Pseudoexfoliation syndrome and relating factors in a rural Japanese population: the Kumejima Study
Abstract
Purpose
To report the prevalence of pseudoexfoliation (PEX) syndrome and glaucoma associated with PEX (PEX-G) and their relating factors of them in a south-western island of Japan.
Methods
A population-based survey of all residents aged ≥ 40 years was conducted in Kumejima, Okinawa, Japan, and 3762 subjects (participation rate, 81.2%) underwent detailed ophthalmic examinations. Presence of PEX material on the lens capsule, iris surface and/or pupillary margin was determined by slit-lamp biomicroscopy. Glaucoma was diagnosed according to the criteria of the International Society of Geographical and Epidemiological Ophthalmology.
Results
In subjects aged ≥ 40 years, prevalence rates of PEX syndrome, PEX without glaucoma and PEX-G in at least one eye were 1.5% (95% confidence interval [CI]: 1.0–1.9%), 1.3% (95% CI: 0.9–1.7%) and 0.1% (95% CI: 0.0–0.2%), respectively, excluding eyes after cataract surgery; and 2.8% (95% CI: 2.3–3.3%), 2.2% (95% CI: 1.8–2.6 %) and 0.4% (95% CI: 0.2–0.6%), respectively, including eyes after cataract surgery. Cataract surgery had been performed in 61% of subjects with PEX in at least one eye; presence of PEX showed no significant effects on the intraocular pressure (IOP). A multivariate analysis showed that PEX was associated with older age (p < 0.0001, odds ratio: 1.10 [95% CI: 1.07–1.13]) and working outdoors (p = 0.0395, odds ratio: 2.18 [95% CI: 0.99–4.82]).
Conclusions
The prevalence rates of PEX syndrome and PEX-G in a south-western island of Japan were reported. PEX showed no significant effect on IOP, and age and working outdoors were significantly related with PEX.
Introduction
Pseudoexfoliation (PEX) syndrome is a condition characterized by the production and progressive accumulation of a fibrillar extracellular material in many ocular tissues as well as systemic distribution (Naumann et al. 1998; Ritch 2014). Pseudoexfoliation (PEX) occurs worldwide, and its prevalence increases universally with age; however, there are substantial ethnic and geographical variations in the prevalence For example, the reported prevalence rates of PEX in individuals ≥ 40-50 years of age ranged ≤ 1% (for example, in Singapore and central Japan (Yamamoto et al. 2005; Sumasri et al. 2008)) to ≥ 10% (in Greece (Kozobolis et al. 1997)). Geographical variation was reported even within the same ethnicity (Kozobolis et al. 1997; Miyazaki et al. 2005; Thomas et al. 2005; Topouzis et al. 2007; Jonas et al. 2013). Further, PEX is a major risk factor for development of glaucomatous optic neuropathy and glaucoma associated with PEX (PEX-G) is reportedly more progressive than primary open-angle glaucoma (POAG)(Naumann et al. 1998; Ritch 2014), accounting for the majority of the glaucoma-caused blindness in some of the Asia-Pacific regions (Krishnadas et al. 2003; Thomas et al. 2005; Abdul-Rahman et al. 2008).
In Japan, one population-based study in an inland urban city of central Japan, Tajimi, reported prevalence rates of PEX and PEX-G to be 1.0% and 0.2%, respectively, in the population aged ≥ 40 years (Yamamoto et al. 2005), while another study carried out in an inland urban town of south Japan, Hisayama, reported a prevalence rate of PEX of 3.4% in the population aged ≥ 50 years (Miyazaki et al. 2005). Residents in Kumejima, a south-western island of Okinawa, Japan, had a much higher prevalence rate of primary angle-closure glaucoma than that in Tajimi (2.2% versus 0.6%) (Iwase et al. 2004; Yamamoto et al. 2005), while the prevalence rate of POAG was similar between Kumejima and Tajimi (4.0% versus 3.9%) (Sawaguchi et al. 2012; Yamamoto et al. 2014).
As the cause of PEX and PEX-G, genetic background including LOXL1 and CACBA1A genes (Aboobakar et al. 2017) and environmental factors including solar exposure, especially activities over water or snow (Stein et al. 2011), are attracting recent attention. As the genetic background is reportedly different between people in Okinawa including Kumejima and those in mainland Japan (Hammer et al. 2006; Yamaguchi-Kabata et al. 2008), and differences in solar exposure between people on a small south-eastern island, Kumejima, and those in inland urban areas such as Tajimi or Hisayama are to be expected, the prevalence rates of PEX and PEX-G in Kumejima are of clinical interest. Herein, we report the prevalence rates of PEX and PEX-G and associated factors among people of Kumejima Island.
