The sine qua non of glaucoma is the progressive loss of retinal ganglion cells and their axons which comprise the ganglion cell layer, retinal nerve fiber layer, and optic nerve. This structural damage respects the horizontal raphe, affects retinal ganglion cells across the retina, alters the morphology of the neuroretinal rim, and is often associated with spatially consistent functional impairment. In clinical practice, individuals thought to be at increased risk of glaucoma or who might have early disease but do not meet a clinical threshold for diagnosis are typically called “glaucoma suspects” or to have a “glaucoma-like” disc. Oftentimes, such a diagnosis reflects the presence of various systemic (eg, older age, genetic susceptibility, ancestry, low blood pressure, etc.) and/or ocular (eg, ocular hypertension, thinner central corneal thickness, exfoliation syndrome, etc.) risk factors. Glaucoma suspects are asked to return periodically for longitudinal examinations to determine if treatment initiation or escalation is warranted. In the United States, the number of individuals with a glaucoma suspect diagnosis substantially exceeds the number of patients with actual glaucoma, ranging from 4.5% of Medicare beneficiaries over age 651 to as high as 20% among high-risk populations.2 In Europe, nearly a quarter of the referrals for evaluation of possible glaucoma are estimated to be due to a suspicious disc, only half of which are eventually confirmed to have glaucoma.3 This is a burgeoning and burdensome subpopulation of individuals without clear-cut disease who nonetheless must remain under surveillance. Other than data from the Ocular Hypertension Treatment Study, it remains far from certain at what rate each of the glaucoma suspect subcategories convert to actual progressive neurodegeneration. We believe that a renewed focus on the diagnostic implications surrounding the use of the term glaucoma suspect, based solely on the appearance of the optic disc, is critical. In contrast to all other risk factors that connote a risk of developing glaucoma, a glaucoma-like disc suggests that the patient might already have experienced or is experiencing early glaucomatous neurodegeneration. Three common clinical reasons for this categorization are an enlarged cup-to-disc ratio, cup-to-disc ratio asymmetry between the eyes, and an atypical neuroretinal rim appearance (eg, rim sloping, excavation, notching, etc.). Unfortunately, the limitations of the clinical examination and the wide variability in size, shape, and insertion of the normal optic nerve makes these assignations highly subjective, resulting in poor interobserver reproducibility.4 Given the cross-sectional nature of the initial patient encounter, there is usually no prior available baseline documentation of disc appearance to determine whether it has changed over time. This often leaves clinicians wondering whether the disc has progressed from an earlier, healthier state, which contribute to uncertainty and proliferation of glaucoma suspect diagnoses. This tendency is further compounded by the ophthalmologist’s preference for high sensitivity at the expense of specificity so as not to miss disease.
Because of the blurred clinical distinction between the suspicious, glaucoma-like disc and the pathognomonic clinical features of the disease, the differentiation between the glaucomatous disc and the glaucoma-like disc remains an important clinical conundrum. Valuable health care resources are applied to surveillance of this condition, often with low yield. For these reasons, we believe it would be beneficial to modify the way we evaluate glaucoma suspects and that the use of the term glaucoma suspect based solely on the appearance of the optic disc should be curtailed.
HOW CAN THIS BE ACCOMPLISHED?
A glaucoma suspect diagnosis implies that the clinician lacks sufficient information or evaluation skills to confidently determine whether glaucomatous damage is present or absent. Historically, this has been due to the subjective nature of optic disc evaluation and the lack of a standardized approach. Many attempts have been made by glaucomatologists to make this process more effective. Optic Reading Centers in prospective, randomized clinical trials have enhanced study design and progression endpoint determination. Methodological approaches for clinical optic nerve evaluation such as the FORGE I (Focusing Ophthalmologists on Reframing Glaucoma Evaluation) didactic program5 or the definition proposed by the International Society of Geographical and Epidemiological Ophthalmology (ISGEO)6 attempt to standardize the definition of the glaucomatous disc. However, clinical evaluation of the optic disc for glaucoma, a fundamental aspect of the eye examination, remains largely subjective even among glaucoma experts. High-resolution optical coherence tomography (OCT) imaging data, which are now readily available to most practitioners in developed countries, offers further opportunities to establish a glaucoma diagnosis.
