Management of Retinal Detachment with Choroidal Coloboma LINGAM GOPAL, MD/ MOHANDAS M. KINI, MD, 1•2 SENGAMEDU S. BADRINATH, MD/ TARUN SHARMA, MD 1
Abstract: Seventeen eyes with retinal detachment secondary to retinal breaks in the colobomatous area were managed by vitrectomy procedures. Endodrainage was done through the break in the colobomatous area, with simultaneous fluid-air exchange. Endolaser photocoagulation was performed along the colobomatous margin posteriorly, while the anterior portion was treated by transscleral cryopexy. Silicone oil was then exchanged with air. In nine cases, the silicone oil had to be removed for complications such as emulsification, glaucoma, and keratopathy; in three of these eyes, oil removal resulted in recurrent retinal detachment. At the 2-month follow-up visit, there was 100% anatomic success, and 12 eyes (70.6%) recovered visual acuity of 10/200 or better. Of the 11 eyes with follow-up of more than 6 months, in 9 (81.8%) the retina remained reattached at last-follow up, and in 6 eyes (54.5%) a visual acuity of 10/200 or better was obtained. Ophthalmology 1991; 98: 1622-1627
Coloboma of the choroid results from a faulty closure of the embryonal fissure. Although the reported incidence is only 0.14% of the general population, 40% of these individuals will develop a retinal detachment sometime during their lifetime. 1 The colobomatous area consists of a thin layer of hypoplastic retinal tissue. The choroid and retinal pigment epithelium are not developed in this region, and the sclera underlying the colobomatous area is usually thin and ectatic, producing a staphyloma. Retinal detachment caused by retinal breaks outside of the colobomatous area can be managed by conventional scleral buckling techniques. These conventional techniques fail, however, if the detachment is caused by a break in the hypoplastic retinal tissue within the colobomatous area. Furthermore, other difficulties may confound the man-
Originally received: February 5, 1991. Revision accepted: June 19, 1991. 1
Sankara Nethralaya, Medical Research Foundation and Vision Research Foundation, Madras, India. 2 Massachusetts Eye and Ear Infirmary, Boston. Presented in part as a poster at the American Academy of Ophthalmology Annual Meeting, Atlanta, Oct/Nov 1990. Reprint requests to Lingam Gopal, MD, Sankara Nethralaya, 18, College Road, Madras 600 006, India.
1622
agement of coloboma-related retinal detachments. Nystagmus can be an associated feature and may be severe enough to make preoperative evaluation difficult. The lack of contrast in the area of the coloboma can make the identification of retinal breaks difficult. Retinal breaks in the colobomatous area are difficult to seal; retinopexy is not effective because of the absence of choroid and retinal pigment epithelium, and the posterior location of the breaks makes it extremely difficult to buckle the same. Hence, in the management of coloboma-related retinal detachments, attempts have been directed toward isolating the coloboma from the rest of the retina, rather than toward sealing the retinal break. Schepens2 advocated drainage of subretinal fluid, followed by the production of chorioretinal adhesion around the periphery of the coloboma using cryopexy, photocoagulation, or both. Patnaik and Kalsi 3 achieved reattachment of the retina in one case by applying a radial buckle that extended to the optic disc. When both sides of a coloboma are involved in the detachment, Wang and Hilton 4 recommend that two radial buckles be applied along the two edges of the coloboma. Wang and Hilton 4 reviewed the literature and found that of 42 eyes with coloboma-related retinal detachment, 18 (43%) had successful reattachment. In their own series of 20 eyes, success was achieved in 7 eyes (35%). They concluded that vitrectomy with intraocular gas tamponade and creation of chorioretinal adhe-
GOPAL et al
•
RETINAL DETACHMENT WITH CHOROIDAL COLOBOMA
Table 1. Preoperative Visual Acuity Visual Acuity
No. of Eyes
Light perception Hand motions Counting fingers
8 5 3 1
1/60 (3.3/200)
Table 2. Status of Fellow Eye Clinical Status Coloboma of the choroid, attached retina, and useful vision Coloboma of the choroid, spontaneous reattachment of retina, and no vision Coloboma of the choroid and retinal detachment Extreme microphthalmos and no vision Phthisis Normal
No. of Eyes
6 1 6 2 1 1 Fig 1. Coloboma of the choroid with retinal break (arrows) within the coloboma.
sion along the edge of the coloboma is the preferred treatment. Silicone oil also has been used as a temporary tamponade in the management of these cases. 5 In our management of coloboma-related retinal detachments, vitrectomy procedures, silicone oil injection, and endolaser photocoagulation to surround the colobomatous area appeared to improve the surgical results considerably. Herein we describe the technique used and the outcome in 17 eyes treated for coloboma-related retinal detachment.
moscope and slit-lamp biomicroscopy with a 90 diopter lens revealed a break in the colobomatous area in 8 eyes (47%) (Fig 1); in the remainder, a break within the coloboma was identified intraoperatively. A peripheral break outside of the coloboma also was present in 2 eyes ( 11 .7%). Proliferative vitreoretinopathy was present in 3 eyes ( 17.6%): grade C2 and 0 2 in I eye each while 1 eye had only subretinal bands.
