99 years (range, 2-49 years) (Table 1). Nine patients (30%) were younger than 10 years and 19 (63.3%) were male.

Table 1.

Demographic and Clinical Details of Patients Undergoing Surgery.

	N = 31
Mean age at presentation, y	18 ± 11.99
Prophylactic laser barrage performed	2
Types of retinal detachment
Total	18
Superior	4
Inferior	4
Temporal	3
Nasal	1
Subtotal	1
Types of coloboma
Ida Mann type 1	15
Ida Mann type 2	6
Ida Mann type 3	7
Ida Mann type 4	2
Ida Mann type 5	1
Mean duration of follow-up, mo	44.07 ± 30.97
Total eyes having redetachments, No. (%)	14 (45.2)
Mean duration eyes stayed without endotamponade, mo	41.8 ± 30.05
Eyes filled with oil at last follow-up, No. (%)	13 (41.9)

CC was present in 53 of 60 eyes (88%) and bilateral colobomas constituted 46 of 60 (76.7%) eyes. Presence of coloboma was associated with microcornea in 23 of 60 (38.3%) eyes, nystagmus in 12 of 60 (20%) eyes, and iris coloboma in 41 of 60 (68.3%) eyes. A total of 36 of 53 (67.9%) CC eyes presented with RD, with 10 of 53 (18.9%) eyes showing bilateral detachments. Six eyes (6 of 60, 10%) were blind, owing to anophthalmos (1 eye) and previous RD (5 eyes, 3 eyes phthisical).

A total of 31 eyes with RD were managed surgically at our center. The macula was included within the margins of the coloboma in 14 eyes. Breaks in only the colobomatous region were detected in 5 of 31 eyes (16.1%), only the peripheral region in 12 of 31 (38.7%) eyes (P = .172), and both types of breaks in 3 of 31 eyes (9.7%). No breaks could be identified in 11 of 31 (35.5%) eyes. PVR grade C or greater was present in 4 of 31 (12.9%) eyes. The mean VA at presentation was 1.728 ± 0.91 logMAR, with macula-spared eyes having VA of 1.43 ± 0.89 logMAR and macula-involved eyes having VA of 2.04 ± 0.82 logMAR (P = .06). Significant cataractous changes were present in 12 eyes.

Surgical Management

All 31 eyes underwent 23-gauge PPV. Additional pars plana lensectomy was performed in 12 eyes and a belt buckle was put in 22 eyes. PVD was preoperatively determined using B-scan ultrasonography and found in 14 of 31 (45.2%) eyes. In eyes that did not already have PVD, it was induced from above the optic disc using suction. PVR cases underwent membrane peeling to release traction over the retina. Fluid-air exchange was performed along with subretinal fluid drainage either through a visible break or an iatrogenic retinotomy made in the noncolobomatous retina inferiorly. Silicone oil intraocular tamponade was used in 29 eyes (24 eyes received 1000 cSt, 5 received 5000 cSt), with 2 eyes receiving intravitreal nonexpansile concentration of gas (C₃F₈). Initial anatomical success was defined as retinal attachment on examination on postoperative day 1. This initial anatomical success was found to be 100% in our cohort.

The mean duration of follow-up of eyes in our study beginning with the first surgery was 44.07 ± 30.97 months (median 31.5; range, 4-119 months). Fourteen (45.2%) eyes developed redetachment, of which 7 (22.6%) occurred with silicone oil in situ, 1 with gas in situ, and 6 (19.4%) after initial oil removal. Total redetachments were 20, and redetachments occurring per eye were 20 of 14, that is, 1.4 (range, 1-4 redetachments). Among the eyes that redetached, 8 (57.1%) had colobomas involving the macula, whereas 5 (35.7%) had developed PVR greater than grade C. Among eyes that did not redetach, 7 developed PVR greater than grade C (P = .762). Revision surgeries included shaving of remaining vitreous base, peeling of membranes, and release of traction. The average duration before an eye suffered from a redetachment was 10.07 ± 12.68 months (median 6.5; range, 1-48 months), with 7 eyes (7 of 14, 50%) showing the first redetachment within 6 months. Of the eyes with RD, 12 eyes underwent a second silicone oil injection (SOI).

