Ocular abnormalities in a large Western China patient cohort with Retinitis Pigmentosa

Background: To report the prevalence of ocular abnormalities and investigate visual acuity in a large Western China cohort of retinitis pigmentosa (RP) patients. Methods: A retrospective study was performed, reviewing the medical records and ophthalmic examination reports of 2,127 eyes from 1,065 RP patients in one eye hospital. The authors investigated the prevalence of ocular abnormalities and the relationship between best corrected visual acuity (BCVA) and macular abnormalities. Results: Nyctalopia (58.2%) and blurred vision (27.1%) were the leading consultation causes. BCVA measurements in the better eyes at rst clinical presentation showed that 304 patients (28.5%) were categorised as blind and 220 patients (20.7%) as low vision. The most common ocular abnormalities were cataracts (43.1%) and macular abnormalities (59.7%), including epiretinal membranes (51.1%), cystoid macular oedema (18.4%), vitreomacular traction syndrome (2.4%), macular holes (2.3%) and choroidal neovascular membranes (0.05%). Glaucoma was found in 35 eyes (1.6%). The proportions of epiretinal membranes (P = 0.001) and macular holes (P = 0.008) increased signicantly with age. The proportions of vitreomacular traction syndrome (P = 0.003) and epiretinal membranes (P < 0.001) in pseudophakia and aphakia eyes were signicantly higher than in eyes that had not received operations (including cataracts and clear lens). Cystoid macular oedema was signicantly associated with poorer visual acuity in RP patients with clear lens (P = 0.002). Conclusion: Cataracts and macular abnormalities are common in RP patients. In the macular abnormalities, cystoid macular oedema may have a negative effect on BCVA in RP patients with clear lens.


Background
Retinitis pigmentosa (RP) is the most common type of inherited retinal dystrophy, causing progressive degeneration of the retinal pigment epithelium (RPE) and photoreceptors [1]. RP prevalence is approximately 1/4,000 and > 1.5 million patients are affected worldwide [2]. Nyctalopia and blurred vision are the most common RP symptoms, but other rare symptoms (e.g., photophobia, blurred vision) also prompt RP patients to see doctors [1][2][3]. However, little systematic information has been published on the clinical symptoms RP patients experience.
Ocular abnormalities (e.g., glaucoma, cataracts, maculopathy, etc.) may occur as RP progresses [4]. The typical histopathological change in RP is thinning of the photoreceptor's outer segments, which worsens with RP progression [4][5]. Although central vision acuity could remain normal for several years, anatomical macular abnormalities may occur in early-stage RP [6][7]. Epiretinal membranes (ERMs) and cystoid macular oedema (CME) are the most common macular abnormalities in RP patients, as detected by optical coherence tomography (OCT). Other macular abnormalities also accompanied by, such as macular holes (MH), vitreomacular traction syndrome (VMT) and choroid neovascularisation membrane (CNVM) [5][6][7][8][9]. To the authors' knowledge, visual acuity and prevalence of ocular abnormalities, have not yet been reported in a large cohort of Western Chinese RP patients.
This study, therefore, assesses the ocular abnormalities in a large cohort of Western Chinese RP patients. It also investigates correlations between visual acuity and macular abnormalities.

Study design and subjects recruitment
The authors retrospectively extracted medical records of patients diagnosed with RP between January 2014 and January 2019 at Southwest Hospital/ Southwest Eye Hospital, Third Military Medical University (Army Medical University), Chongqing, China. These records included information on each patient's age, gender, medical and surgical history, family history, complaints, best corrected visual acuity (BCVA), intraocular pressure, lens status, slit-lamp anterior segment and dilated fundus examination from the rst clinical presentation. RP diagnosis was based on: (1) presence of night blindness or blurred vision and peripheral vision eld restriction; (2) characteristic fundus changes, such as pale optic disc, attenuated vessels and bone-spicule-like pigmentation deposits in the mid-or far-periphery; and (3) reduced or nondetectable full-eld electroretinogram (ffERG) rod and cone amplitudes [1,4,5]. Systemic syndrome RP patients were included in the study. The exclusion criteria were: (1) trauma history; (2) vitreoretinal surgery and intravitreal therapy history; (3) pathological myopia; (4) other vascular retinopathy, such as hypertensive retinopathy, diabetic retinopathy, retinal periphlebitis, etc.; (5) age-related macular degeneration; (6) atypical RP, such as unilateral pigmentary retinopathy or sectorial pigmentary retinopathy; (7) secondary retinal pigmentosa; and (8)

