From the past to the present, optical coherence tomography in glaucoma: a practical guide to a common disease

Fundoscopic Examination

The fundoscopic examination in the assessment of the optic nerve head (ONH) is based on the evaluation of the neuroretinal rim (NRR), which is an area between the optic disc margin (DM) and cup edge.²³ The characteristic ONH changes in glaucoma include a focal or diffuse NRR thinning, specifically in the superior or inferior quadrant of the optic disc with an enlargement of the ONH cup to disc ratio > 0.5.²⁴ The proposal by Jonas et al.,²⁵ the inferior > superior > nasal > temporal (ISNT) rule, states that the thickness of the NRR decreases in the previously mentioned order, and that the NRR in glaucomatous optic discs disturb this relationship. However, several papers also deny the usefulness of ISNT assessment.²⁶

Visual field

An important examination for detecting and monitoring functional impairment of the optic nerve in glaucoma is the assessment of VF using static automated perimetry (SAP). VF abnormalities are associated with loss of RGCs and their axons. The glaucomatous visual field defect can the diagnosed using the Hoddap–Parrish–Anderson criteria,²⁴ which include the following:

Glaucoma diagnosis should be performed in the case of dissymmetry of VF between the eyes. The characteristic defects in VF in glaucoma mainly include narrowing of the peripheral VF, temporal island of VF, temporal wedge, Roenne’s nasal step, and different types of scotomas (paracentral, ring-shaped, Seidel’s, and Bjerrum’s).²⁴

VF examination is highly subjective depending on patient compliance and concentration. The quality of the exam is determined by different factors, such as the patient noncompliance and fatigue especially in elderly patients. In addition to artefacts, such as blepharoptosis, lens defects and refractive errors also decrease the quality of the exam.²⁷ The perimetric examination should be performed on the same device and with the same protocol (e.g., 10-2, 24-2, 30-3) at least three times a year in the first year after the diagnosis – this increases the reliability of the results.¹ The result maps out the patients’ VF, which can be helpful in assessing the baseline and tracking disease progression. In the case of a ganglion cell loss of less than 40–50%, there are no reliable VF defects in threshold perimetry, thus, the concept of preperimetric glaucoma was proposed as typical glaucomatous eye changes without the presence of VF defects detected by a perimetric examination.²⁴ Nowadays, the availability of a new technology, such as OCT-based fundus imaging, improves the assessment of preperimetic glaucoma progression.²⁸

OCT

The OCT technique is superior because the automated perimetry examination can lack reproducibility and has some degree of non-objectiveness. Moreover, the OCT examination shows evidence of structural changes, including ONH damage and RNFL thinning, before functional loss is detected by standard automated perimetry. Furthermore, OCT is reliable when performed on both normal and glaucoma patients, and shows superiority in monitoring patients who are unable to undergo VF testing, including children, the elderly, and those with dementia.²⁷ In ophthalmology, the OCT technique is a basic tool in daily clinical practice.²⁹ Hence, its superior resolution renders it as a reliable tool for retinal morphology assessment and the measurement of retinal thickness, which have become important in evaluating patients with wet age-related macular degeneration.³⁰ Retinal thickening is also frequent in patients with type 1 or 2 diabetic retinopathy, approximately 15% of whom develop macular oedema, where OCT can also be one of the first-choice diagnostic tools. The third and leading reason for OCT evaluation is patients with glaucoma, which causes RNFL atrophy and visual loss in advanced stages, is that OCT evaluation can detect RNFL thinning at an early stage, often before a patient’s visual loss, leading to early-stage glaucoma detection. Additionally, OCT is used in nearly all fields of ophthalmological disorders, such as monitoring and follow-up for eye surgery, macular holes, vascular occlusions, and examination of pathologies in the anterior segment of the eye.³¹

