Clinical presentations of X-linked retinoschisis in Taiwanese patients confirmed with genetic sequencing

Patients

Twenty-three male patients from 16 families were recruited from our clinics. They were all of unexceptional health except eye disease with widely varying age. The diagnosis of XLRS was made clinically and based on the presence of macular schisis (with or without peripheral retina schisis changes) and electrophysiological findings, and then the diagnosis was confirmed using genetic screening for the RS1 gene. This study was conducted with the approval of the Institutional Review Board (protocol No. 99–1063B) of Chang Gung Memorial Hospital, and the study followed the protocols of the Declaration of Helsinki. Informed consent was obtained from all participants after a full explanation of the nature and possible consequences of the study was provided.

Ophthalmic examinations

Patients underwent ophthalmic examinations including best-corrected visual acuity (BCVA), fundus photography, FAF imaging, spectral domain OCT (SD-OCT), and full-field ERG. Fundus photography was performed using a digital fundus camera (Nonmyd-α-DIII, Kowa Optimed, Chou-ku, Japan).

Fundus autofluorescence imaging and fluorescence angiography

FAF examinations were conducted using a confocal scanning laser ophthalmoscope (Heidelberg Retina Angiograph, Heidelberg Engineering, Heidelberg, Germany). Automated eye tracking and image alignment allowed for better resolution. One patient (patient N;II-1) received the fluorescent angiography (FA) examination using the Heidelberg Retina Angiograph.

Optical coherence tomography

SD-OCT images were obtained using either a Spectralis HRA-OCT (Heidelberg, Germany) or RTVue (Optovue, Inc., Fremont, CA). In cases with macular edema treated with topical dorzolamide, the OCT images were compared for morphological changes using the same machines.

Electroretinography

Full-field ERGs were performed using Burian-Allen contact lens electrodes as described previously [9]. Stimuli were generated with a commercial recording unit (Utas-E3000; LKC Technologies, Inc., Gaithersburg, MD) using standards from the International Society for Clinical Electrophysiology of Vision (ISCEV) [10], which included dark-adapted (DA) responses to flash intensity of 0.01 and 3.0 cd.s.m⁻² (DA 0.01 and DA 3.0) and light-adapted (LA) response to a flash intensity of 3.0 cd.s.m⁻² (LA 3.0) and 30 Hz flicker.

Genetic examinations

The coding region of RS1 (exons 1–6) was sequenced for disease-causing mutations in all participants. DNA was extracted from peripheral blood using the QIAamp DNA Mini Kit (Qiagen Inc., Valencia, CA) as described elsewhere [9,11]. Extracted genomic DNA was amplified with polymerase chain reaction (PCR) with a MyCycler Thermal Cycler (Bio-Rad, Hercules, CA) using primer sequences as described in Table 1 [8,12]. The cycling conditions were an initial 10-min incubation at 95 °C, 35 cycles of 30 s at 95 °C, 30 s at 55 °C, and 1 min at 72 °C, and a final step of 7 min at 72 °C. The coding regions of the RS1 gene that encode retinoschisin were directly sequenced after purification.

Clinical examinations

Table 2 and the Figure 1 and Figure 2 show the demographic, clinical, and genetic characteristics of the 23 patients. The median age at diagnosis was 18 years (range 4–58 years). The BCVA ranged from no light perception to 20/25. Eighteen of the 23 patients (78%) presented with decreased visual acuity, three patients with leukocoria, one patient with strabismus, and one with photophobia. Of the three patients (13%) with leukocoria, one was due to congenital cataract (M,II-1), one to Coats’-like exudation of the retina (M,II-3, Figure 2A), and one to total exudative retinal detachment with neovascular glaucoma, which was initially diagnosed as Coats’ disease (G,III-6).

All patients had fundus abnormalities on indirect ophthalmoscopy. Fourteen of the 23 patients (61%) had a typical spoke-wheel pattern in the macula (Figure 2B), one patient had subretinal fibrosis (Figure 2G), and one patient had macula atrophy (Figure 2F) mimicking maculopathy. Vitreous hemorrhages were present in four patients (17%; B,II-1; D,II-1; I,II-1; M,II-1, mean age: 9 years), and all four patients had peripheral retinoschisis. Peripheral retinoschisis was present in ten patients (43%; Figure 2C,D,G,H,K), and peripheral pigmentary changes were observed in five patients (22%; Figure 2D,H,J,K). A golden-yellow reflex was seen in patient K,II-1 (Figure 2I). Two patients had white dots in the posterior fundus mimicking fundus albipunctatus (Figure 2B,K), and one patient had sheathed vessels (Figure 2D).

