Retinoschisis in Coats Disease: Clinical Picture, Therapeutic Considerations, and Management Outcomes

Abstract

Purpose:

Retinoschisis in inflammatory and exudative retinopathy is a known but rare entity. We describe the presentation and clinical and surgical outcomes of a series of patients with retinoschisis in the setting of Coats disease.

Methods:

This retrospective case series of patients with Coats disease recorded visual acuity and the number and type of treatments (eg, angiography-guided photocoagulation, intravitreal bevacizumab, sub-Tenon triamcinolone, or vitreoretinal surgery).

Results:

Eighteen of 133 patients with Coats disease were diagnosed with retinoschisis by color imaging. All (100%) of the areas of schisis or macrocyst were associated with dense subretinal exudation, whereas only 7 (38.8%) were associated with exudative retinal detachment. Initial best-corrected visual acuity in the schisis cohort was 1.14 ± 1.19 (Snellen equivalent, 20/276) vs 1.42 ± 1.40 (Snellen, 20/526) in those without (P = .56). Final best-corrected visual acuity was 1.76 ± 1.37 (Snellen, 20/1150) and 1.45 ± 1.44 (Snellen, 20/563), respectively (P = .43). The mean number of treatments in individuals with schisis was 4.8 ± 2.9 for angiography-guided photocoagulation, 4.5 ± 2.9 for intravitreal bevacizumab, and 1.7±1.2 for sub-Tenon triamcinolone. Intraocular surgery was required in 39% (7 of 18) of patients with schisis vs 22% (25 of 115) for patients without schisis (P = .14). One eye required enucleation, and histopathology of the schisis pocket was obtained.

Conclusions:

The pathophysiology of retinoschisis in Coats disease is not completely understood. We propose that telangiectatic leakage, hypoxia, and toxicity to the retinal pigment epithelial cells play a role in cavity development, and therefore may represent a clinical feature of advanced or uncontrolled disease.

Introduction

It is believed that patients with Coats disease have an inherent dysregulation in the retina-blood barrier that predisposes them to inflammation and the release of various growth factors, including vascular endothelial growth factor (VEGF). ¹ These mediators increase the permeability of vessels and can ultimately manifest as extensive exudation, exudative retinal detachment, and progressive ischemia of the retina.

The mainstay of treatment in early and intermediate disease focuses on ablation of the abnormal retinal vasculature via photocoagulation in an attempt to preserve vision and prevent progression to retinal detachment. ^{2 -6} Adjunctive medical therapy, such as intravitreal anti-VEGF, steroids, or a combination of the 2, has been associated with improved outcomes. ^2,3 Pars plana vitrectomy (PPV), scleral buckle (SB), and enucleation are surgical procedures that may be necessary in cases with advanced disease. ^4,5

As ischemia advances, the vitreoretinal interface and intracellular junctions are affected and may result in the splitting of the neurosensory retina, also referred to as “retinoschisis” or “retinal macrocysts.” The terms retinal macrocyst and retinoschisis will be used interchangeably throughout this article because they refer to the same clinical picture. This is a rare clinical finding in Coats disease with limited description in the literature. ^{6 -9} Although the pathophysiology is currently unknown, its presence may represent disease activity that could ultimately guide clinical and surgical management.

Here, we outline the clinical presentation, therapeutic management, and clinical outcomes in patients with Coats disease and acquired retinoschisis.

Methods

A retrospective case series of patients with a diagnosis of Coats disease and retinoschisis was performed at the Bascom Palmer Eye Institute from January 2014 to May 2018. Patients with a diagnosis of Coats disease with color images (Optos Daytona, RetCam [Natus Medical Inc]) were included in the study. Retinoschisis was confirmed by the treating physician’s chart medical-record report, color pictures demonstrating schisis cavities, or echography confirming hypoechogenic round images consistent with schisis cavities. Fluorescein angiography (FA) was requested routinely in all patients to confirm Coats disease, and it might have demonstrated the schisis cavities; however, it was not diagnostic of retinal macrocysts. A minimum follow-up of 5 months was required for inclusion.

Demographic information was recorded: sex, age at presentation, initial and final best-corrected visual acuity (BCVA) (reported as logMAR), presence or absence of anterior segment neovascularization, number of angiography-guided photocoagulation (AGP) sessions, number of intravitreal bevacizumab injections, number of sub-Tenon triamcinolone (STT) injections, and number of surgical procedures (PPV and SB). Statistical analysis was performed with IBM SPSS Statistics 24. Fisher exact test was used for categorical data, and t test to compare groups. A P value of less than .05 was considered significant.

