Myopia and associated pathological complications

Authors


Correspondence and reprint requests to: A/Prof. Seang-Mei Saw.
Tel.: (65) 6 874 4989; Fax: (65) 6 779 1489.
E-mail address: cofsawsm@nus.edu.sg

Abstract

Besides the direct economic and social burden of myopia, associated ocular complications may lead to substantial visual loss. In several population and clinic-based cohorts, case–control and cross-sectional studies, higher risks of posterior subcapsular cataract, cortical and nuclear cataract in myopic patients were reported. Patients with high myopia (spherical equivalent at least –6.0 D) are more susceptible to ocular abnormalities. The prevalent risks of glaucoma were higher in myopic adults, and risks of chorioretinal abnormalities such as retinal detachment, chorioretinal atrophy and lacquer cracks increased with severity of myopia and greater axial length. Myopic adults were more likely to have tilted, rotated, and larger discs as well as other optic disc abnormalities. Often, these studies support possible associations between myopia and specific ocular complications, but we cannot infer causality because of limitations in study methodology. The detection and treatment of possible pathological ocular complications is essential in the management of high myopia. The ocular risks associated with myopia should not be underestimated and there is a public health need to prevent the onset or progression of myopia.

Myopia has reached epidemic proportions and is already a large public health problem in certain parts of the world, including East Asia (Grosvenor, 2003). The rates of high myopia, and possibly pathological myopia, appear to be rising in Asia and other parts of the world. Cross-sectional prevalence surveys across the United States (National Health and Nutrition Examination Survey, Baltimore Eye Survey, Beaver Dam Eye Study, and the Framingham Offspring Eye Study) suggest that there are higher myopia prevalence rates with more recent birth cohorts (Sperduto et al., 1983; Wang et al., 1994; The Framingham Offspring Eye Study Group, 1996; Katz et al., 1997). The evidence for this phenomenon from repeated prevalence surveys, however, needs to be further substantiated because of differences in methodology (Lin et al., 1988,1999). The apparent worldwide rise in the prevalence of myopia has a large public health impact because of the associated concomitant increase in potentially blinding ocular conditions.

Common definitions of high myopia or myopia with increased risks of ocular morbidity, include spherical equivalent (SphE) of at least –6.0 D, SphE at least –8.0 D, or SphE at least –10.0 D. Ocular pathology is usually due to excessive elongation of the eyeball and associated with pathological changes in the fundus (Goldschmidt, 1988; Tokoro, 1988). The terms ‘malignant myopia’, ‘degenerative myopia’ and ‘pathological myopia’ have also variously been used to describe myopia accompanied by degenerative changes in the sclera, choroid, retinal pigment epithelium and associated compromises in visual function (Duke-Elder, 1970; Daubs, 1982). Tokoro defined pathological myopia as myopia caused by pathological axial elongation and showed that the prevalence increased from 0.5% in junior high school Japanese students to a peak of approximately 3% in 29-year-old adults (Tokoro, 1988).

We aim in this review to summarize the best evidence of myopia-associated ocular pathologies, including cataract, glaucoma, chorioretinal abnormalities, optic disc abnormalities and age-related macular degeneration. Acronyms are summarized in Table 1.

Table 1.  Table of acronyms
SphESpherical equivalent
DDiopters
ARAutorefraction
SRSubjective refraction
PSCPosterior subcapsular
RRRelative risk
OROdds ratio
CIConfidence interval
AREDSAge-Related Eye Disease Study
CDRCup–disc ratio
POAGPrimary open-angle glaucoma
ALAxial length
VFDVisual field defect
OHTSOcular Hypertension Treatment Study
IOPIntraocular pressure
AMDAge-related macular degeneration