Fundus autofluorescence imaging and fluorescence angiography

FAF imaging was performed in six patients, and fluorescence angiography in one. One patient (Figure 3A) had a spoke-wheel pattern of hyper- and hypo-FAF, two patients had an irregularly shaped area of hyper- and hypo-FAF (Figure 3B), and three patients had concentric areas of hyper-FAF (Figure 3C,D). Patient N,II-1 was a referral case and presented with non-FA leakage cystoid macula edema (Figure 3D), which suggested XLRS as a possible diagnosis.

Optical coherence tomography

Thirty-eight eyes from 23 patients underwent SD-OCT examinations. Macular schisis was observed in 31 eyes (82%), and foveal atrophy was observed in seven eyes (18%). Bilateral foveal atrophy was seen in three older patients (mean age: 49 years; Figure 4B). Two patients had macular schisis in one eye and foveal atrophy in the other eye (H,II-1; N,II-1, Figure 4C). Two patients had schisis and a lamellar hole in one eye and macular schisis with no hole in the other eye (J,II-2; K,II-1, Figure 4E,F).

Seven patients received topical dorzolamide for macular schisis, and four (30.8%) of 13 eyes had a decreased central foveal thickness (Figure 5A–D). Patient J,II-5 (OD, Figure 5A, OS, Figure 5B) showed significant resolution of the schisis/cystoid macula edema 7 months after treatment with topical dorzolamide. Patient K, II-1 showed a positive response to treatment in his right eye (Figure 5C), while the other eye (lamellar hole, Figure 4F) showed no response. Patient M,II-1 had reduced macula thickness 3 months after treatment (Figure 5D).

Electroretinography

In total, 12 patients underwent ERGs (Figure 6, Figure 7). The DA 0.01 ERG was reduced in all patients, and the DA 3.0 ERG showed a reduced b/a ratio in all patients except patient L,II-1 (Figure 6). Seven patients (C,II-1; F,II-1; G,III-3; K,II-1; N,II-1; P,II-1; P,II-2) had a b/a ratio <1 on the DA 3.0 ERG, which achieved the definition of an electronegative ERG, while five patients had a b/a ratio >1. Patient L,II-1 showed a normal b/a ratio of >1, and the amplitudes of the a- and b-waves decreased.

All patients showed a delay in implicit time and a decrease in the amplitude of the b-wave in the LA 3.0 ERG. A delayed LA 30 Hz flicker peak time and decreased amplitude were observed in all patients (Figure 6, Figure 7).

Genetic studies

All patients had genetic mutations in the RS1 gene, and the mutations were localized in exons 4, 5, and 6 encoding the discoidin domain. One deletion-caused frameshift mutation was found in exon 5 (Family D), one nonsense mutation in exon 5 (Families L and E), one nonsense mutation in exon 6 (Families F, G, M, O, and P), and the rest were missense mutations (Table 2, Appendix 1). Among these nine mutations, three were known, and six were novel (Table 2, Appendix 1). The brother of patient J;II-2 was born using the technique of preimplantation genetic diagnosis. Multiple species sequence alignment of the portion of the discoidin domain showed that the mutation sites were highly conserved (see Figure 8).

Clinical features

Although a spoke wheel–like appearance in the macula is the most typical characteristic feature of XLRS, this feature may be obscured when there is a vitreous hemorrhage in a young patient or may be absent in older patients. Therefore, clinicians should suspect XLRS as one of the differential diagnoses in young boys who present with unsatisfactory BCVA with or without vitreous hemorrhage or leukocoria, because a sudden visual decrease in XLRS rarely occurs and is often associated with complications such as vitreous hemorrhage or retinal detachment. Diagnosis of XLRS in older patients or patients without macula schisis is also challenging because it may present with various fundus appearances, such as peripheral retinal pigmentation and macular atrophy.

Fundus autofluorescence imaging and fluorescence angiography

FAF is used for the differential diagnosis of retinal degeneration or dystrophy. The typical FAF finding in patients with XLRS is a spoke-wheel pattern of hyper- and hypo-FAF in the macular area (Figure 3A), which is most likely due to changed light transmission in the area of retinoschisis [13]. However, in our study, three out of six patients (50%) presented with concentric areas of hyper-FAF (Figure 3C,D), which must be differentiated from other maculopathy or retinal dystrophies. Although fluorescence angiography is not required for the diagnosis of XLRS, FA may reveal nonleaking cystoid macular edema (Figure 3D), which may help clinicians better differentiate nicotinic acid toxicity, enhanced S-cone syndrome [11], and optic pit–related macular edema.

Optical coherence tomography

It has been suggested that retinoschisin functions as a cell adhesion protein to maintain the cellular organization of the retina and the structural integrity of the photoreceptor–bipolar synapse [13]. Mutation in the RS1 gene may cause cystic cavity formation in different layers of the retina. OCT has become one of the most important examinations for patients with XLRS since Stanga et al. first demonstrated a cleavage plane in the neural retina [14], and since then, several classifications have been proposed based on clinical and OCT examinations [15,16]. Using SD-OCT, clinicians can easily evaluate young patients when dilated ophthalmoscopy and ERG are difficult to perform. Although cystoid macular edema is the typical OCT finding for patients with XLRS, intraretinal cysts with no increase in the thickness of the macula, macular lamellar holes, and RPE atrophy can also be present in patients with XLRS (Figure 4).