MATERIALS AND METHODS .
SURGICAL TECHNIQUE
Seventeen eyes of 16 patients (12 male and 4 female) who were diagnosed as having retinal detachment caused by a break in a coloboma of the choroid are the subject of.this report. The average age at presentation was 19.37 years (range, 6 to 37 years). The duration ofloss of vision at the time of surgery ranged from 5 days to 5 years; 9 patients presented more than than 2 months after the onset of visual disability. The preoperative visual acuity ranged from accurate light projection to 1/60 (3.3/200) (Table 1). Nystagmus was an associated feature in seven patients. Coloboma of the iris was present in 14 eyes (82.35%) and microphthalmos was present in 4 eyes (23.53%). Coloboma of the choroid was associated with coloboma of the disc in 13 eyes (76.47%), and this has special relevance to the management discussed below. The status of the fellow eye is given in Table 2. As is shown, the fellow eye had normal vision in only one patient; in ten patients the fellow eye had no useful vision. Retinal detachments attributable to peripheral breaks not within the coloboma were not included in this study. The retinal detachment was considered to be coloboma-related ifthere was extension to the colobomatous area. Careful preoperative evaluation with the binocular indirect ophthal-
A three-port vitrectomy using the Ocutome System (Cooper Medical Devices, San Leandro, CA) was performed in all cases. Lensectomy was not performed routinely but was necessary for various reasons in 12 eyes (70.6%). Careful excision of the vitreous base was performed in eyes with peripheral proliferative vitreoretinopathy. The colobomatous area was carefully inspected with an endoilluminator and high magnification of the operating microscope, and a retinal break within the coloboma was identified in all these cases. When necessary, membrane peeling always began beyond the margins of the coloboma and was then extended to the colobomatous area to avoid problems related to improper identification of tissues. Once the retina was mobilized, fluid-air exchange was performed. Endodrainage was done through the break in the colobomatous area in 12 eyes; in the other 5 eyes, endodrainage was performed through a preexisting or accidental peripheral retinal break. In one case, a retinotomy created for removal of a subretinal band was used for endodrainage. Fluid-air exchange was performed using a back-flush brush needle, with visualization achieved by use of a -80 diopter lens in phakic eyes and without any contact lens in aphakic eyes. An iridectomy 1623
OPHTHALMOLO GY
•
NOVEMBER 1991
Fig 2. Diagram shows retinal tissue stretched across mouth of coloboma, with subretinal air following fluid-air exchange: fundus view.
Fig 3. Diagram shows retinal tissue stretched across mouth of coloboma, with subretinal air following fluid-air exchange: sagittal section.
at the 6-o'clock meridian was required only in aphakic eyes with no coloboma of the iris. In a few cases, at the end of the fluid-air exchange we found the diaphanous retinal tissue stretched taut over the coloboma (Figs 2, 3). This situation is analogous to that created by localized unrelieved traction with shortening of the retina encountered in routine cases of proliferative vitreoretinopathy and may result in subretinal air. This situation could be relieved by relaxing incisions that extended radially toward the edge of the coloboma (Figs 4, 5). 1624
•
VOLUME 98
•
NUMBER 11
Fig 4. Diagram shows retinal flaps adherent to ectatic sclera following relaxing radial retinotomies: fundus view.
Fig 5. Diagram shows retinal flaps adherent to ectatic sclera following relaxing radial retinotomies: sagittal section.
Once the retina was settled, endolaser photocoagulation was applied around the colobomatous area and any peripheral retinal break. Endophotocoagulation was performed in two or three rows of overlapping burns along the entire margin of the coloboma. Where the coloboma involved the optic disc, the functional border of the disc was left untreated. The extreme anterior portion of the colobomatous margin was treated with transscleral cryopexy. When the coloboma did not extend to the ora anteriorly, it could be surrounded completely by endolaser
GOPAL et al
•
RETINAL DETACHMENT WITH CHOROIDAL COLOBOMA
Table 3. Visual Results at 2-month Follow-up
~lATUS
Visual Acuity
No. of Eyes
Percentage
6/36 (20/120) 6/60 (20/200) 3/60 (1 0/200) 1/60 (3.3/200)
1
5.9 29.4
Hand motions
5 6 3
2
.vvv .vv
,,./
Visual Acuity
No. of Eyes
Percentage
6/18 (20/60) 6/24 (20/80) 3/60 (1 0/200)
1 1
9.1 9.1 36.4 45.5
Hand motions
4
5
~
m. Vv ~
Table 4. Final Visual Results*
~
1·v v
35.3
17.6 11.8
OF RETlNA AT LAST EXAMINATIONU
/
..Vv
2 •.Vv ._v v ,_v v
·~~
,/v'-----/
~ EYESWffliSII.ICXJNEOI..
L
/
/
IIETINAATTICHED
/1/ET/NADETICHED
EYf8 WfrH SIIXXlHE Ol.lfEMCNED
Fig 6. Bar graph depicts anatomical status of retina at final examination.
