A, Baseline disc stereophotograph. B, Follow-up disc stereophotograph (after 6 years) demonstrating βPPA progression. In addition, there was substantial neuroretinal rim thinning during the same period. Note that the βPPA location correlated spatially with the longest distance to the central retinal vessel trunk in the lamina cribrosa, as well as the most marked loss of rim in the disc. C, Baseline corresponding 24-2 Swedish Interactive Thresholding Algorithm standard automated perimetry visual fields of this patient. D, Follow-up visual fields (after 6 years). Note the progression of a superior arcuate defect.
Customize your JAMA Network experience by selecting one or more topics from the list below.
De Moraes CG, Murphy JT, Kaplan CM, et al. β-Zone Parapapillary Atrophy and Rates of Glaucomatous Visual Field Progression: African Descent and Glaucoma Evaluation Study . JAMA Ophthalmol. 2017;135(6):617–623. doi:10.1001/jamaophthalmol.2017.1082
Does race affect the association between β-zone parapapillary atrophy and the velocity of visual field progression?
In a longitudinal cohort study of 634 patients (1090 eyes) with ocular hypertension or glaucoma followed up for a mean of 9 years, patients of European descent with glaucoma and β-zone parapapillary atrophy experienced faster visual field progression compared with those of African descent with β-zone parapapillary atrophy.
The results of this study suggest that the effect of β-zone parapapillary atrophy on the velocity of visual field progression depends on race and glaucoma status.
β-zone parapapillary atrophy (βPPA) has been reported as a risk factor for glaucoma onset and progression. Previous studies have shown that the prevalence of βPPA differs between individuals of African descent (AD) and European descent (ED).
To test whether the association between the presence and progression of βPPA vs visual field progression of glaucoma differs between these 2 ancestry groups.
Design, Setting, and Participants
In a prospective, multicenter, longitudinal cohort study, 634 individuals (1090 eyes) enrolled in the African Descent and Evaluation Study (ADAGES) with a diagnosis of glaucomatous optic neuropathy (GON) or ocular hypertension (OHT) and at least 2 disc stereophotographs were included. Two graders masked to clinical and ancestry data reviewed and graded the baseline and last disc stereophotographs for the presence of βPPA at baseline and βPPA progression (development or enlargement). Mixed-effects linear models were tested with visual field mean deviation as a dependent variable and time (alone and with interaction terms) as independent variables. ADAGES enrollment began in January 2003 and ended in July 2006; follow-up ended in 2016.
Main Outcomes and Measures
Progression of βPPA in AD and ED individuals.
In 634 patients, a total of 814 eyes of AD (395 eyes) and ED (419) patients with GON and 276 eyes of AD (106) and ED (170) patients with OHT who were enrolled in ADAGES were analyzed. There were 336 (53.0%) women in the study; mean (SD) age was 61.9 (12.7) years. In the OHT group, the association between βPPA at baseline and visual field progression was not significantly different between AD and ED eyes (β = 0.071; 95% CI, −0.016 to 0.158; P = .11), nor was the association between βPPA progression and visual field progression (β = 0.020; 95% CI, −0.465 to 0.506; P = .93). In the GON group, ED eyes with baseline βPPA progressed faster than did AD eyes with baseline βPPA (β = −0.124; 95% CI, −0.241 to −0.007; P = .04), although the association between βPPA progression and visual field progression did not differ significantly between race groups (β = −0.101; 95% CI, −0.323 to 0.119; P = .37).
Conclusions and Relevance
Race had a significant effect on the association between baseline βPPA and rates of visual field progression in eyes with GON. Progression of βPPA was not associated with faster visual field progression in either racial group.