Six-month anatomical success was defined as attachment of the retina at the sixth month of follow-up after the first surgery, irrespective of the number of resurgeries performed by that time. This 6-month success was found in 27 of 31 eyes (87.1%). Final anatomical success was defined as the presence of the attached retina (with or without silicone oil tamponade) at the patient’s last follow-up, which was achieved in 26 of 31 eyes (83.8%) in our series. The total number of SOIs performed in the 31 eyes was 48, with the mean interval between oil injection (primary or for redetachment) and removal or reinjection for redetachment being 8.12 ± 6.6 months (range, 3-36 months). At the last follow-up, 13 of 31 eyes (41.9%) were still oil filled (9 with 1000 cSt oil, 4 with 5000 cSt oil). Eighteen eyes (18 of 31, 58.1%) maintained retinal attachment at last follow-up without endotamponade, and the mean duration for which these eyes stayed attached without tamponade was 41.8 ± 30.05 months (median 36.5, range, 6.5-107 months).

Functional outcomes at final follow-up have been summarized in Table 2. Mean BCVA improved to 1.339 ± 0.804 logMAR at 6 months (P = .08) and to 1.52 ± 0.936 logMAR at the final follow-up (P = .38). Overall, 16 eyes (16 of 31, 51.6%) showed improved vision from baseline and 11 of 31 (35.5%) eyes showed deterioration of vision. VA improved to greater than or equal to 6/60 Snellen in 13 of 31 eyes (41.9%), of which 5 eyes had macula-involving colobomas. Counting fingers vision or less was seen in 10 of 31 eyes (32.2%), and all of these eyes had macula-involving colobomas.

Table 2.

Anatomical and Functional Outcomes of Eyes With Choroidal Coloboma Undergoing Surgery for Retinal Detachment.

	Percentage
Initial anatomical success rate	100
Six-month anatomical success rate	87.1
Final anatomical success rate	83.8
	Mean ± SD	P Value
BCVA at presentation
Overall (n = 31)	1.728 ± 0.91
Macula-involved CC eyes (n = 14)	2.04  + 0.82	.9
Macula-spared CC eyes (n = 17)	1.43 ± 0.89
BCVA at final follow-up
Overall (n = 31)	1.52 ± 0.936	.38
Macula-involved CC eyes (n = 14)	1.99 ± 0.84	.10
Macula-spared CC eyes (n = 17)	0.91 ± 0.91
Best BCVA achieved after surgery
Overall (n = 31)	1.19 ± 0.844	.02 a
Macula-involved CC eyes (n = 14)	1.86 ± 0.79	<.001 a
Macula-spared CC eyes (n = 17)	0.73 ± 0.55

Abbreviations: BCVA, best-corrected visual acuity; CC, choroidal coloboma.

^a Bold indicates significance at a value of less than 0.05.

We documented the best VA achieved at any point during the entire follow-up duration in the operated-on eyes, which was significantly better than the presenting BCVA (P = .02). Interestingly, the best VA in the macula-spared coloboma eyes was significantly better than eyes with macular involvement (P < .001), although there was no significant difference between the final follow-up VAs achieved in them.

Overall, the eyes that underwent redetachment achieved poor VA at final follow-up (1.72 ± 1.02 logMAR), which was not significantly different from their preoperative VA (1.78 ± 0.94 logMAR). However, their best VA (1.32 ± 0.97 logMAR) was better than their outcomes at final follow-up, although not significant (P = .29). We also compared the final BCVA of eyes that redetached within 6 months of surgery (1.92 ± 1.07 logMAR) with those that redetached after 6 months of surgery (1.52 ± 1.0 logMAR) and found there was no significant difference (P = .47).

Of 23 CC eyes with microcornea, 16 developed RD. At final follow-up, 9 of 16 eyes (56.2%) achieved final VA greater than or equal to 6/60 Snellen. Nystagmus was present in 11 eyes with RD, and VA greater than or equal to 6/60 Snellen was seen in 6 of 11 eyes (54.5%).