Results

A total of 138 patients were found with a diagnosis of Coats disease from January 2014 to May 2018. Only 133 patients had color pictures and were included in the analysis. Eighteen patients had retinoschisis (14%) diagnosed by color pictures and echography. All (100%) areas of schisis and macrocyst were associated with dense subretinal exudation, whereas only 7 (38.8%) were associated with exudative retinal detachment. Of the total population, 100 (75%) were male and 33 (25%) were female. Of the schisis group, 13 (72%) were male and 5 (28%) were female vs 87 (76%) and 28 (24%), respectively, in individuals without schisis (P = .77). The mean age of the entire population was 16 ± 1.5 years: 6.5 ± 5.0 years in the schisis group, and 17.8 ± 15 years in those without (P ≤ .001). The mean initial BCVA from the complete series was 1.40 ± 1.38 (Snellen, 20/502). Mean initial BCVA in patients with schisis was 1.14 ± 1.19 (Snellen, 20/276) and 1.42 ± 1.40 (Snellen, 20/526) in those without (P = .56). Mean final VA was 1.50 ± 1.42 (Snellen, 20/632) for the total series. The mean final BCVA in those with and without schisis was 1.76 ± 1.37 (Snellen, 20/1150) and 1.45 ± 1.44 (Snellen, 20/563), respectively (P = .43) (Table 1).

Table 1.

Patient Demographics.

	Without schisis	Schisis cohort	P value
Sex
Male, No.	87	13
Female, No.	28	5
Mean age, y	17.8 ± 15	6.5 ± 1.5	< .001
Initial BCVA, logMAR	1.42 ± 1.40	1.14 ± 1.19	.56

Abbreviations: BCVA, best-corrected visual acuity.

On average, our patients were followed for 28 months with multiple examinations under anesthesia for clinical evaluation (eg, indirect ophthalmoscopy, color imaging, FA, B-scan ultrasonography, and/or optical coherence tomography imaging) and treatment (eg, laser, intravitreal steroids, intravitreal anti-VEGF, or PPV with silicone oil tamponade). Four patients had documentation of schisis at initial diagnosis (Figure 1), one of whom developed additional cysts during treatment (Figure 2), whereas the rest were documented to have schisis after initial examination (Figure 3). On FA, the schisis cavities appeared hyporeflective with surrounding hyperfluorescence due to pooling and vascular filling (Figure 4). The location of the schisis cavities with respect to other retinal pathology varied—they localized at or adjacent to retinal detachment, within areas of exudation, and/or within networks of abnormal vasculature.

Figure 1.

A 4-and-a-half-year-old boy with Coats disease presented out of concern for retinal detachment. (A) Examination under anesthesia revealed a massive retinal macrocyst in the inferior retina with exudation surrounding the borders of the cyst. There were areas of telangiectasias and exudation infiltrating the fovea. (B) Optical coherence tomography revealed splitting of the neurosensory retina and intraretinal exudation. (C) After 3 rounds of angiography-guided photocoagulation and intravitreal bevacizumab over 15 months, the exudation largely quiesced and the retinal macrocyst collapsed. His final best-corrected visual acuity was 20/400.

Figure 2.

An 11-year-old girl presented for a second opinion regarding an exudative retinal detachment after receiving angiography-guided photocoagulation at an outside facility. Her best-corrected visual acuity was 20/80. (A) A large pocket of retinoschisis was noted nasal to the optic disc. (B-D) After 2 examinations under anesthesia, with 2 rounds of angiography-guided photocoagulation along with injections of sub-Tenon triamcinolone and intravitreal bevacizumab, the patient developed additional pockets of retinoschisis in the inferior (vertical arrows) and superior (horizontal arrows) poles. A pars plana vitrectomy with endolaser, sub-Tenon triamcinolone, intravitreal bevacizumab, and silicone oil was ultimately performed. (E) There was collapse of the nasal and superior retinal macrocysts. Her final best-corrected visual acuity was 20/70.

Figure 3.