Myopia and cataract

Cataract is the leading cause of blindness worldwide (Resnikoff et al., 2004). Data from studies of cataract as a possible complication of myopia include population-based cohort studies conducted in the United States (Beaver Dam Eye Study), Australia (Blue Mountains Eye Study) and Barbados (Barbados Eye Study) (Lim et al., 1999; Wong et al., 2001; Leske et al., 2002). Population-based cross-sectional studies (both in Australia) and clinic-based case–control studies (UK) are also presented in Table 2 (Brown and Hill, 1987; McCarty et al., 1999; Younan et al., 2002). Refraction measurements were determined by either autorefraction (AR) or subjective refraction (SR) or both. Cataract was confirmed by lens photography. In the Blue Mountains Eye Study (n = 2334; follow-up = 5 years) of adults 49 years and older, the multivariate adjusted odds ratio (OR) of incident PSC was 4.4 (95% CI 1.7, 11.5) for moderate myopia (SphE at least –3.5 D) compared with emmetropia, and 3.3 (95% CI 1.5, 7.4) for nuclear cataract in adults with high myopia (SphE at least –6.0 D) (Younan et al., 2002). There were also associations with prevalent posterior subcapsular (PSC), cortical and nuclear cataract. Refraction-related increasing odds were found between PSC cataract and myopia: low myopia (OR 2.1; 95% CI 1.4, 3.5), moderate myopia (OR 3.1; 95% CI 1.6, 5.7) and high myopia (OR 5.5; 95% CI 2.8, 10.9). PSC, cortical, and late nuclear cataract were associated with high myopia (Lim et al., 1999). The multivariate adjusted OR of incident nuclear cataract in myopic adults (SphE at least –0.5 D) in the Barbados Eye Study of adults aged 40–84 years (n = 2609; follow-up = 4 years) was 2.8 (95% CI 2.0, 4.0) (PSC and cortical cataract results were not reported) (Leske et al., 2002). Myopia was not associated with incident cataract in the Beaver Dam Eye Study of adults 43–84 years (n = 4470; follow-up = 5 years), but only with prevalent nuclear cataract [OR 1.74 (95% CI 1.28, 2.37)] (Wong et al., 2001).

Table 2.  Summary of published data on cataract as a complication of myopia (for abbreviations see Table 1)
Author (Year) (Country) EthnicityStudy design Study population (n)Methodology and definitionsResults
Brown and Hill (1987) (UK) WhiteCase-control studyCases (cataract patients) and controls (patients without cataract) aged 40 years or older, matched by age at a London ophthalmic clinic. High myopia (SphE at least –12 D) excluded (n = 220)SR Lens photography (cataract)The crude OR of myopia (not defined) = 1.06 (95% CI 0.6, 1.9) (computed from table 7 using unmatched analysis)
Lim et al. (1999) (Australia) WhitePopulation-based cross-sectional studyBlue Mountains Eye Study 49 years and older (7308 eyes)AR and SR Lens photography (cataract)OR of prevalent cataract for high myopes (SphE at least –6.0 D), adjusted for age, sex, smoking, hypertension, diabetes, steroid use and sun-related skin damage, were: 4.9 (95% CI 2.1, 11.4) for PSC 2.9 (95% CI 1.4, 6.0) for cortical cataract 1.4 (95% CI 0.8, 2.4) for nuclear cataract
McCarty et al. (1999) (Australia) WhitePopulation-based cross-sectional studyVisual Impairment Project 40 years and older (n = 5147)AR Lens photography (cataract)OR of prevalent cataract for myopia (SphE at least –1.0 D) were: 3.59 (95% CI 2.50, 5.15) for PSC, adjusted for age, rural, diuretics, vitamin E, vitamin A, ultraviolet light 1.76 (95% CI 1.30, 2.40) for cortical cataract, adjusted for age, gender, iris, arthritis, diabetes, gout, beta-blockers, ultraviolet light, glaucoma 2.73 (95% CI 1.90, 3.92) for nuclear cataract, adjusted for age, gender, diabetes, smoking and education
Wong et al. (2001) (US) WhitePopulation-based cohort studyBeaver Dam Eye Study 43–84 years (n = 4470) (FU = 5 years)AR (baseline) Lens photography (prevalent cataract at baseline and incident cataract at 5 years)OR of prevalent cataract for myopia (SphE at least –1.0 D), adjusted for age, gender, diabetes, smoking, and education, were: 1.23 (95% CI 0.75, 2.03) for PSC 0.86 (95% CI 0.64, 1.16) for cortical cataract 1.74 (95% CI 1.28, 2.37) for nuclear cataract Myopia was not associated with incident cataract
Younan et al. (2002) (Australia) WhitePopulation-based cohort studyBlue Mountains Eye Study 49 years and older (n = 2334) (FU = 5 years)AR and SR (baseline) Lens photography (cataract at 5 years)OR of incident cataract were: 4.4 (95% CI 1.7, 11.5) of moderate myopia (SphE at least –3.5 D) for PSC, adjusted for age, sex, education, obesity, hypertension, nuclear cataract 0.5 (95% CI 0.2, 2.0) of high myopia (SphE at least –6.0 D) for cortical cataract, adjusted for age, sex, education, alcohol, ultraviolet light, diabetes, obesity, stroke, nuclear cataract 3.3 (95% CI 1.5, 7.4) of high myopia (SphE at least –6.0 D) for nuclear cataract, adjusted for age, sex, smoking, education, iris, inhaled steroids
Leske et al. (2002) (Barbados) Caribbean AfricansPopulation-based cohort studyBarbados Eye Study 40–84 years (n = 2609) (FU = 4 years)AR (baseline) Lens photography (cataract at 4 years)RR of incident cataract for myopia (SphE at least –0.5 D), adjusted for age, gender, body mass index, iris color, diabetes, IOP, and IOP treatment was 2.8 (95% CI 2.0, 4.0) for nuclear cataract (PSC and cortical cataract not reported)