Genetic study and phenotype–genotype correlation

All the mutations in our study that were clustered in the discoidin domain occurred within conserved residues. Several studies have shown that mutation in the discoidin domain causes protein misfolding and retention in the endothelium reticulum, which subsequently results in the phenotype [17,18].

Similar to the results from studies intended to determine phenotype–genotype correlations [12,19,20], no phenotype–genotype correlation was found in our study. We also found intrafamilial variations (clinical features and electroretinography) in two families with the same mutations (Families F and G). From these findings, it is clear that the phenotypes in XLRS can be highly variable in individuals, and even within the same family.

Electroretinography

A “negative ERG” was originally defined by Karpe as a normal a-wave and a diminished b-wave [21]. In the last two decades, the definition has been expanded to include any ERG findings in which the amplitude of the b-wave is smaller than the a-wave [22]. Although an electronegative ERG has been historically considered characteristic of XLRS, seven out of 12 patients (58%) showed electronegative ERGs, which is consistent with a report by Renner et al. (56.5%) [23]. In addition, since one patient in our study (Figure 6L, II-1) showed a normal b/a ratio, a normal b/a ratio in DA 3.0 ERG does not exclude the diagnosis of XLRS.

Vincent et al. reported that nonsense, splice-site, or frame-shifting mutations in RS1 consistently caused electronegative ERG in bright-flash (DA 11.0) stimuli, while missense mutations resulted in a wider range of ERG abnormalities [24]. In our study, seven patients with nonsense or deletion-related frameshift mutations underwent ERG; four had a b-wave smaller than the a-wave on the (DA 3.0) ERG (F,II-1, G,III-3, O,II-1, and P,II-1), and the other three had a b-wave larger than the a-wave on the (DA 3.0) ERG (F,I-2; G,III-2; L,II-1). This inconsistency may be related to the different bright flash intensities (DA 3.0 versus 11.0 ERG) used in the studies.

Dorzolamide for macular schisis/edema

Previous reports have shown that cystic macular lesions in XLRS may spontaneously resolve and further form an atrophic lesion in the macula [25,26]. Carbonic anhydrase inhibitors (CAIs) have been reported to be effective for improving cystoid macular edema in patients with XLRS [27-30]. The clinical effect of CAIs has been demonstrated through modulation of membrane-bound carbonic anhydrase IV receptors present in the RPE layer [31]. The administration of CAIs has also been shown to enhance fluid transport in the RPE barrier and to enhance retinal adhesiveness [32].

Genead et al. reported that 20 of 29 eyes (69%) had a positive response to topical dorzolamide treatment and that 16 eyes of 12 patients (55%) had improvements in BCVA by at least seven letters in at least one eye during the most recent follow-up visit [29]. Khandhadia et al. also reported a positive response to the treatment, which was not related to genotype. However, this response did not correlate with improvements in visual acuity [28]. In our study, although four (30.8%) of 13 eyes showed a decrease in the central foveal thickness (Figure 4I–L), no eyes demonstrated clinically significant improvements in VA. Additional studies with a longer follow-up period on treating younger patients with XLRS are necessary to elucidate whether the treatment of cystic-like lesions can decrease the incidence of late-onset atrophic lesions or improve VA when treatment starts earlier.

In conclusion, XLRS is a clinically heterogeneous disease, and patients from a family with the same mutation (genotype) may have different phenotypes. Young boys can present with vitreous hemorrhage and exudative retinal detachment (RD), with or without neovascular glaucoma that mimics Coats’ disease, and school-age boys may present with poor BCVA, amblyopia, or strabismus. Older patients, however, may present with retinal pigmentation or maculopathy, and not all patients show electronegative (scotopic maximal b/a <1) ERG. Although SD-OCT is a noninvasive, quick method for screening for the presence of macular schisis when dilated ophthalmoscopy and ERG are difficult to perform, XLRS may present as foveal atrophy in older patients. Although some patients have a positive response to topical dorzolamide, vision tends not to improve to a great extent. Although an electronegative (scotopic maximum b/a <1) ERG has been historically considered characteristic of XLRS, not all patients with XLRS have a negative ERG. Therefore, remaining highly aware of the possibility of XLRS and carrying out genetic testing for RS1 gene mutations is still the best course of action for diagnosing XLRS and the identifying its carriers. To the best of our knowledge, this study is the first report of clinical and genetic features in Taiwanese patients with XLRS and provides information for timely genetic counseling and gene therapy in the future.