' Of the 11 eyes that had follow-up of 6 months or more.
treatment alone. Silicone oil ( 1000 centistokes) was exchanged with air. In aphakic eyes, the anterior chamber was reformed with saline. Sclerotomies were closed after ascertaining that the intraocular pressure was normal. To the extent possible, patients were encouraged to remain prone in the early postoperative period. Additional laser treatment to any areas of colobomatous margin that had been inadvertently missed was performed postoperatively. In two cases, repeat surgery had to be performed for localized, recurrent retinal detachment. This second procedure consisted of membrane peeling under silicone oil, reattachment of retina by endodrainage under oil, and additional endolaser treatment. Silicone oil was removed from nine eyes; the average interval between silicone oil injection and removal was 8.44 months (range, 4 to 16 months). In general, oil was removed only if complications such as emulsification occurred. In phakic eyes, silicone oil removal, when done, was combined with lensectomy, since significant cataract had developed by the time of oil removal.
RESULTS The average follow-up was 10.76 months (range, 2 to 28 months); 11 eyes (64.7%) were followed for more than 6 months. At the 2-month follow-up visit, all 17 eyes had anatomical reattachment. Visual acuity at 2 months after surgery is given in Table 3. Twelve eyes (70.6%) had visual acuity of 3/60 (10/200) or better. Data from last follow-up also were analyzed for the 11 eyes with follow-up of greater than 6 months. The average follow-up in this group was 15.5 months. All9 eyes from which silicone oil was removed had follow-up of more than 6 months from primary surgery and an average of 5.9 months (range, 2 to 12 months) follow-up from time of silicone oil removal. Recurrent retinal detachment occurred in 3 of the 9 eyes that underwent silicone oil removal; the redetachment occurred within 2 months of
silicone oil removal. Two of these patients refused further surgery, while in the third patient, the retina could be reattached by outpatient fluid-air exchange. The overall anatomical success at last follow-up visit in these 11 eyes was 81.8%. The visual results at last follow-up in this group are given in Table 4. Six eyes (54.5%) had visual acuity of 10/200 or better. Poor vision with anatomical reattachment was due to long-standing retinal detachment in two cases, while in one case the initial good visual recovery was lost because of uncontrolled secondary glaucoma. Persistent secondary glaucoma was present in three eyes, while band keratopathy occurred in four eyes, although not severe enough to warrant keratoplasty. Figure 6 shows the status of the 17 eyes at last examination with respect to presence or absence of silicone oil in the eye and the anatomical status of the retina.
DISCUSSION Although rare, retinal detachments associated with choroidal coloboma pose a challenge to the vitreoretinal surgeon because such coloboma-related detachments have retinal breaks in the diaphanous retinal tissue that covers the colobomatous area, making closure by conventional scleral buckling almost impossible. Appropriate treatment involves production of chorioretinal adhesion around the periphery of the coloboma, and the retina should remain attached until a firm adhesion is established. In the technique described by Schepens, 2 subretinal fluid is drained to flatten the retina; this is followed by transscleral cryopexy to the periphery of the coloboma and photocoagulation to the posterior borders. However, firm chorioretinal adhesion is not always achieved as fluid may reaccumulate before scarring can occur. Patnaik and Kalsi 3 reported success in one case by using a radial buckle that extended to the disc. However, this involves exposure and buckling of the globe from the ora serrata to disc in its anteroposterior extent, and requires a radial buckle that is slightly broader than the coloboma itself. The width of the coloboma can vary, which limits the usefulness of this 1625
OPHTHALMOLOGY
•
NOVEMBER 1991
technique, because in many cases the radial sponge may not cover the width of the coloboma and adjacent healthy retina. In addition, this procedure involves the sacrifice of two vortex veins. Vitrectomy with endophotocoagulation and silicone oil injection offers an attractive alternative. Vitrectomy allows removal of all traction, allows unequivocal identification of any break(s) in the coloboma, facilitates intraoperative laser treatment, and creates space for a silicone oil tamponade. Lensectomy may sometimes be necessary even in patients with a clear lens, such as those with an iris coloboma, as the equator of the lens can interfere with proper visualization while working on the inferior periphery. Posterior vitreous detachment, which may not be present, can be induced by means of a membrane pick or gentle suction in the vicinity of the disc. Despite the lack of contrast, it is usually possible to identify a break in the coloboma with the help of an endoilluminator and high magnification with the operating microscope. It has been reported that breaks within the coloboma are found usually on the slope of the coloboma. 2 In our experience, breaks were found on the slope but more commonly were located in the center of the colobomatous area. The floor of the coloboma is usually irregular and scalloped, and the margins of the break can sometimes be seen to merge with the boundaries of the scalloped areas. The central breaks were usually oval, and fairly large (more than 1 disc diameter). There may be a demarcation line along part or the entire border of the coloboma, which can be confused with retinopexy scars in patients undergoing resurgery. Fluid-air exchange is relatively simple. Endodrainage is adequately achieved through the break in the colobomatous area. Because the sclera of the coloboma is concave, fluid is displaced into the concavity as air is injected, from which it can be removed with a back-flush needle. There is no need to resort to external drainage. Production of a chorioretinal adhesion around the entire circumference of the coloboma is a very important step and can be achieved by a combination oflaser photocoagulation and cryopexy. If endolaser facilities are available, it is best to complete the treatment during the surgery. When the coloboma involves the disc, it is often difficult to ensure completeness of treatment, since the functional border of the disc is usually left untreated. Theoretically, light treatment in this area may produce chorioretinal adhesion without destroying the nerve fiber layer. In our experience, however, during intraoperative laser treatment, there is less precision in control of the intensity ofburns produced. Therefore, it may be prudent to leave this part of the treatment to the postoperative period. Michels et al 6 have advocated postoperative krypton laser treatment to the peripapillary portion of the coloboma to avoid damage to the nerve fiber layer. Reasonably prolonged tamponade of the inferior retina over the coloboma is required to ensure surgical success. Gases, in the form of C 3F 8-air mixture, have been advocated for this purpose. 4 •6 In our opinion, this is less than satisfactory. The postoperative positioning of the patient would be very difficult for prolonged periods, because it involves the patient remaining prone with the head at a 1626
•
VOLUME 98
•
NUMBER 11
Fig 7. Diagram shows effective tamponading of margin of choroidal coloboma by silicone oil with collection of sub-oil fluid in ectatic area.