From prevalence estimates, we cannot infer that cataract is a complication of myopia because it is also possible that adults with cataract may have higher risks of myopia. Associations with PSC, cortical and nuclear prevalent cataract were found in the Visual Impairment Project in Australia (n = 5147) of adults 40 years and older (McCarty et al., 1999). However, no relationship between myopia and cataract was found in a case-control study in the UK (n = 220) of adults 40 years and above (Brown and Hill, 1987). In the Age-Related Eye Disease Study (AREDS) of 4477 adults aged 60–80 years, participants with myopia had ORs of 0.72 for mild nuclear opacity and 0.85 for mild cortical opacity compared with hyperopes (Age-related Eye Disease Study Research Group, 2001).

From the cross-sectional studies, we cannot exclude the possibility that myopic shifts may have occurred as a consequence of cataract (notably nuclear) development. The cataract–myopia relationship may be in the opposite direction and the progression of opacity in the lens nucleus may initiate the development of myopia. However, there is a large body of evidence from population and clinic-based studies that cataract (PSC, nuclear and occasionally, cortical cataract) may be associated with high and low myopia in European-derived and Caribbean African populations (Brown and Hill, 1987; Lim et al., 1999; McCarty et al., 1999; Wong et al., 2001; Leske et al., 2002; Younan et al., 2002). The mechanism by which myopia may lead to lens changes is unknown. However, myopia has been linked to damage of rod outer segments and increased production of potentially cataractogenic lipid peroxidation by-products (Zigler et al., 1983).

Myopia and glaucoma

Myopic eyes are structurally different from emmetropic eyes: myopic eyes have longer axial lengths and vitreous chamber depths (Scott and Grosvenor, 1993). Eyes with increased axial length appear to have higher cup–disc ratios (CDRs), increased optic nerve fibre layer defects and possibly greater deformability of the lamina cribrosa, leading to higher susceptibility to glaucomatous optic disc changes (Fong et al., 1990). The glaucoma–myopia association has been described in case–control and cross-sectional studies in Europe, Asia, Australia and the USA, but its effect on incident glaucoma has not yet been evaluated in prospective cohort studies (Table 3) (Daubs and Crick, 1981; Ponte et al., 1994; Mitchell et al., 1999; Grodum et al., 2001; Yoshida et al., 2001; Wong et al., 2003). As the temporal relationship between glaucoma and myopia is not well delineated, a cause-effect relationship between myopia and glaucoma cannot be established. The definitions of primary open-angle glaucoma (POAG) differed and refraction measures were assessed by autorefraction (AR) or subjective refraction (SR). All studies found associations of glaucoma with myopia in patients aged 40 years and older [Casteldaccia Eye Study (n = 264): OR = 5.56; Beaver Dam Eye Study (n = 4670): OR = 1.6] (Ponte et al., 1994; Wong et al., 2003). Several studies showed that glaucoma risks increased with more severe myopia. The OR of glaucoma for moderate myopia was 3.3 in the Blue Mountains Eye study of 3654 adults while glaucoma rates increased with increasing myopia in logistic regression models (regression coefficient –0.373, p < 0.001) in 32 918 Swedes (Mitchell et al., 1999; Grodum et al., 2001). In Japan (n = 64 394), there were positive associations between the strength of myopic refraction and the prevalence of OAG for specific age-gender groups (p < 0.001 in men and p < 0.001 in women) (Yoshida et al., 2001). For example, in women aged 65–74 years, 5.4% of those with moderate-to-high myopia had OAG in contrast to 2.0% of those with hyperopia. The relative risk (RR) of glaucoma for high myopia (SphE at least –5.0 D) in the UK (n = 953) was 3.1 (Daubs and Crick, 1981). The Casteldaccia Eye Study, Blue Mountains Eye Study, Beaver Dam Eye Study and Swedish studies were population-based, while eye clinic patients were examined in studies conducted in Japan and the UK.