An 11-year-old boy presented for an evaluation of peripheral exudation. (A) His best-corrected visual acuity was 20/25 in the affected eye. On examination under anesthesia, (B) fundus photography and (C) B-scan ultrasonography showed he had a shallow inferior retinal detachment, and he was treated with angiography-guided photocoagulation (AGP) and intravitreal bevacizumab (IVB). Four months later, he developed a massive microcyst in the inferior retina. Additional AGP, sub-Tenon triamcinolone, and IVB were given. (D) Disease control, as determined by progression of exudation and vascular leakage on fluorescein angiogram, was achieved after a third session of AGP, IVB, and sub-Tenon triamcinolone. The vertical arrow depicts a retinal scar. (E) Collapse of the retinal macrocyst was noted at 2 years with confirmation via optical coherence tomography. The vertical arrow depicts a retinal scar. His final best-corrected visual acuity was 20/20.

Figure 4.

On fluorescein angiogram, schisis cavities appeared hyporeflective with surrounding early hypofluorescence, hyperfluorescence, or both due to pooling and vascular filling. There is typically telangiectatic vasculature over the cavities with variable amounts of hyperfluorescent pooling. There are varying degrees of capillary dropout and diffuse small-vessel leakage throughout the fundus. The location of the schisis cavities with respect to other retinal pathology varied—patients had localized cavities (A and B) at or adjacent to retinal detachment, (C and D) within areas of exudation, and/or (E and F) within networks of abnormal vasculature.

All patients were treated with multiple sessions of AGP, intravitreal bevacizumab (IVB), and STT. The mean number of treatments for patients with schisis was 4.8 ± 2.9 for AGP sessions, 4.5 ± 2.9 for IVB, and 1.7 ± 1.2 for STT. Owing to inconsistent documentation of treatment among the population without schisis, quantitative treatment data were omitted for this group. The need for intraocular surgery (PPV, SB, and/or silicone oil) for individuals with schisis was 39% (7 of 18) vs 22% (25 of 115) for patients without schisis (P = .14) (Table 2).

Table 2.

Treatment Interventions by Group.

	Without schisis	Schisis cohort	P value
Final BCVA, logMAR	1.45 ± 1.44	1.76 ± 1.37	.43
AGP, No.		4.8 ± 2.9
Intravitreal bevacizumab, No.		4.5 ± 2.9
Sub-Tenon triamcinolone, No.		1.7 ± 1.2
Pars plana vitrectomy, No.	25	7	.14

Abbreviations: AGP, angiography-guided photocoagulation; BCVA, best-corrected visual acuity.

Four patients had either resolution (2 patients) or improvement (2 patients) in the size of the schisis cavity after treatment. Three patients did not have further documentation of their schisis cavities. One patient presented with macular involvement of the schisis cavity and had subsequent improvement in vision after the schisis cavity collapsed. One eye required enucleation, after which gross images and histopathology were obtained to characterize the schisis cavity (Figure 5).

Figure 5.

(A) Macroscopic and (B) microscopic pathology specimen from an enucleation caused by total retinal detachment and suspicion of retinoblastoma. Coats disease was diagnosed after histopathological study in which the splitting of the retina was observed and a massive subretinal exudation with telangiectatic vasculature was found.

Conclusions

Retinoschisis refers to the splitting of the layers in the neurosensory retina. Subtypes include hereditary, degenerative, tractional, and exudative. In the degenerative subtype, it is thought that the retina splits because of preexisting cystic lesions that fuse together as the neuroretinal and glial elements within the lesions degenerate. ¹⁰ Theories pertaining to the etiology of schisis involving other retinal diseases include Müller cell dysfunction and genetic mutations in the retinoschisin (RS1) gene. Cell dysregulation is thought to localize to the adhesive protein properties in the photoreceptor and bipolar cells. ¹¹ Retinal macrocysts are also a hallmark of chronic retinal detachment of other etiologies. Shortly after a retinal detachment develops, there are alterations in the proliferation of Müller cells and astrocytes that can lead to alterations in the mechanical properties of the detached retina. This can split the neurosensory retina and even contract the retina so that a progressive funnel detachment is observed. ¹²

It is certainly possible that the macrocysts in Coats are related to chronic detachment. However, in our population of patients with Coats disease, macrocysts were more often associated with massive subretinal exudation than with retinal detachment. It is also notable that the schisis cavities typically fall in areas of prominent vascular abnormalities. Therefore, it is thought that leakage, hypoxia, and toxicity to the retinal pigment epithelial cells likely play a role in cavity development. ¹³

Our clinical series follows 18 patients with clinical and photographic diagnoses of Coats disease with acquired retinoschisis. Our study supports the idea that treatment of the abnormal vasculature may lead to resolution of the schisis, and therefore vascular leakage and exudation may play a role in its development. All of the macrocysts in our study revealed telangiectatic vasculature and exudation at the border of the schisis. In this study, acquired schisis cavities were not associated with worse visual outcomes or likelihood of requiring surgery, but to our knowledge, this is the largest study to characterize the development of acquired schisis in the clinical and surgical setting.