Subjects and methods
The Kumejima Study was a population-based, cross-sectional survey of eye health in one of the Ryukyu (Okinawa) islands, Japan. The design and protocol of the Kumejima Study basically followed the Tajimi Study (Iwase et al. 2004; Yamamoto et al. 2005) and was previously described (Sawaguchi et al. 2012; Yamamoto et al. 2014). The study followed the tenets of the World Medical Association's Declaration of Helsinki and municipal law of Kumejima for protecting privacy information. The ethics committee of the town of Kumejima approved the study protocol. All participants provided written informed consent.
Study population
In total, there were 5249 residents aged 40 years and older in Kumejima according to the registration in the municipal office as of May 2005. After excluding 617 subjects who were identified as nonresidents or had moved or died during the screening period, 4632 subjects were identified as the target population. All residents were encouraged to undergo examinations.
Screening and definitive examination
After obtaining systemic and ophthalmic medical histories, systemic interviews including history of working outdoors (a year unit), cumulative time of working outdoors (‘hr in a day’ × ‘year) and hat use. Height, weight and arterial blood pressure were measured. Refractive status was measured using an autorefractometer (ARK-730; Nidek, Nagoya, Japan), and visual acuity (VA) was measured using a chart of Landolt rings at a distance of 5 m with refractive correction using the data obtained with autorefractometer. Central corneal thickness (CCT) and corneal endothelial cells were measured using a noncontact specular-type pachymeter (SP-2000P, Topcon Co, Tokyo, Japan), and axial length and central anterior chamber depth were determined using a partial coherence laser interferometer (IOLMaster, Carl Zeiss Meditec, Dublin, CA, USA).
The anterior segment of the eye was examined by two experienced glaucoma subspecialists using slit-lamp biomicroscopy. The tear film, corneal epithelium, stroma and endothelium including pigment deposits on the endothelial surface were firstly observed. The pupillary margin and surface of the iris, presence or absence of transillumination of the iris, the anterior lens surface and lens with retroillumination technique, if necessary, were checked carefully. The peripheral anterior chamber depth was determined according to the Van Herick method.
Slit-lamp biomicroscopy was conducted without pupillary dilation in the screening examination and with pupillary dilation in the definitive examination as described later. The intraocular pressure (IOP) was measured twice by Goldmann applanation tonometry (GAT) under topical anaesthesia, and gonioscopy was performed using a Goldmann two-mirror lens (Haag-Streit, Köniz, Switzerland) in the primary position with the narrowest slit beam to minimize the effect of light on the anterior chamber angle configuration in dim lighting conditions. After the chamber angle width and occludability of the eye were assessed (Foster et al. 2004), the slit beam was widened for detailed observation of the angle structure and dynamic gonioscopy technique was used, if necessary. Pairs of sequential stereo-fundus colour photographs were obtained in a dark room through undilated pupils using a digital fundus camera (TRC-NW7S; Topcon Co) with angles of 30 and 45 degrees. The visual fields (VFs) were evaluated using a frequency-doubling technology screener (FDT, Carl Zeiss Meditec) with the C-20-1 screening test. When participants were unable to come to the facility, the doctors visited them at their homes and performed the examinations using a handheld slit lamp, direct and indirect ophthalmoscopes, and a Perkins applanation tonometer or a Tono-Pen XL (Bio-Rad Laboratory, Inc., Hercules, CA, USA) and gonioscopy.
Subjects suspected to have any ocular abnormalities including glaucoma were referred for definitive examination when their screening findings met one or more of the following criteria: (1) at least one of the three glaucoma specialists who independently evaluated the stereo-fundus photographs noted any findings suggestive of the presence of an abnormality including glaucomatous changes, that is vertical cup/disc ratio (v-CDR) of 0.6 or more; difference in the v-CDR of 0.2 or more between both eyes; superior rim width (from the 11 o'clock to the 1 o'clock position) or inferior rim width (from the 5 o'clock to the 7 o'clock position) of less than 0.2 of the disc diameter; and nerve fibre layer defects or splinter disc haemorrhages; (2) corrected VA worse than 20/30; (3) IOP more than 19 mmHg (Iwase et al. 2004); (4) any abnormal findings in the slit-lamp examination; (5) peripheral anterior chamber depth of van Herick grade 2 or less; (6) presence of gonioscopically occludable angle and/or peripheral anterior synechiae formation; or (7) at least one abnormal test point in the FDT-VF test.