Table 3.  Summary of published data on glaucoma as a complication of myopia (for abbreviations see Table 1)
Author (Year) (Country) EthnicityStudy design Study population (n)Methodology and definitionsResults
Daubs and Crick (1981) (UK) WhiteCase–control studyGeneral ophthalmology patients, King's College Hospital, London (n = 953)SR OAG defined as eyes with open angles and characteristic VFDOR of OAG was 3.1 (95% CI 1.6, 5.8) for high myopia (SphE at least –5.0 D), 1.3 (95% CI 1.0, 1.8) for low myopia (SphE > –0.25 D to –5.0 D) compared with hyperopia (SphE at least + 0.25 D), adjusted for age, IOP, sex, family history, season, blood pressure, astigmatism, urinalysis and health
Ponte et al. (1994) (Italy) WhitePopulation-based case–control studyCasteldaccia Eye Study 40 years and older (n = 264)Cases: IOP ≥ 24 mm Hg, or history of glaucoma, or visual fields suggestive of glaucoma. Controls: subjects with IOP ≤ 20 mmHg, cup-disc ratios 0–0.2 and pink discsOR of prevalent glaucoma for myopia (SphE at least –1.5 D) was 5.56 (95% CI 1.85, 16.67), adjusted for diabetes, hypertension, steroid use and iris texture
Mitchell et al. (1999) (Australia) WhitePopulation-based cross-sectional studyBlue Mountains Eye Study 49 years and older (n = 3654)AR and SR OAG defined as cup-disc ratio ≥0.7 or cup-disc asymmetry ≥0.3OR of prevalent OAG was 3.3 (95% CI 1.7, 6.4) for moderate to high myopia (SphE at least –3.0 D) and 2.3 (95% CI 1.3, 4.1) for patients with low myopia (SphE < –3.0 D and ≥−1.0 D), adjusted for sex, family history, diabetes, hypertension, migraine, steroid use and pseudoexfoliation
Grodum et al. (2001) (Sweden) WhitePopulation-based cross-sectional studyResidents of Malmo born between 1918 and 1932 (n = 32 918)AR OAG defined as repeatable VFD on two consecutive tests on different daysPrevalence of newly detected OAG increased with increasing myopia (SphE at least –1 D) (p < 0.01). [0.6% in hypermetropia (≥+1.0 D); 0.9% in emmetropia (SphE > –1 D to < + 1 D) and 1.5% (SphE ≤−1 D) in moderate to high myopia], adjusted for age, gender and IOP
Yoshida et al. (2001) (Japan) JapaneseCross-sectional studyOptometry clinic patients, Yokohama, 6–98 years (n = 64 394)AR OAG defined as glaucomatous VFD associated with abnormal optic disc and/or disc marginPrevalence of OAG higher for moderate to high myopes (SphE at least –3 D) compared with hyperopes in males (p < 0.001) and females (p < 0.001)
Wong et al. (2003) (USA) WhitePopulation-based cross-sectional studyBeaver Dam Eye Study 43–86 years (n = 4670)AR and SR POAG defined as VFD compatible with glaucoma, IOP ≥22 mm Hg, cup–disc ratio 0.8 or more, history of glaucoma treatmentThe age and gender adjusted ORs of prevalent POAG for myopia (SphE at least –1.0 D) was 1.6 (95% CI 1.1, 2.3)

It appears that glaucoma may be more common in patients with longer ALs, though changes in risk estimates with varying ALs have not been addressed in published studies. There is no well-defined value of AL that is associated with higher risks of glaucoma. The severity of glaucomatous VFDs and optic disc abnormalities were shown to be greater in myopic adults (Quigley et al., 1994; Chihara et al., 1997; Jonas and Dichtl, 1997). In the Ocular Hypertension Treatment Study (OHTS) of the effect of topical ocular hypotensive medication in preventing the onset of POAG, myopia (a baseline risk factor) was not associated with incident POAG (Gordon et al., 2002).