lower level. Furthermore, the gas bubble is unlikely to provide adequate tamponade to the entire colobomatous margin for the period required for minimum chorioretinal adhesion to develop, and supplemental gas injection probably would be required. Postoperative laser treatment, if required, is extremely difficult to perform through gas, especially in phakic eyes. In our opinion, silicone oil offers a preferred alternative because of the prolonged internal tamponade. If an eye has been left phakic at the conclusion of the fluid-air exchange, and if there is difficulty in performing endolaser photocoagulation with the -80 diopter lens, visualization can be improved by first exchanging the air with silicone oil, and then proceeding with laser treatment. In our opinion, prolonged tamponade to the colobomatous margin is possible with silicone oil even in the erect position, since it is the ectatic floor of the coloboma and not the margin that is in the most dependent position (Fig 7). Furthermore, the excellent clarity of the media postoperatively permits additional laser treatment should this be needed. The striking feature of the above described technique is the 100% anatomical success at the 2-month follow-up visit. However, the 33% recurrence rate after silicone oil removal is definitely a cause for concern. Several factors could account for recurrent detachment after silicone oil removal. The retinopexy around the coloboma may not be adequate, especially in cases in which the disc is also involved, and where treatment was not performed around the peripapillary area of coloboma. A retinal break posterior to the sclerotomy site could remain asymptomatic but lead to recurrent retinal detachment following removal of silicone oil. In addition, proliferative vitreoretinopathy may be a cause of recurrence. Long-term use of silicone oil can lead to significant complications. 7 Secondary glaucoma is the most important of these, and may persist even after removal of silicone oil or, more rarely, may manifest for the first time subsequent to removal of silicone oil. In patients with coloboma of the disc, it must be noted that disc changes cannot be used as a guide in the management of glaucoma. In a
GOPAL et al
•
RETINAL DETACHMENT WITH CHOROIDAL COLOBOMA
few cases, we observed the silicone oil to protrude into the anterior chamber and contact the posterior corneal surface, despite the iris coloboma. An intact iris diaphragm with a patent 6~o'clock iridectomy is probably more ef~ fective in maintaining the silicone oil bubble in the pupillary plane. Phthisis has been mentioned as a significant complication of conventional surgery in eyes with a choroidal coloboma. These eyes probably react better to vitrectomy procedures as is evident in our series. A postoperative visual acuity of 6/60 (20/200) may be considered good visual recovery in view of the multiplicity of problems associated with this condition, including nystagmus, coloboma of the disc, and often involving the macula. With 10/200 visual acuity, most patients are mobile and a few in our series were able to pursue "academic" activities with low vision aids.
REFERENCES 1. Jesberg DO, Schepens CL. Retinal detachment associated with coloboma of the choroid. Arch Ophthalmol1961; 65:163-73. 2. Schepens CL. Retinal Detachment and Allied Disorders. Vol. 2. Philadelphia: WB Saunders, 1983; 615-7. 3. Patnaik B. Kalsi R. Retinal detachment with coloboma of the choroid. Indian J Ophthalmol 1981; 29:345-9. 4. Wang K, Hilton GF. Retinal detachment associated with coloboma of the choroid. Trans Am Ophthalmol Soc 1985; 83:49-62. 5. Gonvers M. Temporary use of silicone oil in the treatment of special cases of retinal detachment. Ophthalmologica 1983; 187:202-9. 6. Michels RG, Wilkinson CP, Rice TA. Retinal Detachment. St. Louis: CV Mosby 1990; 727. 7. Leaver PK. Complications of Intraocular Silicone Oil. In: Ryan SJ, ed. Retina. Vol. 3: Surgical Retina, St. Louis: CV Mosby 1989; 293.