Definitive examination included slit-lamp examination, GAT, optic nerve head (ONH) evaluation in the dark with a Goldmann 2-mirror lens and VF testing using the HFA central 24-2 Swedish Interactive Threshold Algorithm standard (SITA-S) program (Carl Zeiss Meditec). Unless gonioscopy showed the angle width to be narrow enough to indicate possible angle closure, the pupil was dilated to enable detailed observation of the anterior segment, ONH, and ocular fundus by slit-lamp biomicroscopy, and direct and indirect ophthalmoscopy.
Diagnosis
Pseudoexfoliation was diagnosed on slit-lamp biomicroscopy if there was whitish PEX material along the pupillary margin and/or the anterior lens capsule. Pigment accumulation on the corneal endothelium and/or anterior chamber angle and Sampaolesi's sign were also reviewed as the supporting findings. Glaucoma and suspected glaucoma were diagnosed based on the results of the screening and definitive examinations, and the subject's clinical history, as described previously (Sawaguchi et al. 2012; Yamamoto et al. 2014), based on the International Society of Geographical and Epidemiological Ophthalmology criteria (Foster et al. 2002). Six glaucoma specialists determined the final glaucoma diagnosis based on the results of the screening and definitive examinations and the subject's clinical history. VF was judged to be abnormal when the HFA 24-2 SITA-S program test result met at least one of the criteria of Anderson–Patella (Anderson & Patella 1999). The hemifield was considered abnormal when the pattern deviation probability plot showed a cluster of more than three contiguous points having sensitivity with a probability of < 5% in the upper or lower hemifield and in one of these with a probability of <1% (Anderson & Patella 1999).
Statistical analysis
To protect participants' privacy, all information was stored at the Data Analysis Center of Ryukyu University Faculty of Medicine. The code numbers of all participants were stored separately from all examination data in a Kumejima municipal office. The data were double-checked and validated through inspection.
Differences among the groups were evaluated using the t-test, Wilcoxon's rank-sum test, or Fisher's exact test. The prevalence of PEX and PEX-G and their 95% confidence intervals (CIs) were calculated for each group assuming that the prevalence rates in participants and nonparticipants were equal. The prevalence rates were calculated by direct age standardization from the Kumejima population (Okinawa_Prefectural_Government 2010). Associated factor analysis was carried out using generalized linear mixed model multiple logistic regression analysis using JMP Pro 13 (SAS Institute Inc., Cary, NC, USA). A p value < 0.05 was considered statistically significant. Variables with p < 0.10 by the univariate analyses (Table 2) were induced in the multivariate analysis. After confirming no strong intercorrelation among the variables (Spearman correlation coefficient < 0.65), the multivariate analysis was carried out using a backward elimination approach based on likelihood ratios.
Several previous studies excluded eyes after cataract surgery, because detection of PEX material is less sensitive in these eyes (Mitchell et al. 1999; Arnarsson et al. 2007; Landers et al. 2012; You et al. 2013), while others did not. To better compare the current results to previous ones under the same conditions, the prevalence rates were calculated both excluding and including eyes after cataract surgery.
Results
Subjects
From an eligible 4632 residents in Kumejima, the overall number of participants was 3762 (81.2%). The distribution of the subjects according to age and sex was summarized and reported previously.(Sawaguchi et al. 2012; Yamamoto et al. 2014) Of all the participants, 2399 were referred for definitive examinations, of which 185 (7.7%) declined or were unable to participate. As a result, 2214 subjects underwent definitive examinations including pupil dilation. Of the 185 subjects who did not undergo a definitive examination, a diagnosis was made based on the findings obtained in the screening examination including FDT as alternative for Humphrey VF.