pRFNL area

Thus far, the most popular OCT parameter in assessing glaucomatous eye changes is the pRFNL. pRNFL thinning in OCT is often one of the first signs of glaucoma, and it can be detected even before changes in the VF are observed.³⁹ In the diagnosis and assessment of glaucoma progression, the average pRNFL thickness, especially the pRNFL thickness in the inferior and superior quadrants, should be considered. Thinning of this parameter can have a significant impact on early glaucoma diagnosis, when the VF parameters are normal. The OCT analyses provide the RNFL thickness curve adjusted to the normative databases where green, yellow, and red colours are assigned to normal, borderline, and abnormal values, respectively (RNFL thickness values under the 99th percentile of normal database) (Figure 1). Early-stage glaucoma known as ‘green disease’ occurs when green labelling of OCT parameters indicates normal values despite the presence of glaucoma or progressive glaucomatous damage. Thus, in some cases the RNFL parameter must be still monitored even when the curve is on the green area. One must especially consider the fact that an asymmetrical thinning, comparing both eyes, with a value greater than 9 μm in the average pRNFL thickness, can suggest early glaucomatous degeneration.^40,41 Some cases of reduced RNFL thickness with red labelling of OCT parameters, despite lacking glaucomatous changes, are considered as ‘red disease’, which can occur in cases of myopia. When monitoring for glaucomatous progression using OCT, the physician must be aware of the accompanying and the abovementioned age- and software-related artefacts and the floor effect. Age-related pRNFL thinning rate, which occurs in healthy eyes at a mean rate of −0.48 μm/year to −0.60 μm/year depending on the OCT device, must also be considered. Software-related artefacts also play a significant role in the accurate assessment of glaucomatous eye damage and/or progression; most of them are operator-dependent and include displacement of optic disk boundaries, myopic eyes, and/or significant media opacities, resulting in erroneous pRNFL values.⁴² Glaucomatous progression must be suspected when there is a global decrease of ≥5 μm in the average pRNFL thickness. Another important factor of progression is a decrease of ≥7–8 μm in the sectoral pRNFL thickness. ‘Floor effect’ occurs when the pRNFL values decrease to a point where they are undetectable by OCT. Notably, the average pRNFL floor value ranges from 49.2 μm to 64.7 μm depending on the OCT device. This happens first in the pRFNL because the papillomacular region is more resistant to glaucomatous damage than the pRNFL; accordingly, measuring the thickness of the papillomacular structure is of great significance in the evaluation of glaucoma in its advanced stages. Thus, it is crucial to understand that the floor effect is achieved earlier within the pRNFL, and later in the macular region. In such cases, we should consider the use of macular OCT and VF 10-2 testing.⁴³ Despite the pRFNL being a useful and the most commonly used tool in diagnosing glaucomatous eye changes in cases of high myopia, small or tilted ONHs, and swelling of ONH, the measurement result can be unreliable; in these cases, the macular area parameters should be used as reference points.⁴⁴

Figure 1. Analysis of peripapillary nerve fiber layer thickness; a) in a non-glaucomatous emmetropic eye; b) in a glaucomatous emmetropic eye; c) in a glaucomatous myopic eye.

The retinal nerve fiber layer is situated between the internal limiting membrane (red line) and the boundary separating the retinal nerve fiber layer from the ganglion cell layer (green line).

Images: Department of Ophthalmology, Medical University of Bialystok Clinical Hospital.

Macular area

Early-stage glaucoma can cause thinning of the macula area, which consists of macular RFNL (mRFNL) and ganglion cell layer with inner plexiform layer (GCIPL); GCIPL contains RGC bodies and RGC dendrites, which forms the ganglion cell complex (GCC = mRFNL + GCIPL). Assessing the macular region in this case is particularly important because it has the highest concentration of RGCs in the retina (approximately 50% of the RGCs of the entire retina).³ During OCT evaluation of the macular area, the mRNFL thickness, GCC thickness, GCIPL minimum thickness, and GCIPL average thickness must be considered. When assessing these parameters, epidemiological factors, such as age and sex, must be considered. Additionally, central corneal thickness, axial length, and macular GCIPL attrition with aging, which occurs similarly to mRNFL at a rate of approximately −0.31 μm/year, are important and must be considered. In the assessment of the macular region, depending on the OCT device that is used, one can focus on assessing the GCC complex as a whole parameter or the GCIPL. The GCC thickness values range between 95.08 ± 7.88 μm in normal eyes, 83.30 ± 9.27 μm in early stage perimetric glaucomatous eyes, and 80.13 ± 9.60 μm in the moderate stage, and 75.08 ± 11.79 μm in the advanced stage.⁴⁵ When assessing glaucomatous eye changes, one must be aware that several studies suggest that in early glaucoma the GCC parameter is superior to the pRFNL,⁴⁶ but there is no clear consensus regarding this statement.⁴⁷ In establishing the diagnosis, it is crucial to inspect the average GCC thickness, as it is more reliable than the pRNFL in early stage glaucoma and preperimetric glaucoma. The average and inferior GCC thicknesses can be correlated with progressive visual field loss. The examination has greater diagnostic value with more advanced stages of the glaucomatous neuropathy when the signal strength values are higher. Moreover, it is unaffected by the increase in axial length, which results in more specific discrimination of glaucomatous changes in myopic eyes compared to that of RNFL thickness.⁴⁶ In advanced glaucoma, the early floor effect is achieved when assessing the pRNFL; thus, this parameter is excluded from OCT assessment in advanced stages. In such cases, one should consider a combination of OCT assessment of the macular region and VF 10-2 testing. The static automated perimetry offers various test protocols (10-2, 24-2, 30-2); the test points are localised approximately 6° from each other in the 24-2 and 30-2 protocols.^48,49 Due to the GCC thickness, which is strongly correlated with retinal sensitivity within a central range of 10° of the macula, the VF 10-2 should be considered as the main examination in determining the rate of progression in advanced glaucoma. Moreover, glaucomatous damage in advanced stages, when the RGC bodies can be displaced from their receptive fields of the macular area, can lead to false VF testing results. This can result in failure to assess the relationship between the defects and RGC damage; hence, in this case, a 10-2 test protocol with a 2° grid should be considered as the better choice as it displays more accurate results.⁵⁰ When assessing the GCIPL, Xu et al.⁵¹ stated that the average GCIPL thickness is approximately 75.2 ± 6.8 μm in early glaucoma, which thins to 64.4 ± 8.4 μm in moderate glaucoma, and then to 55.6 ± 7.6 μm in advanced glaucoma. Furthermore, an average GCIPL thickness change of >4 μm is suggestive of glaucomatous progression. When monitoring progression, one must be aware that the macular thickness is visible as an arcuate defect on the thickness and progression-change maps. Although appearing later, an important parameter to consider is the floor effect, which is defined as a point at which no further structural loss can be detected. As Chhavi et al. stated, several studies have shown that in advanced stages of glaucomatous damage, when mRNFL thickness is <55 μm, the GCIPL thickness change still correlates with the damage measured using a 10-2 visual field; as mentioned above, the floor effect occurs when the average thickness of macular GCIPL is approximately 45 μm.³⁹ Several vendors provide different OCT devices, and not every OCT device can segment the GCIPL complex. Due to these limitations, some studies focus on assessing the GCC as a whole, whereas some assess the GCIPL complex as a solitary parameter. Despite this, the assessment of the GCIPL can be fraught with high error risk due to several coexisting factors, which warrants the consideration of the presence of an epiretinal membrane, areas of vitreous adhesion, myopic eyes, macular schisis, and drusen.^52,53