Laser photocoagulation of telangiectatic retinal vessels and nonperfused retina is the mainstay of therapy for active Coats disease. ¹⁴ AGP is used in many institutions. ^9,15 This is thought to both promote resolution of exudation and reduce the risk of neovascularization. Adjuvant intravitreal anti-VEGF therapy and steroid injections have also been shown to hasten the progress of active disease. ¹⁶ For far advanced disease, PPV with silicone oil tamponade has been shown to be effective in maintaining or achieving good visual outcomes. ^{13,17 -19}

Retinoschisis or macrocyst associated with Coats disease has been described earlier. One patient in the cohort was reported previously by our group, ¹² and Berinstein et al described 2 cases of peripheral and central schisis but did not comment on the interval change in the clinical features during its management. ⁶ Munira and colleagues described a single case of a 15-year-old boy who developed a retinal macrocyst that subsequently resolved with laser photocoagulation and IVB. ²⁰

Although these are the only accounts of acquired retinoschisis in Coats disease, there have been various other case reports with different retinal diseases with retinal telangiectatic pathology that have developed retinoschisis. For instance, 2 other case reports describe patients with either unilateral or bilateral progressive retinal telangiectasias and secondary exudative retinal detachment with retinoschisis. ^21,22 In both reports, patients were treated with cryotherapy and had resolution of their retinal detachments and schisis cavities at a median follow-up time of 4 years. Shields et al reported an incidence of 11% of retinal cysts in a series of 150 patients with Coats disease. ⁸ They did not mention specific characteristics of these patients.

The clinical features of retinoschisis are variable with respect to their location (peripheral vs central), their size, and whether they are associated with retinal exudation vascular abnormalities or retinal detachments. Interestingly, prior literature suggests that exudates within the schisis cavity is a rarely reported feature in all types of retinoschisis. ^23,24

Although not validated in this study, our experience suggests that the presence of a schisis cavity may predict active disease and the need for more aggressive treatment (AGP, IVB, STT, and/or PPV with SB). It is certainly possible that the cavities themselves are a function of chronic detachment. However, treatment and resolution of the macrocyst with photocoagulation, local steroids, and intravitreal anti-VEGF therapy suggest there is in fact underlying activity. Further, the schisis pockets were more often associated with subretinal exudation than with detachment.

Limitations in this study include the relatively small sample size, the uneven follow-up between interval visits, as well as the heterogeneity of disease response, disease location (macula involving vs macula sparing), and treatment plan. The reality that many of our patients were of amblyopic age also limits our ability to attribute final BCVA to disease resolution and activity vs amblyopia; recall that our schisis group was significantly younger than the nonschisis group. However, to our knowledge, this is the first time that acquired retinoschisis cases are united to discuss differences in clinical presentation, management, and outcomes.

Little is known about the pathophysiology of acquired schisis in Coats disease, but our study supports the idea that vascular leakage and exudation disrupt the adhesive properties of the retinal layers and can be reversed with thoughtful management. Given the relatively poor visual outcomes associated with advanced Coats disease, early recognition and treatment may offer vision or globe salvage in select cases. Prospective analysis is required to truly characterize the clinical and surgical outcomes in patients with Coats disease who develop retinoschisis.

In summary, 13.5% of our population with Coats disease had retinoschisis. Treatment was based on AGP, IVB, or STT, with or without surgery. There was no significant difference in BCVA or need for surgery when compared with the nonretinoschisis cases. Improvement of the schisis cavities was observed throughout the follow-up. This is the largest case series reporting the clinical picture and outcome after multiple modal management. This is also the first report of a histopathologic characterization of microcyst related to Coats disease. There is no standard of care for this complication because the incidence is low; however, our series suggests that schisis cavities represent a treatable disease activity and aggressive therapy can lead to improvement over time.