Prevalence rates of PEX, PEX without glaucoma, PEX-G and suspected glaucoma associated with PEX (PEX-GS)
The overall prevalence rates of PEX, excluding eyes after cataract surgery, in at least one eye was 1.5% (95% CI: 1.0–1.9%), and prevalence rates of PEX without glaucoma, PEX-G and PEX-GS in at least one eye were 1.3% (95% CI: 0.9–1.7 %), 0.1% (95% CI: 0.0–0.3%) and 0.1% (95% CI: 0.0–0.1%), respectively (Table 1A). Mean ages (standard deviation) for PEX, PEX without glaucoma, PEX-G and PEX-GS were 75.9 (8.8), 75.9 (8.2), 72.5 (13.2) and 88 years, respectively, without significant intergroup difference among the PEX without glaucoma and PEX-G and PEX-GS (p = 0.4756). The prevalence of PEX increased significantly with age (p < 0.001, Fisher's exact test) (Table 1A). In subjects with PEX without glaucoma, 8 had bilateral and 31 had unilateral PEX. All of four subjects with PEX-G and one with PEX-GS had unilateral PEX, and none of them had angle closure. One PEX-G subject had been diagnosed and treated before the current study. Among subjects with PEX, no subjects had low vision (0.05 ≤ best-corrected VA of the better-seeing eye ≤ 0.3), but one subject with PEX without glaucoma had blindness (best-corrected VA of the better-seeing eye < 0.05). Causes of monocular low vision were cataract in three and age-related macular degeneration in one eye with PEX alone, and causes for monocular blindness were cataract in one eye with PEX without glaucoma, ocular trauma in one eye with PEX without glaucoma and glaucoma in one PEX-G eye. In eyes with PEX (regardless of the presence of glaucoma or glaucoma suspect) excluding eyes with history of glaucoma medications, surgery and/or laser therapy, the IOP averaged 15.3 (4.4) mmHg, which was not significantly different from 14.7 (3.5) mmHg for those without PEX (p = 0.2222). In 31 subjects with unilateral PEX alone, the IOP was 15.3 (4.0) mmHg and 14.7 (3.9) mmHg for eyes with and without PEX, respectively, (p = 0.2425). When eyes after cataract surgery were also included, the overall prevalence rates of PEX in at least one eye was 2.8% (95% CI: 2.3–3.3%), and prevalence rates of PEX without glaucoma, PEX-G and PEX-GS in at least one eye was 2.2% (95% CI: 1.8–2.7%), 0.4% (95% CI: 0.2–0.6%) and 0.2% (95% CI: 0.1–0.4%), respectively (Table 1B).
| Age group (years) | PEX | PEX without glaucoma | PEX-G | PEX-GS | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Male | Female | Total | Male | Female | Total | Male | Female | Total | Male | Female | Total | |
| (A) | ||||||||||||
| 40–49 | 0/487 | 0/458 | 0/945 | 0/487 | 0/458 | 0/945 | 0/487 | 0/458 | 0/945 | 0/487 | 0/458 | 0/945 |
| 50–59 | 1/491 | 1/356 | 2/847 | 1/491 | 0/356 | 1/847 | 0/491 | 1/356 | 1/847 | 0/491 | 0/356 | 0/847 |
| 60–69 | 2/313 | 7/299 | 9/612 | 2/313 | 7/299 | 9/612 | 0/313 | 0/299 | 0/612 | 0/313 | 0/299 | 0/612 |
| 70–79 | 7/311 | 12/398 | 19/709 | 7/311 | 10/398 | 17/709 | 0/311 | 2/398 | 2/709 | 0/311 | 0/398 | 0/709 |
| 80– | 4/100 | 10/169 | 14/269 | 3/100 | 9/169 | 12/269 | 1/100 | 0/169 | 1/269 | 0/100 | 1/169 | 1/269 |
| 60 and older | 13/724 | 29/866 | 42/1590 | 12/724 | 26/866 | 38/1590 | 1/724 | 2/866 | 3/1590 | 0/724 | 1/866 | 1/1590 |
