Abstract
Myopia is rapidly escalating globally, especially in East and Southeast Asia, where its prevalence among younger populations reaches alarming levels of 80–90 %. This surge contributes to a myopia epidemic linked to several ocular complications, including glaucoma. As myopic individuals age, the risk of developing glaucoma increases, and an additional concern arises from the growing frequency of refractive surgeries among younger individuals, making precise optic nerve assessments critical before surgery. Evaluating the optic nerve head (ONH) in myopic eyes is challenging, as structural changes due to myopia often resemble glaucomatous alterations. Techniques such as optical coherence tomography (OCT) have improved the examination of ONH microstructures, but interpreting results remains complex due to potential false-positive findings. Myopic eyes exhibit unique changes, such as peripapillary atrophy and altered neuroretinal rim configurations, making it crucial to distinguish these from true glaucomatous signs. Recent advancements in OCT technology and the establishment of myopia-specific normative databases have enhanced diagnostic accuracy. Parameters such as minimum rim width, ganglion cell–inner plexiform layer thickness and temporal raphe sign show promise in differentiating between glaucomatous and nonglaucomatous changes. Ultimately, a comprehensive approach incorporating multiple OCT metrics is essential for accurately diagnosing glaucoma in myopic patients. By integrating various structural evaluations and leveraging advanced imaging techniques, clinicians can better navigate the complexities of glaucoma diagnosis amidst the challenges posed by myopia. This review highlights the need for increased attention and tailored strategies in managing glaucoma risk within this increasingly affected population.
1
Introduction
Myopia is becoming increasingly prevalent worldwide, particularly in developed countries in East and Southeast Asia. In these regions, the prevalence of myopia has reached 80–90 % among younger populations, contributing to a growing myopia epidemic. Myopia is associated with various ocular complications, including macular degeneration, retinal detachment, posterior staphyloma, choroidal neovascularization and glaucoma. As an independent risk factor for glaucoma, myopia poses a growing concern as the risk of glaucoma increases with age. Consequently, the burden of diagnosing glaucoma in myopic eyes is expected to rise as this young, myopic generation ages. Furthermore, with many myopic individuals undergoing refractive surgery, it has become increasingly important to accurately assess the optic nerve and diagnose glaucoma or predict its risk before the procedure.
However, assessing the optic nerve head (ONH) in myopia is challenging. The structural changes in myopic eyes can mimic glaucomatous ONH alterations. This compromises the diagnostic accuracy of conventional ophthalmoscopic tools for glaucoma diagnosis. Peripapillary atrophy (PPA) complicates the evaluation of the retinal nerve fiber layer (RNFL) in the peripapillary region. Additionally, the stretched optic disc and lamina cribrosa (LC) during myopic axial elongation led to flattening of the optic cup. This reduction in the height of the neuroretinal rim makes it difficult to distinguish glaucomatous changes. Highly myopic eyes can also exhibit various perimetric abnormalities that mimic glaucomatous visual field (VF) defects. Notably, glaucoma-like VF defects are present in 10.8 % of highly myopic eyes, with their prevalence increasing with longer axial length.
Optical coherence tomography (OCT) plays a crucial role in assessing the structural changes in myopic eyes. Technological advances have enabled the evaluation of deeper microstructures of the ONH and provided a wider scan area, including both the macula and optic disc. However, in myopic eyes, false-positive red flags can appear on OCT imaging, making glaucoma diagnosis challenging and requiring careful interpretation. It is essential to understand how structural changes in myopic eyes differ from those in nonmyopic eyes to better distinguish glaucomatous optic nerve changes. This review aims to explore and summarize key insights that can assist in the accurate diagnosis of glaucoma in myopic eyes.
2
Typical structural changes of ONH during myopic axial elongation
In nonmyopic eyes, the axons of retinal ganglion cells (RGCs) are bundled within the RNFL and pass through the optic disc via the sieve-like, multilayered porous structure of the LC. Each RNFL bundle follows its own pathway to the optic disc. The RNFL on the temporal side of the fovea travels to the optic disc in an arcuate shape around the fovea without crossing the horizontal meridian. This creates a temporal raphe characterized by relatively few axons. As a result of this anatomy, most glaucomatous changes present in an arcuate-to-crescent shape, located in the temporal area of the macula along the horizontal raphe. Typically, macular changes correspond with the respective RNFL defect and optic disc damage.