1627
Abstract: Seventeen eyes with retinal detachment secondary to retinal breaks in the colobomatous area were managed by vitrectomy procedures. Endodrainage was done through the break in the colobomatous area, with simultaneous fluid-air exchange. Endolaser photocoagulation was performed along the colobomatous margin posteriorly, while the anterior portion was treated by transscleral cryopexy. Silicone oil was then exchanged with air. In nine cases, the silicone oil had to be removed for complications such as emulsification, glaucoma, and keratopathy; in three of these eyes, oil removal resulted in recurrent retinal detachment. At the 2-month follow-up visit, there was 100% anatomic success, and 12 eyes (70.6%) recovered visual acuity of 10/200 or better. Of the 11 eyes with follow-up of more than 6 months, in 9 (81.8%) the retina remained reattached at last-follow up, and in 6 eyes (54.5%) a visual acuity of 10/200 or better was obtained. Ophthalmology 1991; 98: 1622-1627
Coloboma of the choroid results from a faulty closure of the embryonal fissure. Although the reported incidence is only 0.14% of the general population, 40% of these individuals will develop a retinal detachment sometime during their lifetime. 1 The colobomatous area consists of a thin layer of hypoplastic retinal tissue. The choroid and retinal pigment epithelium are not developed in this region, and the sclera underlying the colobomatous area is usually thin and ectatic, producing a staphyloma. Retinal detachment caused by retinal breaks outside of the colobomatous area can be managed by conventional scleral buckling techniques. These conventional techniques fail, however, if the detachment is caused by a break in the hypoplastic retinal tissue within the colobomatous area. Furthermore, other difficulties may confound the man-
Originally received: February 5, 1991. Revision accepted: June 19, 1991. 1
Sankara Nethralaya, Medical Research Foundation and Vision Research Foundation, Madras, India. 2 Massachusetts Eye and Ear Infirmary, Boston. Presented in part as a poster at the American Academy of Ophthalmology Annual Meeting, Atlanta, Oct/Nov 1990. Reprint requests to Lingam Gopal, MD, Sankara Nethralaya, 18, College Road, Madras 600 006, India.
1622
agement of coloboma-related retinal detachments. Nystagmus can be an associated feature and may be severe enough to make preoperative evaluation difficult. The lack of contrast in the area of the coloboma can make the identification of retinal breaks difficult. Retinal breaks in the colobomatous area are difficult to seal; retinopexy is not effective because of the absence of choroid and retinal pigment epithelium, and the posterior location of the breaks makes it extremely difficult to buckle the same. Hence, in the management of coloboma-related retinal detachments, attempts have been directed toward isolating the coloboma from the rest of the retina, rather than toward sealing the retinal break. Schepens2 advocated drainage of subretinal fluid, followed by the production of chorioretinal adhesion around the periphery of the coloboma using cryopexy, photocoagulation, or both. Patnaik and Kalsi 3 achieved reattachment of the retina in one case by applying a radial buckle that extended to the optic disc. When both sides of a coloboma are involved in the detachment, Wang and Hilton 4 recommend that two radial buckles be applied along the two edges of the coloboma. Wang and Hilton 4 reviewed the literature and found that of 42 eyes with coloboma-related retinal detachment, 18 (43%) had successful reattachment. In their own series of 20 eyes, success was achieved in 7 eyes (35%). They concluded that vitrectomy with intraocular gas tamponade and creation of chorioretinal adhe-
GOPAL et al
•
RETINAL DETACHMENT WITH CHOROIDAL COLOBOMA
Table 1. Preoperative Visual Acuity Visual Acuity
No. of Eyes
Light perception Hand motions Counting fingers
8 5 3 1
1/60 (3.3/200)
Table 2. Status of Fellow Eye Clinical Status Coloboma of the choroid, attached retina, and useful vision Coloboma of the choroid, spontaneous reattachment of retina, and no vision Coloboma of the choroid and retinal detachment Extreme microphthalmos and no vision Phthisis Normal
No. of Eyes
6 1 6 2 1 1 Fig 1. Coloboma of the choroid with retinal break (arrows) within the coloboma.
sion along the edge of the coloboma is the preferred treatment. Silicone oil also has been used as a temporary tamponade in the management of these cases. 5 In our management of coloboma-related retinal detachments, vitrectomy procedures, silicone oil injection, and endolaser photocoagulation to surround the colobomatous area appeared to improve the surgical results considerably. Herein we describe the technique used and the outcome in 17 eyes treated for coloboma-related retinal detachment.
moscope and slit-lamp biomicroscopy with a 90 diopter lens revealed a break in the colobomatous area in 8 eyes (47%) (Fig 1); in the remainder, a break within the coloboma was identified intraoperatively. A peripheral break outside of the coloboma also was present in 2 eyes ( 11 .7%). Proliferative vitreoretinopathy was present in 3 eyes ( 17.6%): grade C2 and 0 2 in I eye each while 1 eye had only subretinal bands.