| Prevalence (%)* | 1.8 (0.8–2.8) | 3.3 (2.1–4.6) | 2.6 (1.9–3.4) | 1.7 (0.7–2.6) | 3.0 (1.9–4.1) | 2.4 (1.6–3.1) | 0.1 (0.0–0.4) | 0.2 (0.0–0.6) | 0.2 (0.0–0.4) | 0 | 0.1 (0.0–0.3) | 0.1 (0.0-–0.2) |
| All subjects | 14/1702 | 30/1680 | 44/3382 | 13/1702 | 26/1680 | 39/3382 | 1/1702 | 3/1680 | 4/3382 | 0/1702 | 1/1680 | 1/3382 |
| Mean age (yrs) | 74.5 (8.4) | 76.5 (9.0) | 75.9 (8.8) | 73.4 (7.6) | 77.2 (8.4) | 75.9 (8.2) | 89 | 67.0(8.9) | 72.5(13.2) | 0 | 88 | 88 |
| Prevalence (%)* | 0.9 (0.4–1.4) | 2.1 (1.4–2.9) | 1.5 (1.4–1.9) | 0.8 (0.4–1.2) | 1.8 (1.1–2.2) | 1.3 (0.9–1.7) | 0.1 (0.0–0.2) | 0.2 (0.0–0.3) | 0.1 (0.0–0.3) | 0 | 0.1 (0.0–0.3) | 0.1 (0.0–0.1) |
| (B) | ||||||||||||
| 40–49 | 0/489 | 0/461 | 0/950 | 0/489 | 0/461 | 0/950 | 0/489 | 0/461 | 0/950 | 0/489 | 0/461 | 0/950 |
| 50–59 | 1/501 | 2/360 | 3/861 | 1/501 | 0/360 | 1/861 | 0/501 | 1/360 | 1/861 | 0/501 | 1/360 | 1/861 |
| 60–69 | 3/322 | 8/317 | 11/639 | 3/322 | 8/317 | 11/639 | 0/322 | 0/317 | 0/639 | 0/322 | 0/317 | 0/639 |
| 70–79 | 13/370 | 26/503 | 39/873 | 12/370 | 22/503 | 34/873 | 0/370 | 3/503 | 3/873 | 1/370 | 1/503 | 2/873 |
| 80– | 21/151 | 34/288 | 55/439 | 14/151 | 26/288 | 40/439 | 4/151 | 6/288 | 10/439 | 3/151 | 2/288 | 5/439 |
| 60 and older | 37/843 | 68/1108 | 105/1951 | 29/843 | 56/1108 | 85/1951 | 4/843 | 9/1108 | 13/1951 | 4/843 | 3/1108 | 7/1951 |
| Prevalence (%)* | 4.4 (3.0–5.8) | 6.1 (4.7–7.6) | 5.4 (4.4–6.4) | 3.4 (2.2–4.7) | 5.1 (3.8–6.3) | 4.4 (3.5–5.3) | 0.5 (0.0–0.9) | 0.8 (0.3–1.3) | 0.7 (0.3–1.0) | 0.5 (0.0–0.9) | 0.3 (0.0–0.6) | 0.4 (0.1–0.6) |
| All subjects | 38/1833 | 70/1929 | 108/3762 | 30/1833 | 56/1929 | 86/3762 | 4/1833 | 10/1929 | 14/3762 | 4/1833 | 4/1929 | 8/3762 |
| Mean age (yrs) | 85.0 (4.1) | 77.3(10.0) | 80.1 (9/1) | 78.4(8.9) | 79.8(8.0) | 79.3 (8.3) | 86.0 (3.5) | 77.9 (8.9) | 80.2 (8.5) | 84.0 (5.0) | 75.8(13.8) | 80.0(10.6) |
| Prevalence (%)* | 1.9 (1.4–2.5) | 3.7 (2.9–4.5) | 2.8 (2.3–3.3) | 1.5 (1.0–2.0) | 2.9 (2.2–3.7) | 2.2 (1.8–2.7) | 0.2 (0.0–0.4) | 0.5 (0.2–0.9) | 0.4 (0.2–0.6) | 0.2 (0.0–0.4) | 0.2 (0.0–0.4) | 0.2 (0.1–0.4) |
- PEX indicates pseudoexfoliation, PEX-G pseudoexfoliation glaucoma and PEX-GS pseudoexfoliation glaucoma suspect. Figures indicate mean (standard deviation).
- * The prevalence (95 % confidential interval) calculated by direct age standardization from the Kumejima population (Okinawa_Prefectural_Government 2010), assuming that it was equal between participants and nonparticipants.
Relating factors
Demographics of subjects without PEX and with PEX in at least one eye after excluding eyes after cataract surgery are shown in Table 2. In multivariate logistic regression analysis, factors with p values < 0.10 on univariate logistic regression analyses were taken into account. As high intercorrelations (Spearman's correlation coefficient > 0.75) were seen among history of working outdoors, cumulative time of working outdoors and hat use, history of working outdoors was adopted as a factor representing working outdoors. Presence of PEX was significantly associated with older age (p < 0.0001, odds ratio: 1.10 [95% CI: 1.07–1.13]) and history of working outdoors (p = 0.0395, odds ratio: 2.18 [95% CI: 0.99–4.82]) (Table 3).