During axial elongation in myopic eyes, the macula shifts posteriorly. The ONH is stretched temporally in the direction of elongation. As a result, Bruch membrane opening (BMO) becomes stretched and enlarged temporally. The LC underneath also stretches toward the temporal side. This causes the LC to become thinner and leads to the formation of LC defects on the temporal side. The size and number of these defects are associated with optic disc tilt. Additionally, the presence of LC defects correlates with a reduction in peripapillary OCT vessel parameters in myopic eyes. LC defects are also associated with the presence of papillomacular bundle defects in glaucoma patients with high myopia.
The border tissue of Elschnig on the temporal side tilts temporally, changing its configuration from internally oblique to externally oblique. In contrast, the border tissue on the nasal side remains relatively stable. These changes in the temporal border tissue and the enlargement of the BMO contribute to the development of PPA. Meanwhile, the nasal peripapillary tissue elevates, leading to the thickening of the neuroretinal rim. The elevation of the nasal peripapillary tissue causes the optic disc to appear more tilted in the direction of elongation, increasing its ovality ( Fig. 1 ). The peripapillary superficial major vessels also stretch temporally, altering the course of the RNFL bundles that accompany these vessels.
In highly myopic eyes, greater elongation of the eyeball induces further enlargement of BMO. This results in an increase in the size of PPA, which can encircle the disc margin rather than being confined to the temporal side. The absence of Bruch membrane (BM) in the PPA leads to atrophy of the choriocapillaris, resulting in thinning of the peripapillary choroidal thickness. Additionally, this elongation contributes to the thinning of the LC and the peripapillary scleral flange that connects the inner half of the sclera to the LC.
3
Key considerations for RNFL thickness changes
Myopic eyes are reported to have thinner RNFL, with thickness negatively correlated with spherical equivalent and axial length. However, it is unclear whether the thinning is due to anatomical features or technical issues such as scan circle size and location in OCT imaging. OCT devices usually measure RNFL thickness at a fixed distance from the optic disc. For example, the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA, USA) uses a scan circle with a 1.73-mm radius. In eyes with an axial length greater than the standard 24.46 mm, the scan circle can enlarge due to the magnification effect. This results in measuring the RNFL further from the disc where it is naturally thinner, which could skew results. To address this, adjusted equations using the Littmann formula have been proposed. The Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany) uses a 3.5-mm scan circle and adjusts for ocular magnification by incorporating keratometry values. However, this adjustment is not perfect either, as optic disc size can also increase in myopic eyes. Therefore, both axial length and disc size should be considered when interpreting RNFL thickness in highly myopic eyes.
The location of the scan circle is crucial for RNFL measurements in myopia. With axial elongation, the temporal stretching of major vessels can alter the typical course of RNFL bundles. As myopia increases, the two peaks of RNFL thickness shift closer to the temporal quadrant. This shift may lead to false-positive red flags in the superior and inferior sectors, along with relatively thickened RNFL in the temporal sector ( Fig. 2 ). Chung and Yoo demonstrated that adjusting the scan circle to the temporal side of the optic disc, based on the neural canal opening, reduced the number of abnormal clock-hour sectors in myopic tilted discs. Therefore, it is important to carefully examine the RNFL thickness curve to determine whether the defect is due to the shifting of the RNFL peak or actual thinning of the RNFL in myopic eyes.
The normative databases built into commercially available OCT devices often lack sufficient data on moderate-to-highly myopic eyes. To reduce false-positive errors in these cases, a new normative database specifically for highly myopic eyes has been developed and evaluated for its diagnostic power in glaucoma detection. Biswas et al. created a normative database that included 180 highly myopic eyes (spherical equivalent of −6.0 D or less). They showed that using this myopic normative database improved specificity for detecting glaucomatous RNFL abnormalities in eyes with high myopia. Similarly, Seol et al. developed a normative database comprising 154 myopic eyes, ranging from mild to high myopia. Their work demonstrated improved specificity compared to conventional normative databases without compromising overall sensitivity for detecting glaucoma. However, while these customized databases enhance diagnostic accuracy, their application in busy clinical settings can be impractical. This underscores the need for a more comprehensive normative database that accommodates a wider range of refractions for RNFL thickness measurements.