The effects of myopia on IOP, as well as the incidence and severity of glaucomatous VFDs and optic disc changes, have also been evaluated (Greve and Furuno, 1980; Quinn et al., 1995; Chihara et al., 1997; Ko et al., 2002; Wong et al., 2003). In a cross-sectional survey of children in a Philadelphia hospital (n = 321), the mean IOP of myopic eyes (17.8 mm Hg) was higher compared with non-myopic eyes (17.1 mm Hg) (p < 0.01) (Quinn et al., 1995). This relationship remained after controlling for age, family history of myopia and amblyopia. The increase in VFD was greater (1.04 mean change in rank of VFD) in adults with high myopia (SphE at least –4.0 D) compared with emmetropes (0.53) in 122 POAG Japanese patients (Chihara et al., 1997). In another study of 993 white adults with ocular hypertension, refractive error did not predict incident glaucomatous VFD (Quigley et al., 1994). The optic discs were larger (p < 0.0001), shape more elongated (p < 0.0005) and cup depth significantly shallower (p < 0.0001) in 44 highly myopic (SphE at least –5.0 D) white POAG patients compared with 571 POAG patients with low to moderate myopia or hyperopia (Jonas and Dichtl, 1997).

Myopia and chorioretinal abnormalities

Increased axial elongation in myopes may lead to mechanical stretching and thinning of the choroid and retinal pigment epithelium with concomitant vascular and degenerative changes (Pierro et al., 1992). A myriad of chorioretinal abnormalities associated with myopia have been evaluated in one case–control and five cross-sectional clinic-based studies. These abnormalities include retinal breaks, chorioretinal atrophy, Fuch's spot, lacquer cracks, pigmentary degeneration, lattice degeneration, posterior staphyloma and white without pressure (Table 4) (Hyams and Neumann, 1969; Curtin and Karlin, 1970; Karlin and Curtin, 1976; Celorio and Pruett, 1991; Pierro et al., 1992; The Eye Disease Case–Control Study Group, 1993; Yura, 1998). Data of associations of prevalent retinal abnormalities with varying AL (n = 3), presence of myopia (n = 2), as well as the severity of myopia (n = 1) are available. However, these findings are derived from clinic-based populations and have yet to be confirmed in population-based cohort studies of incident retinal abnormalities. Ideally, population-based cohort studies describing the association between myopia and incident chorioretinal abnormalities should be conducted to provide valuable data.

Table 4.  Summary of published data on chorioretinal abnormalities as complications of myopia (for abbreviations see Table 1)
Author (Year) (Country) EthnicityStudy design Study population (n)Methodology and definitionsResults
Hyams and Neumann (1969) (Israel) WhiteCross-sectional studyAsymptomatic eye clinic patients with myopia at least –1.0 D, aged 10–65 years (332 eyes)Slit-lamp biomicroscopy13% of high myopes (SphE at least –6.0 D), 10.7% of moderate myopes (SphE –3.25 to –6.0 D) and 10.5% of low myopes (SphE –1.0 to –3.0 D) had retinal breaks
Curtin and Karlin (1970)Karlin and Curtin (1976) (USA) WhiteCross-sectional studyEye clinic patients with myopia (1437 eyes) Patients with hyperopia or emmetropia (n = 100)A scan (AL) Binocular indirect ophthalmoscopy (BIO) + direct ophthalmoscopy (Curtin and Karlin) or scleral indentation (Karlin and Curtin)Percentage of chorioretinal atrophy was 0% if AL < 24.5 mm and 23% if AL ≥24.5 mm Percentage with Fuch's spot was 0% if < 26.5 mm and 5.2% if ≥26.5 mm Percentage with lacquer cracks was 0% if < 26.5 mm and 4.3% if ≥26.5 mm Percentage white without pressure increased from 0% at 20–21 mm to 54% at 33 mm Percentage lattice degeneration increased with AL (p < 0.01)
Celorio et al. (1990) (USA) WhiteCross-sectional studyRecords of patients with high myopia (SphE at least –6.0 D) reviewed (n = 218; 436 eyes)A scan (AL) BIO with scleral depression Slit-lamp biomicroscopyPercentage of lattice degeneration decreased with AL (40.9% for AL 26–26.9 mm and 7.0% for 32 mm or greater)
Pierro et al. (1992) (USA) WhiteCross-sectional studyPatients in an ophthalmology clinic with AL ≥24 mm (n = 513)A scan (AL) Slit-lamp biomicroscopy with scleral depressionPercentage of eyes with one of more retinal lesions (white with or without pressure; lattice degeneration; pavingstone degeneration; posterior vitreous detachment) increases with AL
The Eye Disease Case–Control Study Group (1993) (USA) WhiteCase–control studyCases of idiopathic rhegmatogenous retinal detachments and age-sex- race-clinic matched controls (free of retinal disease) from five eye centres, high myopia (SphE at least –8 D) excluded (n = 1391)AR/SR Confirmed retinal detachment from surgical recordsThe OR of retinal detachment for myopes (SphE at least –1 D) was 7.8 (95% CI 5.0, 12.3), adjusted for age, sex, race and clinic
Yura (1998) (Japan) JapaneseCross-sectional studyClinic patients with high myopia (AL's 26–31.99 mm) (n = 542; 970 eyes)AR and SR A scan (AL) BIO with scleral depressionPercentages of posterior staphyloma, but not lattice degeneration increased with AL