Evaluation and management of care for patients with glaucoma typically include assessing risk factors, lowering intraocular pressure (IOP), and evaluating optic nerve structure and function. Among the structural factors, evaluation of the retinal nerve fiber layer, neuroretinal rim tissue, disc hemorrhages, and parapapillary atrophy (PPA) plays an important role when diagnosing the disease and monitoring progression.1
Parapapillary atrophy was initially characterized clinically and morphologically, consisting of 2 distinct morphologic components: a region of variable hyperpigmentation and hypopigmentation of the retinal pigment epithelium found within the α-zone PPA and an area adjacent to the optic disc characterized by atrophy of the retinal pigment epithelium, photoreceptors, and choriocapillaris, resulting in a distinctive appearance of sclera and choroidal vessels termed β-zone parapapillary atrophy (βPPA).2 With the advent of newer technologies, such as spectral-domain optical coherence tomography (SDOCT), additional characterizations have become possible, including the integrity of the retinal pigment epithelium and distance from the Bruch membrane opening.3
African descent (AD) connotes greater incidence, prevalence, and severity, as well as a worse prognosis for glaucoma compared with European descent (ED).4-7 The exact nature of why, how, and to what degree race contributes to progression of the disease remains unknown. The African Descent and Glaucoma Evaluation Study (ADAGES) is a prospective, multicenter, longitudinal study designed with the purpose of gathering a large amount of comprehensive clinical and genomics data to address this question and better understand the differences between these ancestry groups.8
Skaat et al9 recently reported that patients of AD have a greater prevalence of βPPA than do those of ED with glaucomatous optic neuropathy (GON). In the present longitudinal study, we examined (1) the prevalence of βPPA at baseline and its progression (ie, enlargement or development), (2) the association between the presence and progression of βPPA vs visual field progression, (3) how the above associations differ between patients with ocular hypertension (OHT) vs GON, and (4) how the above associations differ between patients of AD and ED.
The 3-site ADAGES collaboration (NCT00221923) includes the Hamilton Glaucoma Center at the Department of Ophthalmology, University of California, San Diego (data coordinating center), Edward S. Harkness Eye Institute at Columbia University Medical Center (site formerly located at New York Eye and Ear Infirmary), and the Department of Ophthalmology, University of Alabama, Birmingham. ADAGES enrollment began in January 2003 and ended in July 2006; follow-up ended in 2016. Participants of AD and ED were included in the cohorts. The institutional review boards at all sites approved the study methodology, which adhered to the tenets of the Declaration of Helsinki10 and to the Health Insurance Portability and Accountability Act. All participants gave written informed consent; there was no financial compensation.
Participants were asked to identify their race by self-report using the National Eye Institute inclusion/enrollment system describing ethnicity and race (https://grants.nih.gov/grants/guide/notice-files/NOT-OD-01-053.html). The ocular testing completed for ADAGES has been described elsewhere.8
All participants had open angles, a best-corrected visual acuity of 20/40 or better, and a refractive error of less than 5.0 diopters sphere and less than 3.0 diopters cylinder. At least 1 high-quality stereophotograph and 2 reliable standard automated perimetry Humphrey 24-2 field test results at baseline were required, defined as less than 33% false-positives, false-negatives, and fixation losses. Both eyes were included, except in cases where only 1 eye met the study criteria. All participants were older than 18 years. Participants with diabetes who had no evidence of retinopathy were included. Each participant underwent standard automated perimetry using a 24-2 program on the Humphrey Field Analyzer II with the Swedish Interactive Thresholding Algorithm, 33 version, 4.1 (Carl Zeiss Meditec Inc). Patients with GON (with or without visual field loss) or OHT were included in the present study. All included eyes had at least 2 good-quality optic disc stereophotographs performed at different visits.
Individuals were excluded if they had a history of intraocular surgery (except uncomplicated cataract surgery or glaucoma surgery); secondary causes of glaucoma; other systemic or ocular diseases known to affect the visual field; significant cognitive impairment; history of stroke, Alzheimer disease, or dementia; problems other than glaucoma affecting color vision, an inability to perform visual field examinations reliably, or a life-threatening disease that precluded retention in the study.