The above-mentioned groups of parameters comprise the basis of glaucomatous damage assessment. Importantly, in the clinical assessment of glaucoma progression, the expected results may vary due to a series of artefacts that can occur in the diagnostic process, such as the segmentation error artefact; decentration, especially in myopic eyes; and accompanying pathological changes in the eye, such as media opacities in the form of corneal haze, cataracts, vitreous debris, myelinated RNFL, epiretinal membrane, and swelling of the ONH; these artefacts may falsely decrease or increase RNFL measurements.³⁹

Optical Coherence Tomography Angiography in Glaucoma

Although the complete understanding of glaucoma pathogenesis remains a challenge, a prevailing theory posits that vascular dysfunction within the ONH may indeed play a pivotal role in the development and progression of glaucoma. A noninvasive method, such as OCT-angiography (OCTA), enables the assessment of perfusion in targeted layers of the retina and optic nerve head, providing high resolution and replicability. OCTA techniques elevate flow signals through the assessment of reflectance signal amplitude across successive scans comparing the static to dynamic tissues. This enables the identification of dynamic structures, such as red blood cells within the vasculature, depicted as bright pixels, while regions of static reflectance, resembling constant tissue, are visualized as dark pixels. Therefore, unlike traditional fluorescein or indocyanine green angiography, OCTA examination allows for the visualization of retinal vasculature without the need for dye injection and facilitates the assessment of vascular networks across distinct retinal layers.

In glaucoma, the OCT angiography parameters predominantly utilized for evaluating ocular vascular flow are vessel density (VD) and flow index (FI). Decreased FI and VD in the vascular network of peripapillary retina and ONH have been observed in both preperimetric and perimetric POAG according to various studies.^54–56 VD is less affected by the floor effect in comparison with metrics assessing retinal thickness, making VD a valuable parameter in the advanced stages of glaucomatous progression.

Additionally, to that it is worth mentioning that some studies propose the possibility of identifying changes in OCTA metrics in specific sectors before deviations in RNFL thickness are identified.

The main limitations of OCTA include segmentation artifacts, typically associated with patient movement and a lack of fixation, reduced optical transparency, such as cataracts and vitreous floaters, and the absence of a normative database for clinical assessment and classification. Moreover, elevated intraocular pressure reduces the actual vascular density measurements.

However, more advanced OCTA technologies enable a reduction in the number of artifacts associated with patient non-compliance and improve the segmentation of retinal layers, providing a more accurate visualization of the vascular network.^3,57

OCTA is an immensely crucial diagnostic method. However, owing to its extensive range of applications, the continual development of new modalities within this technique, and the inherent challenges of the technology in glaucoma diagnosis, potentially resulting in varied diagnostic approaches, further comprehensive research in this field is necessary.