The prevalence of lattice degeneration, which is a risk factor for developing retinal breaks, increased with AL in 2 studies by Curtin and Karlin (1,437 eyes of all ages), and Pierro (513 patients) in white patients aged 10 years and above in the USA (Curtin and Karlin, 1970; Karlin and Curtin, 1976; Pierro et al., 1992). No association was, however, found in a study of 542 Japanese patients of all ages (Yura, 1998). We noted that in a clinic-based study of 218 white patients of all ages in the USA, the prevalence was actually lower in patients with extreme AL (>32 mm) (Celorio and Pruett, 1991). The prevalence of retinal breaks in Israeli eye clinic patients aged 10–65 years (332 eyes) was 13% with high myopia (SphE at least –6.0 D), but only 10.5% in patients with low myopia (SphE –1.0 to –3.0 D) (Hyams and Neumann, 1969). This difference, though, may not be clinically significant and patients who present at the clinic may be more likely to have dual pathology (retinal breaks and high myopia). The prevalence of posterior vitreous detachment, which is involved in the development of many retinal breaks, was 12.5% in a case series of patients with high myopia and 60.7% in patients with AL of more than 30.0 mm (Morita et al., 1995). Other vitreous changes observed in myopic eyes include posterior vitreous lacunae and extensive vitreous liquefaction (Stirpe and Heimann, 1996). The multivariate adjusted OR of idiopathic rhegmatogenous retinal detachment in myopic adults was 7.8 (95% CI 5.0, 12.3) in white patients aged 21–80 years in the Eye disease Case–Control Study (The Eye Disease Case–Control Study Group, 1993). However, the prevalence of retinal tears, holes and detachment did not increase with longer AL (Pierro et al., 1992).

The evidence from the available literature assessing chorioretinal abnormalities is not strong because there are no large studies that delineate in a prospective fashion the association between refractive error or AL. Possible myopic pathology includes Fuch's spot, lacquer cracks, lattice degeneration, and retinal breaks. Fuch's spots and lacquer cracks are entities often defined clinically as occurring in the setting of pathological myopia. Choroidal neovascularization and macular holes in myopic adults have been well characterized clinically in case studies, but not cross-sectional, case–control or cohort studies (Avila et al., 1984; Stirpe and Michels, 1990).

Myopia and optic disc abnormalities

The risks of optic disc abnormalities in adults with myopia and high myopia were evaluated in East Asian university-based and white population-based studies (Rotterdam Study and Blue Mountains Eye Study) (Table 5) (Hyung et al., 1992; Chihara and Chihara, 1994; Ramrattan et al., 1999; Vongphanit et al., 2002). In all four studies described in Table 5, disc abnormalities were assessed using stereoscopic photographs. In the Japanese study (n = 210) by Chihara and Chihara (1994), myopic patients had significantly higher rates of larger, tilted, rotated discs, larger disc areas, longer disc-foveola distances, and larger long:short axis ratios. In the Blue Mountains Eye Study (n = 3,583; aged 49 years or older), subjects with tilted discs were more likely to be myopic (66.2%), compared with subjects with no tilted discs (p < 0.001) (Vongphanit et al., 2002). Increasing severity of myopia was associated with an increase in prevalent optic disc abnormalities in the clinic-based Korean (n = 61) and the population-based Rotterdam study (n = 5,114) (Hyung et al., 1992; Ramrattan et al., 1999). In a study by Fulk and colleagues of 224 subjects aged 8–25 years, myopic refractive error was associated with large optic nerve crescents (Fulk et al., 1992). While myopic optic disc abnormalities may appear innocuous, they are important clinically because they can be difficult to evaluate. The detection of both glaucoma and progression of glaucomatous optic neuropathy may be delayed in these patients. There is also evidence that larger discs may be more susceptible to the effects of intra-ocular pressure and intra-ocular pressure-related stress (Bellezza et al., 2000).