Complications

Although VA was favorable for several patients, a variety of complications, apart from PVR, were observed in 24 of 31 eyes (77.4%) (Table 3). PVR was seen in 12 of 31 eyes (38.7%) after the first surgery (5 of them redetached). The most common complication was glaucoma (17 of 31, 54.8%), followed by cataract formation. All the glaucoma patients were kept on medications and by the final follow-up, 82.3% of eyes were on more than 3 glaucoma drugs. A total of 12 eyes had undergone lensectomy along with the primary surgery. Of the remaining 19, 13 eyes (68.4%) developed cataract during the follow-up, for which a cataract extraction with or without intraocular lenses (IOL) was performed. Other complications were phthisis bulbi (2 of 31, 6.4%), neovascularization of the iris (1 of 31), ciliary staphyloma (1 of 31), bullous keratopathy (2 of 31, 6.4%), and foveoschisis (1 of 31). The case with foveoschisis was settled with 1 intravitreal triamcinolone acetonide injection.

Table 3.

Complications Arising After Surgery Until Final Follow-Up.

	No. (%), N = 31
Total eyes with glaucoma	17 (54.8)
Glaucoma requiring more than 3 glaucoma medications	14 of 17 (82.3)
Postvitrectomy cataract requiring lens extraction with or without intraocular lens implantation	13 (41.9)
Proliferative vitreoretinopathy	12 (38.7)
Phthisis bulbi	2 (6.4)
Neovascularization of iris	1 (3.2)
Ciliary staphyloma	1 (3.2)
Bullous corneal decompensation	2 (6.4)
Foveoschisis	1 (3.2)

Discussion

CC arises from defective embryonic fusion of the optic cup, leading to a cleft or fissure in the inferior aspect of the cup. It may be associated with coloboma of the iris, microphthalmos, microcornea, cataract, posterior staphyloma, nystagmus, and strabismus. Hermann Schubert studied the histopathology of CCs extensively and found that near the margin of the coloboma, the extracolobomatous retina splits into 2 layers: The inner nerve fiber layer continues as the thin ICM over the coloboma, and the outer layers turn back, become disorganized, and join the RPE, which may become hyperplastic. ¹¹ The choroid also becomes thickened and hyperpigmented near the margin. The RPE pump function is expectedly absent in the region of the ICM. ¹²

Coloboma RDs are challenging cases to repair, with varied success rates reported in the literature ranging from 35% to 100%. ^2,6,12,13 We summarized outcomes of previous studies with follow-up in Table 4 and compared these results with our outcomes. Our cases achieved an initial success rate of 100% and a final success rate of 83.8%. Hussain et al ¹⁵ reported a final attachment rate of 85% after a mean follow-up of 40.8 months, and Gopal and colleagues ¹ reported a 75% attachment rate after a mean follow-up of 13.4 months. Our patients had a median follow-up duration of 31.5 months, which is much longer than previously reported in the literature. Hence, we could evaluate the natural course of the disease after the RD surgery more completely, along with documentation of multiple redetachment management techniques and associated complications.

Detachments have been described as occurring because of a variety of breaks: (1) in the ICM, (2) in the margin, (3) in the peripheral retina, (4) in the peripheral retina and margin, (5) in the margin and the ICM, and (6) in the peripheral retina, margin, and ICM. ¹ Similar to the previous literature, we also found peripheral causative breaks in the peripheral retina, indicating an abnormal vitreoretinal interface in colobomatous eyes. ¹⁵

Several different treatment modalities have been described for RDs associated with colobomas, ranging from radial scleral buckles, to diathermy and cryotherapy, but most of them are of only historical importance today. PPV with endotamponade has been described as an effective method of removing vitreous traction and settling the retina and ICM. ¹

A total of 25.8% (8 of 31) of eyes had ICM breaks in our study, as compared with 38% to 100% in previously reported studies. ^9,14,17 Gopal et al observed that almost two-thirds of ICM breaks were present within 2 disc diameters of the coloboma margin, and the rest were oval and atrophic breaks seen inside the ICM center without any evidence of being caused by tractional forces. ¹⁸ Hence, in all the cases where no break could be identified in the ICM, the breaks might have been present in the margin, which may be extremely difficult to localize. Because the retina overlying the coloboma is extremely atrophic, transparent, and may undergo splitting as well, identifying the small, narrow, slit-like holes within the colobomatous crater may be a difficult task, more so if they are present near the margin. In this regard, ICM breaks may be identified with more certainty and precision during PPV. ^1,9

In many cases, because localization of breaks in the ICM may be difficult, endodrainage through breaks in the ICM may not be possible, and in cases without an associated break in the normal retina, a drainage retinotomy may be made in the extracolobomatous retina or in the ICM and subretinal fluid drained. Drainage retinotomy was made in 64.5% of our patients’ eyes requiring subretinal fluid drainage; however, none of the redetachment cases showed reopening of the retinotomy sites. Moreover, perfluorocarbon liquid may be used to attach the retina and the ICM, which allows easy retinopexy using endolaser.