4
Key considerations for neuroretinal rim evaluation
Evaluating the neuroretinal rim in tilted myopic discs can be more intuitive and less influenced by the temporal shift of major vessels and RNFL peaks. Minimum rim width (MRW), defined as the shortest distance between the BMO and the internal limiting membrane, has shown higher sensitivity than peripapillary RNFL thickness for diagnosing early glaucoma. In myopic eyes, MRW offers comparable sensitivity for glaucoma detection to RNFL thickness, with approximately 71.4 % sensitivity at 90 % specificity. Additionally, in nonglaucomatous myopic eyes, MRW exhibits fewer false-positive signs than RNFL thickness. Similarly, three-dimensional neuroretinal rim thickness, which measures the distance between the BMO and the vitreoretinal interface and reflects the minimum cross-sectional rim area in each direction, demonstrates significantly lower false-positive rates than RNFL thickness (2.1 % vs. 26.9 %, P < 0.001). This improvement is particularly notable in the superior and inferior regions, where false-positive RNFL findings are common in myopic eyes, providing better diagnostic accuracy for glaucoma in these patients ( Fig. 2 ).
However, evaluating the neuroretinal rim in highly myopic eyes can be challenging. The BM margin often shifts away from the temporal optic disc border, which can result in a misleadingly enlarged optic disc size and a falsely reduced neuroretinal rim thickness. Moreover, in approximately 30 % of highly myopic eyes, the BMO may be indiscernible in at least one meridian—typically in regions most prone to glaucomatous neuroretinal rim loss. As a result, BMO-based neuroretinal rim thickness evaluation may have limitations in highly myopic eyes.
5
Key considerations for macular parameters
The macula contains over 50 % of RGCs, making it advantageous for detecting glaucomatous RGC loss. Unlike peripapillary RNFL thickness, macular inner retinal thickness is generally less affected by the degree of myopia and myopia-related structural changes in the ONH. The most commonly used macular parameters are the measurements of inner retinal layer thickness in the macular region, including the RNFL, ganglion cell layer (GCL) and inner plexiform layer (IPL). Cirrus HD-OCT reports ganglion cell–inner plexiform layer (GCIPL) thickness, the sum of the thicknesses of the GCL and IPL. RTVue OCT (Optovue, Fremont, CA, USA) measures ganglion cell complex (GCC) thickness, which is the sum of the thicknesses of the RNFL, GCL and IPL. Spectralis OCT provides measurements for RNFL, GCL and IPL thickness, respectively.
Using Cirrus HD-OCT, the most effective parameters for distinguishing glaucomatous eyes from highly myopic eyes were inferior RNFL thickness (area under the receiver operating characteristic curve [AUROC] 0.906) and inferotemporal GCIPL thickness (AUROC 0.852). In contrast, for the non–highly myopic group, the optimal parameters were average RNFL thickness (AUROC 0.920) and minimum GCIPL thickness (AUROC 0.908). Another study assessed the diagnostic utility of inferotemporal GCIPL thickness in myopic preperimetric glaucoma, identifying it as the most reliable parameter, with significantly greater diagnostic power than average RNFL thickness, rim area and minimum GCIPL thickness. A recent study using swept-source OCT (DRI OCT, Topcon, Tokyo, Japan) demonstrated that inferotemporal GCL+ (equivalent to GCIPL thickness) and GCL++ (equivalent to GCC thickness) had the highest AUROC for detecting myopic glaucoma. In this study, macular GCL++ thickness (87.6 %) and GCL+ thickness (87.5 %) showed higher AUROC values compared to macular GCIPL thickness measured with spectral-domain OCT (Cirrus HD-OCT). Additionally, Jeong et al. recently demonstrated that inferotemporal GCIPL thickness offered the highest diagnostic accuracy for glaucoma in highly myopic eyes, outperforming other OCT parameters such as mean RNFL thickness, ONH rim area and the University of North Carolina OCT index.
Analyzing the asymmetry of GCIPL thickness across the horizontal raphe, often referred to as the “temporal raphe sign”, can be an excellent tool for detecting glaucomatous changes in myopic eyes ( Fig. 3 ). Similar to the glaucoma hemifield test of the Humphrey visual field, which is used to identify glaucomatous VF defects, this asymmetry can serve as an early indicator of glaucomatous changes. By revealing subtle differences in GCIPL thickness, the temporal raphe sign offers valuable diagnostic information that can aid in the timely detection of glaucoma in elderly patients with large optic disc cupping or optic neuropathy. Notably, the temporal raphe sign exhibits superior diagnostic capability for glaucoma in highly myopic eyes compared to other OCT parameters, highlighting its significance in clinical assessment ( Fig. 4 ).