MATERIALS AND METHODS .
SURGICAL TECHNIQUE
Seventeen eyes of 16 patients (12 male and 4 female) who were diagnosed as having retinal detachment caused by a break in a coloboma of the choroid are the subject of.this report. The average age at presentation was 19.37 years (range, 6 to 37 years). The duration ofloss of vision at the time of surgery ranged from 5 days to 5 years; 9 patients presented more than than 2 months after the onset of visual disability. The preoperative visual acuity ranged from accurate light projection to 1/60 (3.3/200) (Table 1). Nystagmus was an associated feature in seven patients. Coloboma of the iris was present in 14 eyes (82.35%) and microphthalmos was present in 4 eyes (23.53%). Coloboma of the choroid was associated with coloboma of the disc in 13 eyes (76.47%), and this has special relevance to the management discussed below. The status of the fellow eye is given in Table 2. As is shown, the fellow eye had normal vision in only one patient; in ten patients the fellow eye had no useful vision. Retinal detachments attributable to peripheral breaks not within the coloboma were not included in this study. The retinal detachment was considered to be coloboma-related ifthere was extension to the colobomatous area. Careful preoperative evaluation with the binocular indirect ophthal-
A three-port vitrectomy using the Ocutome System (Cooper Medical Devices, San Leandro, CA) was performed in all cases. Lensectomy was not performed routinely but was necessary for various reasons in 12 eyes (70.6%). Careful excision of the vitreous base was performed in eyes with peripheral proliferative vitreoretinopathy. The colobomatous area was carefully inspected with an endoilluminator and high magnification of the operating microscope, and a retinal break within the coloboma was identified in all these cases. When necessary, membrane peeling always began beyond the margins of the coloboma and was then extended to the colobomatous area to avoid problems related to improper identification of tissues. Once the retina was mobilized, fluid-air exchange was performed. Endodrainage was done through the break in the colobomatous area in 12 eyes; in the other 5 eyes, endodrainage was performed through a preexisting or accidental peripheral retinal break. In one case, a retinotomy created for removal of a subretinal band was used for endodrainage. Fluid-air exchange was performed using a back-flush brush needle, with visualization achieved by use of a -80 diopter lens in phakic eyes and without any contact lens in aphakic eyes. An iridectomy 1623
OPHTHALMOLO GY
•
NOVEMBER 1991
Fig 2. Diagram shows retinal tissue stretched across mouth of coloboma, with subretinal air following fluid-air exchange: fundus view.
Fig 3. Diagram shows retinal tissue stretched across mouth of coloboma, with subretinal air following fluid-air exchange: sagittal section.
at the 6-o'clock meridian was required only in aphakic eyes with no coloboma of the iris. In a few cases, at the end of the fluid-air exchange we found the diaphanous retinal tissue stretched taut over the coloboma (Figs 2, 3). This situation is analogous to that created by localized unrelieved traction with shortening of the retina encountered in routine cases of proliferative vitreoretinopathy and may result in subretinal air. This situation could be relieved by relaxing incisions that extended radially toward the edge of the coloboma (Figs 4, 5). 1624
•
VOLUME 98
•
NUMBER 11
Fig 4. Diagram shows retinal flaps adherent to ectatic sclera following relaxing radial retinotomies: fundus view.
Fig 5. Diagram shows retinal flaps adherent to ectatic sclera following relaxing radial retinotomies: sagittal section.
Once the retina was settled, endolaser photocoagulation was applied around the colobomatous area and any peripheral retinal break. Endophotocoagulation was performed in two or three rows of overlapping burns along the entire margin of the coloboma. Where the coloboma involved the optic disc, the functional border of the disc was left untreated. The extreme anterior portion of the colobomatous margin was treated with transscleral cryopexy. When the coloboma did not extend to the ora anteriorly, it could be surrounded completely by endolaser
GOPAL et al
•
RETINAL DETACHMENT WITH CHOROIDAL COLOBOMA
Table 3. Visual Results at 2-month Follow-up
~lATUS
Visual Acuity
No. of Eyes
Percentage
6/36 (20/120) 6/60 (20/200) 3/60 (1 0/200) 1/60 (3.3/200)
1
5.9 29.4
Hand motions
5 6 3
2
.vvv .vv
,,./
Visual Acuity
No. of Eyes
Percentage
6/18 (20/60) 6/24 (20/80) 3/60 (1 0/200)
1 1
9.1 9.1 36.4 45.5
Hand motions
4
5
~
m. Vv ~
Table 4. Final Visual Results*
~
1·v v
35.3
17.6 11.8
OF RETlNA AT LAST EXAMINATIONU
/
..Vv
2 •.Vv ._v v ,_v v
·~~
,/v'-----/
~ EYESWffliSII.ICXJNEOI..
L
/
/
IIETINAATTICHED
/1/ET/NADETICHED
EYf8 WfrH SIIXXlHE Ol.lfEMCNED
Fig 6. Bar graph depicts anatomical status of retina at final examination.