Table 5.  Summary of published data on optic disc abnormalities as complications of myopia (for abbreviations see Table 1)
Author (Year) (Country) EthnicityStudy design Study population (n)Methodology and definitionsResults
Hyung et al. (1992) (Korea) Koreans ECross-sectional studyMale Koreans, eye department, Seoul National University Hospital with IOP < 21 mm Hg and astigmatism (cylinder <1.5 D) (n = 61; 109 eyes)Cycloplegic streak retinoscopy, stereoscopic photographsWith increasing myopia, the temporal slope of the disc cup decreased (p < 0.01), the ratio of vertical to horizontal disc diameter (p < 0.01) and ratio of width of peripapillary atrophy to vertical disc diameter increased (p < 0.01)
Chihara and Chihara (1994) (Japan) JapaneseCross-sectional studyHealthy Japanese at Kyoto University (n = 210)Stereoscopic photographsPatients with myopia (SphE at least -5.0 D) had a longer disc–foveola distance (p < 0.001), larger long:short axis ratio (p < 0.001), larger discs (p < 0.01), higher likelihood of tilted (p < 0.001), rotated disc (p < 0.01)
Ramrattan et al. (1999) (The Netherlands) WhitePopulation-based cross-sectional studyRotterdam study 55 years and older (n = 5114)AR + SR, Stereoscopic photographsThe disc area increased by 0.033 mm2 (95% CI 0.027, 0.038), the neural rim area by 0.029 mm2 (95% CI 0.025, 0.034), and the prevalence of parapapillary atrophy [zone alpha by 0.4% (95% CI 0.03%, 0.8%), and zone beta by 1.3% (95% CI 0.57, 1.9%)] for each D increase towards myopia
Vongphanit et al. (2002) (Australia) WhitePopulation-based cross-sectional study49 years or older in the Blue Mountains, West Sydney (n = 3583)SR, Stereoscopic photographs. Tilted disc defined as inferior or nasal tilting of the optic discIn eyes with tilted discs (77 eyes), 66.2% were myopic (SphE at least –1.0 D), but in eyes without tilted discs (7,089 eyes), 11.3% were myopic (p < 0.001)

Myopia and age-related macular degeneration

Three recent population-based studies: the Blue Mountains Eye Study (3654 white subjects aged 49 years or older; cross-sectional study), Beaver Dam Eye Study (3684 white subjects aged 43–86 years; cohort study), and National Health and Nutrition Examination Survey (less than 10 000 white subjects aged 45 year or older; cross-sectional study) have evaluated the association of myopia and AMD (Goldberg et al., 1988; Wang et al., 1998; Wong et al., 2002). No significant associations of myopia with prevalent or incident AMD were found.

Pathological myopia as a cause of visual impairment

Pathological myopia is a potentially blinding disease: cataract and glaucoma are leading causes of blindness and visual impairment. Myopic degeneration is also a significant cause of visual impairment worldwide. In the Rotterdam study of 6775 subjects aged 55 years and older, myopic degeneration was the predominant cause of impaired vision (accounting for 23.0% in adults younger than 75 years) (Klaver et al., 1998). There were 12 cases of impaired vision younger than 75 years and eight cases of myopic degeneration due to impaired vision in adults younger than 75 years and only 11 cases of myopic degeneration in total. Thus, the total number of adults affected is small. A study of 1000 inhabitants aged 60–80 years in Copenhagen revealed that myopic macular degeneration was one of the most common causes of bilateral blindness (accounting for 10%) (Buch et al., 2001). Again, the total number of cases of myopic macular degeneration due to blindness was only two and thus the overall impact of myopic degeneration may not be large. On the other hand, in a survey of adults 50 years or older in Taiwan, myopic macular degeneration was the second most common cause of visual impairment (contributing 25.0% of cases) (Liu et al., 2001). Similarly, a survey by Iwano and colleagues in Japan of 2263 adults aged 40–79 years showed that the OR of visual impairment for myopic adults was 2.9 (95% CI 1.4, 6.0) (Iwano et al., 2004). In Asian countries, myopia may be a leading cause of visual impairment because of the higher overall rates of myopia and high myopia. Further large surveys of the contribution of myopia to visual impairment in different populations should be conducted. Myopia may also be under corrected if the lens prescription is not updated or myopia remains undiagnosed. Under corrected myopia is also one of the leading causes of visual impairment and is easily correctable.