Data collection and processing
The age of onset (that is, of symptoms) was de ned as the patient's age subtracted from the year with positive disease history. BCVA was measured with a Tumbling E chart and converted into the logarithm of the minimum angle of resolution (logMAR) value for analysis [10]. BCVA was classi ed according to the World Health Organization's (WHO) category of vision as follows [2]: BCVA worse than 3/60 in the better eye was considered blindness; BCVA of 3/60-6/18 in the better eye was considered low vision; and BCVA of 6/18 or more was considered normal. The researchers did not classify visual acuity according to vision eld. According to the International Society of Clinical Electrophysiology of Vision (ISCEV) standards, ffERG testing was performed [11]. The macular microstructure in RP patients was examined with either Spectral Domain OCT (Cirrus HD-OCT, Carl Zeiss Meditec, Dublin California, USA) or Heidelberg Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). Two experienced ophthalmologists independently evaluated the images. If the results differed, a third ophthalmologist re-evaluated them. Macular abnormalities were documented as follows: ERM, CME, MH (including lamellar and full-thickness MH), VMT and CNVM [1,12].
A specialist diagnosed glaucoma based on the presence of glaucomatous optic neuropathy, intraocular pressure over 21 mmHg, with or without the presence of iridotrabecular contact [13]. Lens condition was classi ed as clear, cataract, pseudophakic and aphakic. Some patients voluntarily underwent molecular diagnosis, and inherence patterns were categorised according to genetic test reports: autosomic dominant (AD), autosomic recessive (AR), X-linked (X-L) and sporadic cases (i.e., patients showing negative genetic reports or no evidence of other affected family members). RP patients' ages were divided into four groups (≤ 15 years, 16-44, 45-64 and ≥ 65) for statistical analysis, according to the International Classi cation of Disease.
Statistical analyses SPSS 22.0 was used to conduct analyses. Continuous variables, such as counselling age, age of onset and BCVA (logMAR), were expressed as means ± standard deviation (SD) and were compared with independent sample t-tests. Categorical variables (gender, complaints, inherence pattern, age group, lens condition and macular abnormalities) were presented as counts and percentages and compared with Chi-squared or Fisher's exact tests. Multiple linear regression investigated the relationship between BCVA (logMAR) and macular abnormalities. Coe cients of the estimated regression (β), the corresponding statistical signi cance (P), the exponential parameter and its con dence interval were presented for each factor. A P-value of < 0.05 was considered statistically signi cant.

Page 4/14
2127 eyes belong to 1,065 patients (493 [46.3%] female and 572 [53.7%] male, respectively.) were sampled. Table 1 describes the patients' demographic characteristics. The number of eyes was odd because three eyeballs in three female patients were enucleated because of glaucoma. The mean ± SD counselling age (i.e., age at patients' rst doctor visit) was 41.9 ± 15.7 years (range: three months to 83 years; females: 43.4 ± 16.0; males: 40.6 ± 15.3; P = 0.000). The mean ± SD age of onset for RP patients was 21.9 ± 19.2 years (females: 24.1 ± 19.7; males: 20.0 ± 18.6; P = 0.000). In both age of onset and mean counselling age, female patients were older than males. 352 of 1,065 patients (33.1%) chose molecular diagnosis, and the most common inheritance pattern was AR (57.7%), followed by Sporadic (27.6%), AD (8.8%) and X-L (6.0%) ( Table 1). Typical OCT and corresponding fundus photography of macular abnormalities in RP patients were provided in Fig. 3 (a-j), and the prevalence of macular abnormalities in this study were distributed as follows: MH and VMT were not found in patients ≤ 15 years old. ERM (P = 0.001) and MH (P = 0.008) were signi cantly more prevalent in elder RP patients, and that prevalence increased with age (Fig. 4b, Supplemental Table 4). No differences were observed in CME (P = 0.283) and VMT (P = 0.619) distributions among age groups. Because parts of patients had undergone cataract surgery (pseudophakic and aphakic eyes), the authors compared macular abnormality distribution between lens status. ERM (P < 0.001) and VMT (P = 0.003) were signi cantly more frequent in pseudophakic and aphakic eyes than in unoperated eyes (clear lens and cataracts) (Fig. 4c, Supplemental Table 5). The researchers also analysed the relationship between macular abnormalities and BCVA (logMAR) for RP patients with clear lens, and poor BCVA seemed signi cantly associated with CME (P = 0.002) ( Table 2).