Glaucomatous optic neuropathy was defined as excavation, neuroretinal rim thinning or notching, localized or diffuse retinal nerve fiber layer defect, or vertical cup-disc ratio asymmetry of more than 0.2 between eyes (not explained by differences in disc size) based on masked grading of stereophotographs by 2 graders at the IDEA Reading Center (Hamilton Glaucoma Center). β-zone PPA was not considered a criterion for classification of GON. Disagreement regarding GON status was resolved by adjudication by a third experienced grader or by consensus. Only stereophotographs of adequate quality were used for evaluation. Ocular hypertension was defined if the patient had a history of elevated IOP (>22 mm Hg), absence of GON, and normal, repeatable visual field results at study entry. Normal standard automated perimetry visual fields were defined as a mean deviation (MD) and pattern standard deviation (PSD) within the 95% confidence limits and a glaucoma hemifield test result within the 99% normal limits.8
Participants from the ADAGES database with at least 2 color optic nerve stereophotographs of sufficient quality at separate visits were included in the present analysis to determine whether βPPA enlarged or remained stable between the first and last stereophotographs. Eyes with significant cataracts were not included in the study because they affected the stereophotograph quality and could significantly influence the analysis of visual field progression. Two experienced graders at Columbia University Medical Center Edward S. Harkness Eye Institute (J.T.M. and C.M.K.) were masked to participant diagnosis, GON grading results from the IDEA Reading Center, ancestry, and all other identifying characteristics. The stereophotographs were graded for the presence, absence, and progression of βPPA. The 2 graders assessed the stereophotographs independently and masked from each other’s assessments. β-zone PPA was defined as an area adjacent to the disc margin with notable atrophy of the retinal pigment epithelium, visible sclera, and visible large choroidal vessels. In cases where baseline stereophotographs did not show βPPA, progression was defined as development of βPPA that was not present at baseline. Cases of disagreement (22 of 1090 [2.1%]) were adjudicated by a third, experienced grader (J.M.L.).
We analyzed progression using trend analysis of the visual field MD. In brief, the slope measuring the rate of MD change over time was calculated using all reliable visual field data points performed between the baseline disc stereophotograph and final follow-up visit. Mixed-effects linear models (with random intercepts and random slopes) were used to compute these slopes.
Measures of center and dispersion are described as means (SDs). Mixed-effects linear models were tested with visual field MD as the dependent variable and time as the independent variable. Three-way interaction terms (race × baseline βPPA × time and race × βPPA progression × time) were included to test the hypothesis that the association between baseline βPPA and βPPA progression vs rates of visual field progression differed between ancestry groups. The following covariates were included in the model: age, baseline visual field MD, central corneal thickness, spherical equivalent, and mean follow-up IOP. One model was performed for patients with GON (with or without standard automated perimetry abnormalities) and a separate model for patients with OHT. Statistical analyses were performed using commercially available software (Stata, version 14; StataCorp LP). Statistical significance was defined at P < .05.
In 634 patients, a total of 814 eyes of AD (395 eyes) and ED (419) patients with GON and 276 eyes of AD (106) and ED (170) patients with OHT were analyzed. A total of 336 (53.0%) of the participants were women, and the mean (SD) age was 61.9 (12.7) years. The mean time between optic disc stereophotographs was 9.12 (3.8) years. Patient characteristics are summarized in Table 1.
In the entire sample of eyes (N = 1090), there was no significant difference between mean follow-up IOP and the presence of βPPA at baseline (β = −0.10; 95% CI, −0.40 to 0.20; P = .52), although eyes with βPPA progression were more likely to have lower mean follow-up IOP (β = −0.97; 95% CI, −1.78 to −0.15; P = .02). In addition, eyes with baseline βPPA were more myopic (β = −0.16; 95% CI, −0.30 to −0.02; P = .03), although there was no significant difference in spherical equivalent between eyes with and without βPPA progression (β = −0.07; 95% CI, −0.45 to 0.31; P = .72).
Among OHT eyes, the presence of βPPA at baseline was not significantly different between AD and ED eyes (odds ratio [OR], 1.17; 95% CI, 0.71 to 1.93; P = .53). There was also no significant difference in βPPA progression between the 2 groups (OR, 0.31; 95% CI, 0.03 to 2.72; P = .29). In the entire OHT group there was an association between the presence of βPPA at baseline and faster rates of visual field progression (β = −0.071; 95% CI, −0.140 to −0.003; P = .04). However, βPPA progression was not associated with faster visual field progression in this group (β = −0.039; 95% CI, −0.498 to 0.419; P = .86). In addition, there was no effect of race on the association between the presence of βPPA at baseline and visual field progression (β = 0.071; 95% CI, −0.016 to 0.158; P = .11) or between βPPA progression and visual field progression (β = 0.020; 95% CI, −0.465 to 0.506; P = .93) (Table 2).