| Numbers of subjects or mean (standard deviation) | Result of univariate logistic regression analysis Odds ratio (95% confidence interval) | p values* | ||
|---|---|---|---|---|
|
3330 subjects without PEX |
44 subjects with PEX |
|||
| Age (years) | 59.8 (13.2) | 75.9 (8.7) | 1.10 (1.07–1.13) | <0.0001 |
| Male/Female | 1685/1645 | 14/30 | 2.20 (1.16–4.15)† | 0.0123 |
| Height (cm) | 155 (9.2) | 148 (7.7) | 0.92 (0.89–0.95) | <0.0001 |
| Body mass index | 25.1 (3.7) | 24.3 (3.3) | 0.94 (0.86–1.02) | 0.1430 |
| Mean blood pressure (mmHg) | 99.8 (15.5) | 97.2 (11.9) | 0.99 (0.97–1.01) | 0.2831 |
| Mean ocular perfusion pressure (mmHg)* | 51.2 (10.2) | 49.4 (7.2) | 0.98 (0.95–1.01) | 0.1224 |
| Hypertension (−/+/unknown) | 1988/1332/10 | 222/22/0 | 0.67 (0.37–1.22) (−) vs.(+) | 0.1871 |
| Diabetes (−/+/unknown) | 3020/304/6 | 39/5/0 | 1.27 (0.50–3.26) (−) vs.(+) | 0.6135 |
| Smoking (−/+/unknown) | 1936/1389/5 | 32/12/0 | 0.52 (0.27–1.02) (−) vs.(+) | 0.0566 |
|
Best-corrected visual acuity Median (range) |
1.0 (hand motion–2.0) | 0.9 (hand motion–1.2) | 0.15 (0.06–0.37) | <0.0001 |
| Spherical equivalent (dioptres) | 0.03 (1.96) | 0.88 (1.46) | 1.27 (1.08–1.49) | 0.0052 |
| Axial length (mm) | 2.34 81.0) | 23.2 (0.7) | 0.75 (0.52–1.07) | 0.1190 |
| Anterior chamber depth (mm) | 3.1 (0.4) | 2.9 (0.4) | 0.24 (0.10–0.55) | 0.0006 |
| Central corneal thickness (mm) | 0.52 (0.03) | 0.50 (0.03) | 0.13 (0.00–1.32) | 0.0573 |
| Intraocular pressure (mmHg)‡ | 14.7 (3,5) | 15.5 (4.6) | 1.07 (0.98–1.15) | 0.1306 |
| Pterygium (−/+) | 2641/649 | 34/10 | 1.22 (0.60–2.27) | 0.5910 |
| Glaucoma (−/+) | 3196/134 | 40/4 | 1.19 (0.47–3.14) | 0.4114 |
| History of working outdoors (−/+) | 1214/2116 | 8/36 | 2.58 (1.20–5.57) | 0.0082 |
| Cumulative time of working outdoors (hrs) | 114.3 (142.6) | 190.5 (176.6) | 1.00 (1.00–1.01) | 0.0012 |
| Hat use (−/+) | 1447/1883 | 12/32 | 2.05 (1.05–3.99) | 0.0274 |
| Sunglasses use (−/+) | 3035/245 | 42/2 | 0.60 (0.14–2.49) | 0.4468 |
| Main occupation, farming (−/+)§ | 2471/859 | 28/16 | 1.66 (0.89–3.05) | 0.1253 |
| Main occupation, fishing (−/+)§ | 3222/108 | 44/0 | 0 | 0.0896 |
| Main occupation, office work (−/+)§ | 2789/541 | 41/3 | 0.38 (0.17–1.22) | 0.0614 |
| Main occupation, homemaker (−/+)§ | 2685/645 | 36/14 | 0.52 (0.27–0.98) | 0.0519 |
- PEX indicates pseudoexfoliation. In calculating mean or median of measures of ocular factors, one random chosen eye of a subject without pseudoexfoliation and that of a subject with bilateral pseudoexfoliation were adopted.
- * Likelihood ratio test.
- † 2/3 × (mean blood pressure − intraocular pressure).
- ‡ Eyes under ocular hypotensive therapy were excluded.
- § Main occupation was based on the participants' self-reports.
| Odds ratio (95% confidence interval) | p value* | |
|---|---|---|
| Age (years) | 1.10 (1.07–1.13) | <0.0001 |
| History of working outdoors | 2.18 (0.99–4.82) | 0.0395 |
- * Likelihood ratio test.
Discussion
Prevalence rate
In this report, we studied the prevalence rates of PEX and glaucoma associated with PEX and clinical relating factors in Kumejima, one of the rural islands in Okinawa in south-western Japan. The prevalence of PEX in at least one eye in Kumejima residents aged 40 years or older was 1.5% when eyes after cataract surgery were excluded, and 2.8% when eyes after cataract surgery were included. An apparently higher value in the latter group should be mainly attributed to possibly higher association of PEX and cataract (Krause & Tarkkanen 1978; Rudkin et al. 2008; Landers et al. 2012; Vijaya et al. 2015). It should be also associated that eyes with PEX and cataract had undergone early cataract surgery to minimize possible intraoperative complications (Naumann et al. 1998). In fact, if we include eyes after cataract surgery, 61% of subjects with PEX had undergone cataract surgery.