The colobomatous margin has been found to be weaker in children and achieves strength in adulthood, hence extension of extracolobomatous RDs into the colobomatous area via the sub-ICM space or vice versa is a possibility. Hence, as long as the marginal vessels in the coloboma are intact, offering adequate vascular supply with a thickened choroid and hyperplastic RPE, the margin would offer a strong seal from extension of detachment. However, with increasing age, colobomas may also suffer from mechanical stress along the extent of the margin, with stretching of the ocular walls and vitreous liquefaction causing tractional forces that may lead to disturbance of the coloboma margin and formation of atrophic breaks. ¹⁹ In our series, 70% of the patients were older than 10 years.

Although breaks in the margin are extremely difficult to localize, prophylactic treatment along the margin has been described using laser photocoagulation, which forms strong chorioretinal adhesions in the region, thereby helping restrict extension of ICM detachments even without treating ICM breaks. ²⁰ Previous studies evaluating the coloboma margin using optical coherence tomography showed that focal communications exist between the sub-ICM and subretinal spaces in these RD eyes. ^21,22 Determining the exact location of these communications is a challenging task, and new connections may develop during vitrectomy, hence surgeons may consider lasering the coloboma margin instead of searching for and lasering each ICM break individually. However, there has been concern regarding nerve fiber layer damage and visual field loss due to this photocoagulation around the coloboma margin, and hence macular-sparing photocoagulation is preferred.

With almost 40% of eyes having peripheral breaks (round holes and horse-shoe tears) associated with RD, it seemed prudent to laser 360° around the ora in addition to the coloboma margin. Moreover, as Hussain et al suggested, if the macula is included within the margins of the coloboma, the risk of vision loss due to photocoagulation of the papillomacular bundle may outweigh the benefits of RD prophylaxis. ¹⁵ Hence, 360° of peripheral laser seems to have an additive beneficial action, especially in cases in which the macula is spared from laser. However, although this 360° photocoagulation has been a routine practice in our center in such coloboma RD eyes, because we do not have a comparative arm, we could not quantify the exact benefit. Also, in eyes that developed redetachments, few cases had new additional break formation in the periphery or reopening of existing breaks.

Eyes that developed redetachment, requiring repeated membrane peeling along with endotamponade, had no statistically significant difference in the rate of PVR greater than grade C from eyes that did not redetach. We believe that the redetachments in these eyes were mostly because of difficult-to-visualize breaks in the ICM, the reopening of existing breaks, or the formation of new breaks. Apart from primary injection of 5000 cSt silicone oil in 1-eyed patients, providing longer tamponade proved to be important in redetachment cases. PVR development may be an important cause of redetachment in coloboma eyes. Our cohort showed a PVR grade greater than C in 12 of 31 eyes (38.7%); however, this number did not reach statistical significance as a risk factor for redetachment. Hussain and colleagues ¹⁵ reported a 100% incidence of PVR in their cases, whereas Gopal et al ¹ had reported a 22.4% incidence of PVR with 9% of eyes having a severe PVR of greater than C3.

The functional outcomes of our cases were optimal, with VA improving to more than or equal to 6/60 Snellen in 41.9% of eyes, although a number of these eyes suffered from redetachments. The VA at final follow-up was not significantly different from preoperative VA, although 51.6% of eyes showed improvement in VA from baseline, and the best VA achieved during follow-up was significantly better than baseline. This best VA may indicate that the surgical outcomes are better than the final follow-up VA (if both are different). Subgroup analysis revealed that although the best VA in eyes with macula-sparing colobomas was significantly better than that of the macula-involved eyes, at the final follow-up their outcomes were similar, reinstating that however good the management may be, functional end results are not too impressive. Hence, it can be concluded that once the retina is detached, coloboma eyes are difficult to salvage in the long run. However, surgery with effective tamponade must be attempted, whereby comes the importance of close and long-term follow-up, along with 5000 cSt oil in 1-eyed cases or when the physician strongly suspects imminent redetachment.