Among GON eyes, βPPA at baseline was more common in AD than ED eyes (OR, 1.59; 95% CI, 1.16 to 2.18; P = .003). Notwithstanding, βPPA progression was less common in AD compared with ED eyes (OR, 0.51; 95% CI, 0.30 to 0.89; P = .02). In the entire GON group, the presence of βPPA at baseline was not associated with faster visual field progression (β = 0.040; 95% CI, −0.048 to 0.130; P = .37). Similarly, βPPA progression was not associated with faster progression in this group (β = 0.022; 95% CI, −0.163 to 0.208; P = .81). Nonetheless, the 3-way interaction term race × baseline βPPA × time revealed that ED eyes with baseline βPPA progressed faster than AD eyes with βPPA (β = −0.124; 95% CI, −0.241 to −0.007; P = .04). The 3-way interaction term race × βPPA progression × time was not statistically significant (β = −0.101; 95% CI, −0.323 to 0.119; P = .37), suggesting that race did not have an effect on the association between βPPA progression and rates of visual field progression (Table 3).
In this prospective cohort, we found that baseline βPPA was more common in AD than ED eyes with GON, although no significant difference between ancestry groups was seen among OHT eyes. Progression of βPPA was more common in ED than AD eyes with GON; however, no significant difference was seen among OHT eyes. In addition, when analyzing all GON eyes, neither the presence of βPPA at baseline nor the progression of βPPA was associated with faster visual field progression. However, the association between βPPA at baseline and visual field progression in GON eyes was influenced by race; that is, ED eyes with βPPA progressed faster than did AD eyes with βPPA. Different associations were seen in OHT eyes: the presence of βPPA at baseline (but not βPPA progression) was associated with faster visual field progression in this group as a whole, although race did not have an effect in this association. These findings have potential implications on how clinicians weigh the role of βPPA and its progression when assessing risk in clinical practice.
Longitudinal studies have investigated βPPA and its association with glaucoma and progression of the disease.11-17 Rockwood and Anderson13 measured the degree of loss of pigmentation of the retinal pigment epithelium over a 12-month period in patients with progressive and nonprogressive glaucoma. They reported that these changes were observed near the disc in 21% of patients with progressive glaucomatous cupping, although the changes also occurred as a natural phenomenon in 4% of eyes with nonprogressive glaucoma and in 3% of nonglaucomatous eyes. Budde and Jonas14 found that enlargement of βPPA was significantly more common in eyes with progressive vs nonprogressive glaucoma. However, owing to its low frequency, they suggested that enlargement of βPPA may not be a very useful marker for glaucoma progression. Jonas et al11 reported that a large area of βPPA is an important morphologic indicator for glaucoma progression when an end point was defined as loss of neuroretinal rim as detected by disc stereophotographs. While using visual fields as an end point, Teng et al16 demonstrated that both the presence and extent of βPPA at baseline are significant indicators of future faster visual field progression in eyes with established glaucoma. In a follow-up study, the authors also reported that the location with the largest PPA area was indicative of the hemifield undergoing the fastest progression.12 These findings were later confirmed using rates of retinal nerve fiber layer thinning as progression outcome.17
When investigating risk factors for the development of glaucomatous visual field loss in patients with OHT, Quigley et al15 found a significant indicative value of disc crescents (defined as any nonuniformity in the pigmentation of the peripapillary area that extended for >1 clock hour, including α and β zones). However, this association was no longer significant after adjusting for age, sex, and race. In a retrospective study, Tezel et al18 found that the sensitivity and specificity of βPPA progression on indicating conversion from OHT to glaucoma were 49% and 90%, respectively. More recently, the Ocular Hypertension Treatment Study19 evaluated data collected over a mean of 12 years comparing 279 eyes with OHT that did not convert to glaucoma with 279 age-matched and follow-up time–matched OHT eyes that converted to glaucoma. First, the investigators found that the βPPA area was not significantly different at baseline between OHT eyes that went on to develop glaucoma and those that did not. Then, there was no significant difference in enlargement of βPPA (expressed as ratio of disc area to βPPA area) between case and control eyes. Our findings in the OHT group are therefore consistent with the Ocular Hypertension Treatment Study results with regard to the nonsignificant association between βPPA enlargement and visual field progression.