The prevalence rate of PEX in Kumejima excluding eyes after cataract surgery was lower than that reported in Beijing (5.8% in subjects aged ≥ 50 years) (You et al. 2013), indigenous Australians (5.9% in subjects aged ≥ 40 years) (Landers et al. 2012) and Iceland (10.7% in subjects aged ≥ 50 years) (Arnarsson et al. 2007) and somewhat lower than that reported in the Blue Mountains region of Australia (2.8% in subjects aged > 49 years) (Mitchell et al. 1999) versus 1.8% in the present subjects aged > 49 years) when compared under the same conditions.
On the other hand, the prevalence rate of PEX in eyes including eyes after cataract surgery was in a similar range to those reported under the same conditions in several regions in India (Thomas et al. 2005; Arnarsson et al. 2007; Jonas et al. 2013) and Myanmar (Abdul-Rahman et al. 2008), although the techniques of cataract surgery, which could potentially impact the detection of PEX material in undilated pupils, should vary widely among the studies including ours. It was higher than those in Singapore (Sumasri et al. 2008) and Sri Lanka (Rudkin et al. 2008) and lower than those in Spain (Viso et al. 2010) and Greece (Topouzis et al. 2007).
The large geographical variation in the reported prevalence rates of PEX ranging from low (≤1.0%) to high (≥10%) of the screened population is probably attributable to complexed interactions of various geographical and/or environmental factors with genetic risk variants to produce disease manifestations (Aboobakar et al. 2017). Recent studies suggested ambient temperature and sunlight exposure as possible important triggers of pseudoexfoliation (Naumann et al. 1998; Stein et al. 2011). This hypothesis appears to be consistent with the current results that when compared in eyes including those after cataract surgery, the prevalence rate of PEX in a south-western rural island of Japan, Kumejima, was higher than that in an inland urban city of central Japan (2.8% versus 1.0%) (Yamamoto et al. 2005) in subjects aged ≥ 40 years and similar to that in an inland urban town in south Japan (3.8% versus 3.4%) (Miyazaki et al. 2005) in subjects aged ≥ 50 years, and that working outdoors was a significantly related factor of having PEX besides older age in the current subjects.
Clinically associated factors
In the current univariate analysis, age, height, best-corrected VA, anterior chamber depth and factors associated with working outdoors showed significant difference between subjects or eyes with and without PEX after correction of multiple comparison (p < 0.05/25), but a multivariate analysis showed that only older age and working outdoors were significantly associated with having PEX in the current subjects. Association of older age and PEX is well known (Naumann et al. 1998; Ritch 2014), and association with working outdoors is consistent with the results of recent large-scale studies (Stein et al. 2011; Aboobakar et al. 2017).
Association of PEX with glaucoma has been of clinical interest (Ritch. 2014). Pseudoexfoliation (PEX) is characterized by the accumulation of extracellular fibrillar material in many tissues including eyes (Naumann et al. 1998; Ritch 2014) and is often reported to be associated with higher IOP (Kozobolis et al. 1997; Abdul-Rahman et al. 2008; Rudkin et al. 2008), open-angle glaucoma or angle-closure glaucoma (Naumann et al. 1998; Krishnadas et al. 2003; Thomas et al. 2005; Abdul-Rahman et al. 2008; Ritch 2014) which has a worse prognosis than other types of glaucoma (Naumann et al. 1998; Ritch 2014). Pseudoexfoliation (PEX) materials gradually accumulate in the chamber angle structure and the IOP is thought to gradually increase, finally resulting in glaucoma (Naumann et al. 1998; Ritch 2014). However, slit-lamp grading of PEX material suggested that PEX material deposited on the lens surface and/or pupillary margin in the advanced stage was not necessarily correlated with development of glaucoma (You et al. 2013) and PEX material itself might be not directly linked to the development of glaucoma (Sein et al. 2013). In the current subjects, the IOP of the eyes with PEX showed no significant difference from that of those without PEX. Furthermore, there was no significant inter-eye difference in IOP in subjects with unilateral PEX and multivariate analysis showed no significant association between IOP and PEX. This result is consistent with some previous population-based studies (Krishnadas et al. 2003; Arnarsson et al. 2007; You et al. 2013), although not with others (Kozobolis et al. 1997; Abdul-Rahman et al. 2008; Rudkin et al. 2008).