While looking at functional outcomes, it may also be prudent to consider preoperative structural factors that may also affect the final outcome, like microcornea or nystagmus. A total of 56.2% of eyes having microcornea and 54.5% of eyes with nystagmus appeared to improve beyond 6/60 Snellen, which appears to be an optimal improvement. However, comparison with eyes not having these features to arrive at a meaningful conclusion would be difficult in our cohort because of the presence of multiple different variables, like recurrent detachments, PVR status, or glaucoma, and especially because the numbers in each group were not high enough.

None of the previous studies have reported the management and follow-up of cases over such a long period of time (Table 4). Two recent studies have documented coloboma RD management. Hocaoglu et al followed up 10 operated-on coloboma RD eyes for a mean period of 28.8 months and reported a final attachment rate of 90%. ²³ The other study by Suwal and colleagues involving 12 eyes undergoing surgery mentioned a follow-up period of 6 months, which is again inadequate to properly assess the final outcome. ²⁴ In our series, 45.2% of eyes developed redetachment, with 50% of them showing redetachment within 6 months after the first vitrectomy surgery. Several of these cases required SOIs to be repeated for redetachments. Although in our study 83% of patients showed an attached retina after a median follow-up of 31.5 months, only 18 of 31 (58%) patients maintained the attachment without endotamponade.

Table 4.

Summary of Surgical Outcomes of Different Reports From the Literature.

Authors	Year	No. of Eyes	Gauge of PPV	Final Anatomical Success, %	Follow-Up Mean Duration, mo
Gopal et al 14	1991	17	NA	81.8	6
Gopal et al 1	1998	85	NA	81.2	13.4
Pal et al 13	2006	42	NA	88.1	14
Wei et al 16	2014	5	20	100	20.8
Hussain et al 15	2017	14	NA	86	40.8 (median 11)
Present study	2019	31	23	83.8	44.07 ± 30.97 (median 31.5)

Abbreviations: NA, not available; PPV, pars plana vitrectomy.

Apart from disease-related problems, patients may also develop complications secondary to multiple retinal surgeries, like glaucoma, cataract, or corneal decompensations, that need to be managed in parallel. Because coloboma eyes have a high rate of association with microcornea or microphthalmos, these eyes with shallow anterior chamber depth may already be predisposed to glaucoma owing to possible anterior segment dysgenesis. Moreover, a number of eyes underwent lensectomy along with PPV, and aphakia may also have predisposed patients to increased IOP. Finally, 1-eyed patients and redetachment eyes needed longer endotamponade, which also substantially increased the risk of glaucoma.

Given this guarded visual prognosis and such frequent need for repeated surgeries, one might ponder whether coloboma eyes with RD need to be operated on at all. However, in this regard, Fivgas and Capone in their review found that 38% of pediatric RDs that underwent surgical repair had at least equal to or better VA than their fellow eyes at final follow-up. ²⁵ In our study, almost one-fifth of patients had bilateral RDs and around one-third of eyes were blind at presentation. Hence, considering that these patients were at a very high chance of complete vision loss due to the affliction of the other eye, the decision to operate on the RD eyes under an extremely guarded prognosis was made. Moreover, the eyes with recurrent redetachments ended up having final vision almost similar to the preoperative status. Hence, it may be inferred that intervention should take place, if not for a better outcome, then at least to preserve the VA in these eyes. Several complications, most commonly glaucoma and cataract, may occur in these eyes after surgery. Although cataracts may be amenable to intraocular-lens implantation surgery, glaucoma may plague the patient with multiple medicines and recurrent follow-ups.

Conclusions

To summarize, RD in an eye with CC is a challenging case to tackle, having chances of failure of attachment after surgery with frequent redetachments and subnormal gain of vision over the long term. Colobomas, especially ones involving the disc and the macula, have a poorer prognosis than those that spare the same. But, keeping in mind the bilateral predisposition of this disease and the possibility of having at least 1 detachment in either eye over a lifetime, such cases need to be treated at the earliest, regularly followed up for complications, and prognosticated in an individualistic manner.