' Of the 11 eyes that had follow-up of 6 months or more.
treatment alone. Silicone oil ( 1000 centistokes) was exchanged with air. In aphakic eyes, the anterior chamber was reformed with saline. Sclerotomies were closed after ascertaining that the intraocular pressure was normal. To the extent possible, patients were encouraged to remain prone in the early postoperative period. Additional laser treatment to any areas of colobomatous margin that had been inadvertently missed was performed postoperatively. In two cases, repeat surgery had to be performed for localized, recurrent retinal detachment. This second procedure consisted of membrane peeling under silicone oil, reattachment of retina by endodrainage under oil, and additional endolaser treatment. Silicone oil was removed from nine eyes; the average interval between silicone oil injection and removal was 8.44 months (range, 4 to 16 months). In general, oil was removed only if complications such as emulsification occurred. In phakic eyes, silicone oil removal, when done, was combined with lensectomy, since significant cataract had developed by the time of oil removal.
RESULTS The average follow-up was 10.76 months (range, 2 to 28 months); 11 eyes (64.7%) were followed for more than 6 months. At the 2-month follow-up visit, all 17 eyes had anatomical reattachment. Visual acuity at 2 months after surgery is given in Table 3. Twelve eyes (70.6%) had visual acuity of 3/60 (10/200) or better. Data from last follow-up also were analyzed for the 11 eyes with follow-up of greater than 6 months. The average follow-up in this group was 15.5 months. All9 eyes from which silicone oil was removed had follow-up of more than 6 months from primary surgery and an average of 5.9 months (range, 2 to 12 months) follow-up from time of silicone oil removal. Recurrent retinal detachment occurred in 3 of the 9 eyes that underwent silicone oil removal; the redetachment occurred within 2 months of
silicone oil removal. Two of these patients refused further surgery, while in the third patient, the retina could be reattached by outpatient fluid-air exchange. The overall anatomical success at last follow-up visit in these 11 eyes was 81.8%. The visual results at last follow-up in this group are given in Table 4. Six eyes (54.5%) had visual acuity of 10/200 or better. Poor vision with anatomical reattachment was due to long-standing retinal detachment in two cases, while in one case the initial good visual recovery was lost because of uncontrolled secondary glaucoma. Persistent secondary glaucoma was present in three eyes, while band keratopathy occurred in four eyes, although not severe enough to warrant keratoplasty. Figure 6 shows the status of the 17 eyes at last examination with respect to presence or absence of silicone oil in the eye and the anatomical status of the retina.
DISCUSSION Although rare, retinal detachments associated with choroidal coloboma pose a challenge to the vitreoretinal surgeon because such coloboma-related detachments have retinal breaks in the diaphanous retinal tissue that covers the colobomatous area, making closure by conventional scleral buckling almost impossible. Appropriate treatment involves production of chorioretinal adhesion around the periphery of the coloboma, and the retina should remain attached until a firm adhesion is established. In the technique described by Schepens, 2 subretinal fluid is drained to flatten the retina; this is followed by transscleral cryopexy to the periphery of the coloboma and photocoagulation to the posterior borders. However, firm chorioretinal adhesion is not always achieved as fluid may reaccumulate before scarring can occur. Patnaik and Kalsi 3 reported success in one case by using a radial buckle that extended to the disc. However, this involves exposure and buckling of the globe from the ora serrata to disc in its anteroposterior extent, and requires a radial buckle that is slightly broader than the coloboma itself. The width of the coloboma can vary, which limits the usefulness of this 1625
OPHTHALMOLOGY
•
NOVEMBER 1991
technique, because in many cases the radial sponge may not cover the width of the coloboma and adjacent healthy retina. In addition, this procedure involves the sacrifice of two vortex veins. Vitrectomy with endophotocoagulation and silicone oil injection offers an attractive alternative. Vitrectomy allows removal of all traction, allows unequivocal identification of any break(s) in the coloboma, facilitates intraoperative laser treatment, and creates space for a silicone oil tamponade. Lensectomy may sometimes be necessary even in patients with a clear lens, such as those with an iris coloboma, as the equator of the lens can interfere with proper visualization while working on the inferior periphery. Posterior vitreous detachment, which may not be present, can be induced by means of a membrane pick or gentle suction in the vicinity of the disc. Despite the lack of contrast, it is usually possible to identify a break in the coloboma with the help of an endoilluminator and high magnification with the operating microscope. It has been reported that breaks within the coloboma are found usually on the slope of the coloboma. 2 In our experience, breaks were found on the slope but more commonly were located in the center of the colobomatous area. The floor of the coloboma is usually irregular and scalloped, and the margins of the break can sometimes be seen to merge with the boundaries of the scalloped areas. The central breaks were usually oval, and fairly large (more than 1 disc diameter). There may be a demarcation line along part or the entire border of the coloboma, which can be confused with retinopexy scars in patients undergoing resurgery. Fluid-air exchange is relatively simple. Endodrainage is adequately achieved through the break in the colobomatous area. Because the sclera of the coloboma is concave, fluid is displaced into the concavity as air is injected, from which it can be removed with a back-flush needle. There is no need to resort to external drainage. Production of a chorioretinal adhesion around the entire circumference of the coloboma is a very important step and can be achieved by a combination oflaser photocoagulation and cryopexy. If endolaser facilities are available, it is best to complete the treatment during the surgery. When the coloboma involves the disc, it is often difficult to ensure completeness of treatment, since the functional border of the disc is usually left untreated. Theoretically, light treatment in this area may produce chorioretinal adhesion without destroying the nerve fiber layer. In our experience, however, during intraoperative laser treatment, there is less precision in control of the intensity ofburns produced. Therefore, it may be prudent to leave this part of the treatment to the postoperative period. Michels et al 6 have advocated postoperative krypton laser treatment to the peripapillary portion of the coloboma to avoid damage to the nerve fiber layer. Reasonably prolonged tamponade of the inferior retina over the coloboma is required to ensure surgical success. Gases, in the form of C 3F 8-air mixture, have been advocated for this purpose. 4 •6 In our opinion, this is less than satisfactory. The postoperative positioning of the patient would be very difficult for prolonged periods, because it involves the patient remaining prone with the head at a 1626
•
VOLUME 98
•
NUMBER 11
Fig 7. Diagram shows effective tamponading of margin of choroidal coloboma by silicone oil with collection of sub-oil fluid in ectatic area.