Conclusion

The impact of myopia, an apparently benign ocular disease, may be larger than it seems. A greater understanding of the potentially blinding risks of myopia by ophthalmologists and optometrists may facilitate the screening and management of myopia-related ocular complications. Severe myopia may be associated with increased risks of ocular complications and the prevention of the onset of myopia or the progression of low myopia to high myopia is of utmost importance.

Cohort studies have shown that cataract (PSC, nuclear and occasionally, cortical cataract) is a complication of myopia (Lim et al., 1999; Wong et al., 2001; Leske et al., 2002). A survey of the findings of the literature in European-derived and Asian populations suggests that myopic adults have an increased prevalent risk of glaucoma, although no cohort studies depicting the temporal relationship between myopia at baseline and incident glaucoma risks are available (Daubs and Crick, 1981; Ponte et al., 1994; Mitchell et al., 1999; Grodum et al., 2001; Yoshida et al., 2001; Wong et al., 2003). In clinic-based studies, chorioretinal abnormalities such as lacquer cracks and lattice degeneration have been associated with refractive error or AL (Hyams and Neumann, 1969; Curtin and Karlin, 1970; Karlin and Curtin, 1976; Celorio and Pruett, 1991; Pierro et al., 1992; The Eye Disease Case–Control Study Group, 1993; Yura, 1998). Our synthesis of the literature in European-derived and Asian populations also shows higher prevalence of optic disc abnormalities such as tilted disc in the myopic population (Hyung et al., 1992; Chihara and Chihara, 1994; Ramrattan et al., 1999; Vongphanit et al., 2002). However, data from well-designed cohort studies showed no increased risks of AMD with myopia (Wang et al., 1998; Wong et al., 2002). Several studies show that the risks of cataract, glaucoma, and chorioretinal abnormalies increase with increasing myopic refraction or axial length (Curtin and Karlin, 1970; Karlin and Curtin, 1976; Pierro et al., 1992; Lim et al., 1999; Mitchell et al., 1999; Grodum et al., 2001).

Risk estimates (multivariate RRs) from several large population-based cohort studies demonstrated strong evidence of associations of myopia and ocular pathology such as cataract (Lim et al., 1999; Wong et al., 2001; Leske et al., 2002). The inferences from other studies in the literature are limited because of the cross-sectional nature of the studies, small sample sizes, or unadjusted risk estimates. Ascertainment bias may be present in clinic-based studies because patients who attend the clinic for their myopia may be more likely to have other ocular pathologies. This may lead to spurious associations between glaucoma and myopia, or cataract and myopia. This bias is worse for diseases that are largely asymptomatic. The current best available evidence cannot conclusively support the hypothesis that myopia causes ocular pathology. Only suggestions of possible associations may be inferred. There is also the issue of publication bias. Even if a hypothesis does not have a strong biological plausibility, it may be possible that positive results are reported. Conversely, negative associations from other possibly well-designed studies are often not published.

Several important questions remain unanswered: Are both low myopia and high myopia associated with higher risks of ocular complications? Do the risks of specific ocular pathology increase with more severe myopia or greater AL? Is the pattern of pathological myopia different in Asian and non-Asian countries? Will the rates of pathological myopia rise over the next few decades?

The following issues have not been well addressed thus far. The hereditary susceptibility to pathological myopia in Asian and non-Asian countries may differ. In addition, the public health burden of pathological myopia may be unique in Asian countries. Thus, comparisons of pathological complications in Asian and non-Asian countries would be of utmost clinical importance. The appropriate cutoff limit for high and malignant myopia is unclear; correlations of different levels of severity of myopia and specific pathologies need to be further studied before clinical guidelines can be developed. Perhaps even low degrees of myopia may be associated with significant risks of ocular complications. The evaluation of contributions to visual loss and optimal treatment of associated chorioretinal lesions will assist in the assessment of ocular morbidity contributed by each anomaly. Other myopia-associated pathologies such as myopia-associated cataract may also contribute significantly to visual impairment.

In conclusion, patients with myopia, especially high myopia, may have higher risks of cataract, glaucoma, and chorioretinal abnormalities such as retinal detachment and optic disc abnormalities. Pathological myopia is one of the most common causes of visual impairment and blindness. The early detection and management of degenerative eye diseases is of utmost importance in the care of myopic adults. Additional data from different ethnic populations are needed to describe the effects of varying severity of refractive error and AL on the incident risks of specific ocular pathological complications. Currently, there are many challenges for myopia research. The accurate prediction of the burden of ocular pathology in myopic children and adults has important clinical implications and should be a priority research area for eye care professionals.

Ancillary