In the current study, the overall prevalence rate of PEX-G was 0.1% (0.3% in subjects aged ≥ 60 years), and the prevalence of PEX-G among subjects with PEX syndrome was 9.1% (95% CI: 2.5–21.3%). This figure, 9.1% tended to be higher than the prevalence of open-angle glaucoma in Japan, about 4% (Yamamoto et al. 2014). However, because of a wide range of 95% CI attributable to a small number of PEX-G subjects found in Kumejima residents, this difference was not statistically significant. This result was also confirmed by univariate and multivariate analyses of the current results, where association of glaucoma and PEX syndrome was not statistically significant, and no association of PEX and glaucoma was seen. The overall prevalence rate of PEX-G, 0.1%, was similar to that reported in an inland urban city of central Japan (0.2%) (Yamamoto et al. 2005), while reported prevalence of other previous studies ranged between 0% and 2.1% of the screened population (Krishnadas et al. 2003; Thomas et al. 2005; Yamamoto et al. 2005; Arnarsson et al. 2007; Topouzis et al. 2007; Abdul-Rahman et al. 2008; Sumasri et al. 2008; Landers et al. 2012). The current results indicate that the Japanese belong to the group with relatively low prevalence rate of PEX-G globally.
Previous studies yielded conflicting results on the association between gender and PEX syndrome. Several studies reported higher prevalence rates of PEX syndrome in males (Jonas et al. 2013), while others reported preponderance of female subjects (Arnarsson et al. 2007), suggesting complexed interactions of geographical and/or environmental factors with genetic factors to produce disease manifestations (Aboobakar et al. 2017). In the current study, the prevalence rate of PEX syndrome was significantly higher in females by univariate analysis, but this finding could not be confirmed by multivariate analysis. In an urban area of Japan, prevalence rate of PEX syndrome tended to be higher in females than males (Yamamoto et al. 2005), which was compatible with the current result obtained by univariate analysis. Although it is difficult to draw a conclusion on the association between PEX syndrome and gender in Japanese based on the current result, a possibility of female preponderance in Japanese subjects may not be excluded. Previous studies suggested an association of climatic keratopathy presumably associated with UV exposure (Landers et al. 2012; Sein et al. 2013), lower visual acuity (Krishnadas et al. 2003; Thomas et al. 2005; Jonas et al. 2013) or hypertension (Miyazaki et al. 2005) with PEX, but the current study failed to confirm these previous findings. These differences may be also a factor of the complicated interactions between various geographic and/or environmental factors and genetic risk variants in this disease (Aboobakar et al. 2017). Further, the relatively small number of subjects with PEX found among the current study population might have limited a statistical power for detection.
Limitation
In the present study, pupils were not dilated in 38% of the participants who declined or were not involved in the definitive examination. PEX material may be missed in 10% to 20% of undilated eyes (Forsius et al. 1974), and a previous population-based study (Krishnadas et al. 2003) reported that 32% of eyes with PEX had PEX materials visible only on the lens surface. A significant portion of which might be missed without pupil dilation. If presence of PEX materials was missed in 20% of the cases without pupil dilation (Forsius et al. 1974), the currently estimated prevalence rate of PEX would be an underestimation by approximately 8% (0.38 × 0.8 + 0.62 = 0.92), suggesting that error range in the prevalence rate of PEX in Kumejima excluding eyes after cataract surgery currently estimated would be within 10%. The corrected prevalence rate of PEX syndrome in the current subjects excluding eyes after cataract surgery would be 1.6% instead of 1.5%, and that including eyes after cataract surgery would be slightly greater than 3.0%, as detection of PEX materials should be less sensitive in eyes after cataract surgery (Mitchell et al. 1999; Krishnadas et al. 2003; Arnarsson et al. 2007; You et al. 2013). A large difference in the prevalence rate of PEX between when eyes after surgery were excluded and when those were included (1.5% versus 2.8%) in the current subjects was probably due to the fact that eyes with PEX and cataract had undergone early cataract surgery to minimize possible intraoperative complications (Naumann et al. 1998) and that a highly age-dependent PEX was more likely associated with cataract (Rudkin et al. 2008; Landers et al. 2012). So, to correctly estimate the prevalence rate of PEX in a population where many elders undergo cataract surgery, a method that sensitively detects PEX materials in eyes after cataract surgery must be developed.
In summary, the prevalence rates of PEX and PEX-G in Kumejima residents, a south-western island of Okinawa, Japan, was estimated to be 1.5% and 0.1%, when eyes after cataract surgery were excluded, and 2.8% and 0.4%, when eyes after cataract surgery were included, respectively, which was higher than those found in an inland urban city of central Japan. Age and working outdoors were factors associated with PEX, while an association of higher IOP, glaucoma or gender with PEX could not be confirmed in this population.