lower level. Furthermore, the gas bubble is unlikely to provide adequate tamponade to the entire colobomatous margin for the period required for minimum chorioretinal adhesion to develop, and supplemental gas injection probably would be required. Postoperative laser treatment, if required, is extremely difficult to perform through gas, especially in phakic eyes. In our opinion, silicone oil offers a preferred alternative because of the prolonged internal tamponade. If an eye has been left phakic at the conclusion of the fluid-air exchange, and if there is difficulty in performing endolaser photocoagulation with the -80 diopter lens, visualization can be improved by first exchanging the air with silicone oil, and then proceeding with laser treatment. In our opinion, prolonged tamponade to the colobomatous margin is possible with silicone oil even in the erect position, since it is the ectatic floor of the coloboma and not the margin that is in the most dependent position (Fig 7). Furthermore, the excellent clarity of the media postoperatively permits additional laser treatment should this be needed. The striking feature of the above described technique is the 100% anatomical success at the 2-month follow-up visit. However, the 33% recurrence rate after silicone oil removal is definitely a cause for concern. Several factors could account for recurrent detachment after silicone oil removal. The retinopexy around the coloboma may not be adequate, especially in cases in which the disc is also involved, and where treatment was not performed around the peripapillary area of coloboma. A retinal break posterior to the sclerotomy site could remain asymptomatic but lead to recurrent retinal detachment following removal of silicone oil. In addition, proliferative vitreoretinopathy may be a cause of recurrence. Long-term use of silicone oil can lead to significant complications. 7 Secondary glaucoma is the most important of these, and may persist even after removal of silicone oil or, more rarely, may manifest for the first time subsequent to removal of silicone oil. In patients with coloboma of the disc, it must be noted that disc changes cannot be used as a guide in the management of glaucoma. In a
GOPAL et al
•
RETINAL DETACHMENT WITH CHOROIDAL COLOBOMA
few cases, we observed the silicone oil to protrude into the anterior chamber and contact the posterior corneal surface, despite the iris coloboma. An intact iris diaphragm with a patent 6~o'clock iridectomy is probably more ef~ fective in maintaining the silicone oil bubble in the pupillary plane. Phthisis has been mentioned as a significant complication of conventional surgery in eyes with a choroidal coloboma. These eyes probably react better to vitrectomy procedures as is evident in our series. A postoperative visual acuity of 6/60 (20/200) may be considered good visual recovery in view of the multiplicity of problems associated with this condition, including nystagmus, coloboma of the disc, and often involving the macula. With 10/200 visual acuity, most patients are mobile and a few in our series were able to pursue "academic" activities with low vision aids.
REFERENCES 1. Jesberg DO, Schepens CL. Retinal detachment associated with coloboma of the choroid. Arch Ophthalmol1961; 65:163-73. 2. Schepens CL. Retinal Detachment and Allied Disorders. Vol. 2. Philadelphia: WB Saunders, 1983; 615-7. 3. Patnaik B. Kalsi R. Retinal detachment with coloboma of the choroid. Indian J Ophthalmol 1981; 29:345-9. 4. Wang K, Hilton GF. Retinal detachment associated with coloboma of the choroid. Trans Am Ophthalmol Soc 1985; 83:49-62. 5. Gonvers M. Temporary use of silicone oil in the treatment of special cases of retinal detachment. Ophthalmologica 1983; 187:202-9. 6. Michels RG, Wilkinson CP, Rice TA. Retinal Detachment. St. Louis: CV Mosby 1990; 727. 7. Leaver PK. Complications of Intraocular Silicone Oil. In: Ryan SJ, ed. Retina. Vol. 3: Surgical Retina, St. Louis: CV Mosby 1989; 293.
1627