Differences in the results among the abovementioned studies were likely owing to different designs (prospective vs retrospective, cross-sectional vs longitudinal), sample characteristics (including distribution of ancestry groups and glaucoma status) and end point definitions. The ADAGES database provided a unique opportunity to test how such differences could explain varying results. We found that, in a prospective, longitudinal database including patients followed up over a long period of time with a predefined frequency of optic disc stereophotographs and visual field testing (regardless of clinical status), there is a significant role of ancestry and glaucoma status when interpreting the indicative value of βPPA on glaucoma progression. Moreover, with the advent of SDOCT, one can now assess other features that may act as confounders, such as the microstructural characteristics of PPA, which are concealed during clinical examination, that may also play a role in the indicative value of PPA, as demonstrated in recent studies.20,21 Despite recent advances in understanding the microstructure of PPA using SDOCT, we chose to evaluate the role of this feature as defined clinically (during stereophotograph review), which remains the most commonly used method to describe the presence of PPA in clinical practice.
One potential study limitation is the fact that graders were not masked to other changes in the optic nerve complex, which could have been a confounder in their evaluation of progression. While evaluating the presence of βPPA or its progression, graders could have been influenced by progressive changes seen in the stereoscopic disc stereophotographs, such as alterations in the neuroretinal rim or retinal nerve fiber layer dropout. For example, the Figure depicts an example of a patient who had βPPA progression and visual field deterioration. However, there was significant loss of neuroretinal rim tissue between the 2 disc stereophotographs taken 6 years apart. Notwithstanding this apparent study limitation, the explanation above does not indicate why a significant association would selectively occur among patients of ED but not AD, as we described in our results. Also, despite known differences in disc size between AD and ED patients,22 it has not yet been fully elucidated whether disc size is a significant factor associated with visual field progression. Moreover, complete masking from patients’ race is not possible owing to differences in background retinal pigmentation. As discussed in previous work,9 it is possible that a higher prevalence of βPPA among AD eyes could be an artifact due to differences in background retinal pigmentation (from the retinal pigment epithelium), which increases the contrast between areas with and without retinal pigment epithelium in AD eyes. Misclassification errors, particularly related to potential overlap in the detection of αPPA and βPPA, may have affected the main effects of tested associations. Future studies using objective methods to detect and measure βPPA should test whether this finding holds true. An objective evaluation of the presence of βPPA and its size with automated segmentation (with SDOCT) may not only overcome disagreement between sexes, but could also provide more conclusive results in future studies.
Another limitation is the potential confounding effect of cataract development and progression on the analysis of visual field progression. We tried to minimize this effect by including patients with clear media and visual field results not affected by diffuse loss of sensitivity. Although cataract and its removal could still have influenced our analysis, there does not seem to be any reason to assume that this occurred more selectively in one group than the other, and therefore did not cause systematic bias. Our results may have been influenced by treatment effects, since the differences in mean follow-up IOP between eyes with and without progressing βPPA suggest that eyes with possible structural progression were treated more aggressively. Although unrealistic, studies in which patients were not treated or did not undergo treatment changes during follow-up would provide better estimates of the true association between βPPA progression and visual field changes. Finally, the number of eyes with βPPA progression was small in the OHT group (n = 6), which may help to explain the lack of statistical significance in the analyses of this group.
Whether βPPA is an epiphenomenon or causally connected with GON remains debatable. Nonetheless, our findings suggest that, when pondering the role of βPPA at baseline and its progression, clinicians should take into account where patients stand in the glaucoma continuum as well as their ancestry group. For the moment, clinicians should be aware that the presence of βPPA in individuals of ED increases the risk and rate of glaucomatous visual field progression. More research is needed to clarify the underlying microstructural characteristics of PPA and mechanisms that place some patients at greater risk of progression.