There is widespread agreement that standardized, systematic use of the information in OCT images can help in disease detection. At the most basic level, a healthy retinal nerve fiber layer (RNFL) and ganglion cell layer (GCL) in the presence of a glaucoma-like disc reassures most practitioners that the eye is healthy and not glaucomatous. To enhance specificity, Quigley and colleagues suggested the use of quadrantic OCT and visual field (VF) metrics to establish a diagnosis for clinical research.7 We have previously reported the use of an objective, evaluative approach to the RNFL and GCL to separate glaucomatous from nonglaucomatous eyes that utilizes the localized nature of glaucomatous damage to the optic nerve that functions better than the use of global metrics for image analysis.8 Using this method, we were able to detect eyes with moderate and severe VF loss with extremely high sensitivity and early VF with excellent sensitivity, while maintaining high specificity.9 This approach reclassified many “glaucoma suspects” as normal and others as preperimetric glaucomatous optic nerve disease, allowing more appropriate scheduling of future surveillance and a more judicious use of resources. We suggest that this OCT-based methodology, utilizing localized optic nerve structural damage and structure-structure corroboration (ie, corroborative GCL and RNFL deficits), easily applied in the clinical setting, be utilized by practitioners to reclassify glaucoma-like discs into healthy, glauomatous optic neuropathy, and “uncertain” categories. The eyes that remain “uncertain” would be the only remaining eyes with a glaucoma suspect diagnosis. An artificial intelligence algorithm, especially with standardization of images,10,11 might also be able improve diagnostic accuracy, but the technology remains nascent. Our OCT-based approach, or similar other methodologies, bridge the gap between the present day and the future arrival of artificial intelligence-assisted decision making.12–14
In summary, the use of the term “glaucoma suspect” creates a difficult predicament for both the patient and health care systems and is counterproductive. Patients suffer from the inconvenience, psychological impact and financial implications of a glaucoma suspect diagnosis and health care systems need to cope with provider, technology and cost constraints. Thus, we need to more rapidly integrate new information and technologies into practice applications for the diagnosis of the suspicious disc with the goal of minimizing diagnostic uncertainty, rather than relying on definitions created when these technologies did not exist. The only alternative is to continue to spend valuable health care resources, at considerable human burden and cost, to collect longitudinal data looking for change in individuals who are likely at minimal or no risk of developing disease.
1. Cassard SD, Quigley HA, Gower EW, et al. Regional variations and trends in the prevalence of diagnosed glaucoma in the medicare population. Ophthalmology. 2012;119:1342–1351.
2. Waisbourd M, Pruzan NL, Johnson D, et al. The Philadelphia Glaucoma Detection and Treatment Project. Ophthalmology. 2016;123:1667–1674.
3. Founti P, Topouzis F, Holló G, et al. Prospective study of glaucoma referrals across Europe: are we using resources wisely? Br J Ophthalmol. 2018;102:329–337.
4. Jampel HD, Friedman D, Quigley H, et al. Agreement among glaucoma specialists in assessing progressive disc changes from photographs in open-angle glaucoma patients. Am J Ophthalmol. 2009;147:39–44.e1.
5. Fingeret M, Medeiros FA, Susanna R, et al. Five rules to evaluate the optic disc and retinal nerve fiber layer for glaucoma. Optom J Am Optom Assoc. 2005;76:661–668.
6. Foster PJ. The definition and classification of glaucoma in prevalence surveys. Br J Ophthalmol. 2002;86:238–242.
7. Iyer J, Vianna JR, Chauhan BC, et al. Toward a new definition of glaucomatous optic neuropathy for clinical research. Curr Opin Ophthalmol. 2020;31:85–90.
8. Liebmann JM, Hood DC, de Moraes CG, et al. Rationale and development of an OCT-based method for detection of glaucomatous optic neuropathy. J Glaucoma. 2022;31:375–381.
9. Hood DC, La Bruna S, Tsamis E, et al. Detecting glaucoma with only OCT: implications for the clinic, research, screening, and AI development. Prog Retin Eye Res. 2022:101052. doi:10.1016/j.preteyeres.2022.101052
10. Ting DSW, Wong TY, Park KH, et al. Ocular imaging standardization for artificial intelligence applications in ophthalmology: the Joint Position Statement and Recommendations from the Asia-Pacific Academy of Ophthalmology and the Asia-Pacific Ocular Imaging Society. Asia Pac J Ophthalmol (Phila). 2021;10:348–349.
11. Lee AY, Campbell JP, Hwang TS, et al. Recommendations for standardization of images in ophthalmology. Ophthalmology. 2021;128:969–970.
12. Ran An, Cheung CY. Deep learning-based optical coherence tomography and optical coherence tomography angiography image analysis: an updated summary. Asia Pac J Ophthalmol (Phila). 2021;10:253–260.
13. Ramessur R, Raja L, Kilduff CLS, et al. Impact and challenges of integrating artificial intelligence and telemedicine into clinical ophthalmology. Asia Pac J Ophthalmol (Phila). 2021;10:317–327.
14. Lee EB, Wang SY, Chang RT. Interpreting deep learning studies in glaucoma: unresolved challenges. Asia Pac J Ophthalmol (Phila). 2021;10:261–267.