Discussion
To our knowledge, this is the rst study to report on the prevalence of ocular abnormalities in a large cohort of Western Chinese RP patients and to also investigate the relationship between BCVA with macular abnormalities demonstrated by OCT. Results revealed that the most common ocular abnormalities were cataracts (43.1%) and macular abnormalities (59.7%). For macular abnormalities, CME was signi cantly associated with poorer visual acuity in RP patients with clear lens.
Macular abnormality was the most common ocular abnormality in RP patients, and it accounted for 59.7% of all checked cases and was distributed in our study as follows: ERM (51.1%), CME (18.4%), VMT (2.4%), MH (2.3%), and CNVM (0.05%). ERM have been reported to be the second most frequent macular abnormality in RP patients, with a prevalence rate of 0.6-35.4% (Table 3). However, the prevalence of ERM in our study was much higher (51.1%) than that recorded in previous studies [4,14,15], which may be due to the application of SD-OCT with higher resolution, different genetic backgrounds, and diagnostic methods. We noted the presence of ERM when even a subtle, hyper-re ective lesion adhered to the inner retinal surface, regardless of other abnormalities being present. Testa et al performed a retrospective study investigating the prevalence of macular abnormalities in Usher syndrome patients [12] and found its prevalence to be at 45.1%, and they designated the most frequent abnormalities as CME (20.4% eyes), followed by ERM (15.6%), VMT (5%), and MHs (2%). The mechanisms of ERM formation remains unclear. However, it may include (1) idiopathic preretinal glial cell proliferation, (2) in ammation revealed by elevated aqueous are, and (3) chronic macular-vitreous traction [15][16][17]. Our results demonstrated that CME was the second most common macular abnormality, and this coincided with results from an Italian population for which Testa et al investigated macular abnormalities in 581 RP subjects [1] and found that the most frequent abnormalities was CME (20.4% eyes), followed by ERM (15.6%), VMT (5%), and MH (2%). CME varies between 5.5% and 49% in RP patients [4,14]. The exact mechanism of CME in RP remains unclear; however, it may include (1) the breakdown of the blood-retinal barrier secondary to the degeneration of RPE and/or Müller cells, (2) anti-retinal antibodies, and (3) traction from ERM and VMT. There is no consensus on the relationship between CME and visual acuity in RP patients [15][16]. Sandberg et al. believed that retinal thinning and thickening appeared to be associated with lower visual acuity in RP patients [18]. Yoshida et al. demonstrated that a normal preoperative ellipsoid zone (EZ), also called the inner/outer segment junction (IS/OS), was signi cantly related to better BCVA after cataracts in RP patients [19]. Because cataracts and PSCs were prevalent in RP subjects and were negatively correlated with BCVA, we analyzed the relationship between macular abnormalities and BCVA (logMAR) only in eyes with clear lens. CME appeared to be signi cantly associated with poor BCVA in our study. The exact relationship between maculopathy and visual acuity requires greater attention in future studies. CNVM are rare, and until recently, no data has shown the prevalence of CNVMs in RP patients.
For several years, this information could only be attained through case reports [9,20,21]. In our study, a CNVM was observed in only one eye from one female patient (prevalence: approximately 0.09%). It has been proposed that photoreceptor cell degeneration and choriocapillaris damage may lead to the formation of CNVM [20].  (Table 3) [22]. Posterior subcapsular cataracts (PSCs) are the most typical morphological abnormalities and occur in 63-83.9% of RP patients [19,23,24]. However, lens status was determined through medical records, and cataract type was unidenti able in the study. Glaucoma is another ocular abnormality prevalent among RP subjects. There is some evidence to suggest similar genetic backgrounds for glaucoma and RP [13,25]. Ko YC et al. reported a 3.64-fold greater odd of developing PACG in patients with RP than in the general population [26]. In our study, the prevalence of glaucoma was 2%, lower than the 11.5% reported by Onakpoya et al [2]. and the 7.5% reported in Eballe et al. [27], but similar to the prevalence in the general population (2-3%) [13]. The reason for the lower rate may be due to our larger sampled cohort and our study being retrospective study.
More than half of the RP patients in our study presented visual acuity deterioration at rst clinical presentation, and the proportion of blindness and low vision de ned by BCVA were 28.5% and 20.7%, respectively. We de ned low vision or blindness according to central visual acuity and did not consider visual eld defects and blindness. The low vision rates in the RP subjects were actually much higher than these results show.
This study had the advantage of a large sample size and assessed various ocular abnormality distributions and visual acuity simultaneously. However, it included several limitations. First, the study was retrospective, and other ocular abnormalities and details, such as corneal nebula, cataract and glaucoma types, and so on, remained unexplored. Secondly, some patients had no molecular diagnosis, and we could not su ciently investigate ocular abnormalities in different genetic subtypes. Further studies including prospective investigations and more patients with genetic diagnoses, are needed to explore the relationship between the course of RP and BCVA or to clarify the relationship between the genetic phenotypes of RP and BCVA.     Supplementary Files