Accepted for Publication: March 20, 2017.
Corresponding Author: C. Gustavo De Moraes, MD, MPH, Bernard and Shirlee Brown Glaucoma Research Laboratory, Edward S. Harkness Eye Institute, Department of Ophthalmology, Columbia University Medical Center, 635 W 165th St, PO Box 69, New York, NY 10022 (firstname.lastname@example.org).
Published Online: May 11, 2017. doi:10.1001/jamaophthalmol.2017.1082
Author Contributions: Dr De Moraes had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: De Moraes, Murphy, Skaat, Cioffi, Medeiros, Weinreb, Liebmann.
Acquisition, analysis, or interpretation of data: De Moraes, Murphy, Kaplan, Reimann, Skaat, Blumberg, Al-Aswad, Cioffi, Girkin, Medeiros, Zangwill, Liebmann.
Drafting of the manuscript: De Moraes, Murphy, Reimann, Al-Aswad, Weinreb.
Critical revision of the manuscript for important intellectual content: De Moraes, Murphy, Kaplan, Skaat, Blumberg, Al-Aswad, Cioffi, Girkin, Medeiros, Weinreb, Zangwill, Liebmann.
Statistical analysis: De Moraes, Murphy, Blumberg.
Obtained funding: Al-Aswad, Cioffi, Girkin, Medeiros, Zangwill, Liebmann.
Administrative, technical, or material support: Kaplan, Skaat, Al-Aswad, Cioffi, Girkin, Medeiros, Weinreb, Zangwill, Liebmann.
Supervision: De Moraes, Al-Aswad, Cioffi, Weinreb, Zangwill, Liebmann.
Other - Presentation at Association for Research in Vision and Ophthalmology: Reimann.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Girkin has received research support from Heidelberg Engineering GmbH, Optove Inc, Merck Inc, Topcon Medical Systems Inc, and Carl Zeiss Meditec Inc. Dr Medeiros has received financial support from Alcon Laboratories Inc, Carl Zeiss Meditec Inc, Pfizer Inc; served as a paid consultant for Alcon Laboratories Inc, Allergan Inc, and Pfizer Inc (not directly relevant to the present study); and received research support from Alcon Laboratories Inc, Allergan Inc, Carl Zeiss Meditec Inc, Pfizer Inc, and Reicherts Inc. Dr Weinreb has received financial support from Carl Zeiss Meditec Inc, Heidelberg Engineering GmbH, Optovue Inc, Topcon Medical Systems Inc, and Nidek Inc; served as a paid consultant for Alcon Laboratories Inc, Allergan Inc, Bausch & Lomb, Aerie, Carl Zeiss Meditec Inc, and Optovue Inc; and received grants from Quark and Genentech. Dr Zangwill has received grants and royalties for intellectual property licensed by the University of California, San Diego; financial support from Carl Zeiss Meditec Inc, Heidelberg Engineering GmbH, Optovue Inc, and Topcon Medical Systems Inc, and research support from Heidelberg Engineering GmbH. Dr Liebmann has served as a paid consultant for Alcon Laboratories Inc, Allergan Inc, and Carl Zeiss Meditec Inc and as a consultant for Dyopsis Inc and Topcon Medical Systems Inc. No other disclosures were reported.
Funding/Support: Funding for the study was provided by National Eye Institute grants U10EY14267, EY08208, EY11008, EY019869, and EY13959; Eyesight Foundation of Alabama; Alcon Laboratories Inc; Allergan Inc; Pfizer Inc; Merck Inc; Santen Inc; and the Edith C. Blum Foundation Fund of the New York Glaucoma Research Institute, (Dr Skaat); and an unrestricted departmental grant from the Research to Prevent Blindness (Department of Ophthalmology, Columbia University Medical Center and University of California, San Diego).
Role of the Funder/Sponsor: The funding organizations and sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Meeting Presentation: Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology; May 1, 2016; Seattle, Washington.