Optic disc photographs of patients with glaucoma. A, A patient with primary open-angle glaucoma exhibiting moderate damage at baseline. B, Follow-up of the same eye shown in A. Optic disc damage does not exhibit prominent progression during a follow-up period of 5 years. C, A patient with normal-pressure glaucoma exhibiting more advanced damage at baseline. D, Follow-up of the same eye shown in C. Neural rim area exhibits a progressive decrease during a follow-up period of 5 years.
Tezel G, Siegmund KD, Trinkaus K, Wax MB, Kass MA, Kolker AE. Clinical Factors Associated With Progression of Glaucomatous Optic Disc Damage in Treated Patients. Arch Ophthalmol. 2001;119(6):813-818. doi:10.1001/archopht.119.6.813
Copyright 2001 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2001
Reducing intraocular pressure (IOP) in glaucomatous eyes does not always prevent disease progression.
To determine the clinical factors associated with progressive optic disc damage in glaucomatous eyes receiving treatment to reduce IOP.
Baseline and follow-up optic disc photographs as well as demographic and clinical data were retrospectively studied in 186 eyes of 93 patients with primary open-angle glaucoma, and in 138 eyes of 69 patients with normal-pressure glaucoma. The patients with primary open-angle glaucoma were included in the study only if their treated IOPs during a follow-up period of 5 years were less than 21 mm Hg. The patients with normal-pressure glaucoma were included only if their IOPs were reduced by at least 20% during the follow-up period. The association of progressive optic disc damage with patient- and eye-specific characteristics was examined using multivariate analysis.
During the 5-year study period, 141 (43.5%) of the 324 eyes exhibited progressive optic disc damage defined by at least a 5% decrease in the neural rim area-to-disc area ratio. Using multivariate analysis, the following were found to be strongly associated with progressive neural rim damage: a baseline smaller neural rim area-disc area ratio (P<.001); a baseline larger zone β area-disc area ratio (P= .04); a baseline larger parapapillary atrophy length-disc circumference ratio (P = .05); a diagnosis of normal-pressure glaucoma(P = .01); and combined medical and surgical treatment prior to the study period (P = .01).
Clinical factors other than IOP may be important indicators of subsequent progression of glaucomatous optic disc damage. Our findings suggest that eyes with advanced glaucomatous optic disc damage and normal-pressure glaucoma are more likely to progress despite receiving treatment to reduce IOP.
CURRENT therapeutic management of glaucoma aims to halt disease progression by reducing intraocular pressure (IOP) to clinically acceptable levels using medical, laser, or surgical treatments. Numerous studies have reported a beneficial effect of pressure reduction in most patients with glaucoma.1,2 However, reduction of IOP to a preselected level does not always prevent progression.3- 5 The proportion of glaucoma patients who exhibited progressive visual field loss despite treatment in various studies has been reported to be between 5% and 80%, with an average of approximately 10% per year.6- 9 This suggests that factors independent of IOP may be responsible, in part, for the sustained progression of glaucoma.10- 13
In this study, we aimed to determine the clinical factors associated with progressive glaucomatous damage and to examine whether there is a relationship between the level of glaucomatous optic disc damage and further progression. We therefore retrospectively analyzed serial optic disc photographs as well as demographic and clinical data of patients with glaucoma who were receiving treatment to reduce IOP.
We studied 186 eyes of 93 patients with primary open-angle glaucoma and 138 eyes of 69 patients with normal-pressure glaucoma. All of the patients studied met the following inclusion criteria: (1) A diagnosis of primary glaucoma, either primary open-angle glaucoma or normal-pressure glaucoma. (2) For patients with primary open-angle glaucoma, treated IOP levels had to be less than 21 mm Hg in both eyes. This was achieved by medical, laser, and/or surgical treatment prior to the study period. (3) For patients with normal-pressure glaucoma, IOP levels had to be reduced by at least 20% in both eyes from the untreated levels measured at the time of initial diagnosis. This was achieved by medical, laser, and/or surgical treatment prior to the study period. (4) The IOPs had to be maintained below these levels during a study period of 5 years. This was assessed by IOP readings at follow-up examinations performed at least yearly during the study period.
The inclusion and exclusion criteria for a diagnosis of primary open-angle glaucoma or normal-pressure glaucoma have been described previously.14 Briefly, normal-pressure glaucoma consisted of the presence of open iridocorneal angles, no evidence of untreated IOP level greater than 23 mm Hg on multiple measurements, glaucomatous changes in visual fields and optic nerve cupping, and the absence of alternative causes of optic neuropathy. The diagnostic criteria for primary open-angle glaucoma were similar to that of normal-pressure glaucoma except that treated or untreated IOP levels had to have been greater than 23 mm Hg on at least 1 occasion. There were no obvious causes for elevated IOP such as pseudoexfoliation, pigment dispersion, or traumatic anterior chamber angle recession. No patient had retinal or optic nerve disease or substantial lens opacities that could interfere with visual field testing and interpretation. None of the patients had a refractive error that exceeded ± 3.0 diopters. All of the patients had complete ophthalmologic examinations at regular follow-up visits, at least yearly, that included IOP readings using applanation tonometry. During the study period, 120 patients(74.1%) were examined at least twice per year. Visual field examinations were performed with the Humphrey Field Analyzer 30-2 program (Allergan-Humphrey Inc, San Leandro, Calif). The criteria for visual field abnormalities on the computerized perimetric tests included a corrected pattern SD with a P value less than .05 or a glaucoma hemifield test outside normal limits obtained with at least 2 sequential and reproducible visual field examinations. A fixation loss of less than 20% and false-positive and false-negative rates less than 33% were the criteria for reliable visual fields.
After we obtained informed consent from the patients, stereoscopic color optic disc photographs were taken at baseline and during the last year of the study period, and were used for morphometric analysis of optic disc and parapapillary atrophy parameters. The morphometric analysis was performed using NIH Image (National Institutes of Health, Bethesda, Md) by one of us(G.T.). The principles of the morphometric analysis were similar to that previously described.14- 16 Briefly, optic disc photographs of all patients were mixed, numbered, and scanned by a person who was not familiar with the optic discs or the diagnosis or follow-up. The borders of the optic disc, optic cup, and parapapillary atrophy were then defined on digitized images with a resolution of 1200 dots per inch per centimeter using a mouse. All of the measurements of the selected areas or lengths were expressed as a ratio to optic disc size to minimize the effects of individual-, age-, and refraction-related errors. The border of the optic disc was defined as the inner edge of the scleral ring and the optic cup was defined on the basis of contour (not color or pallor) while simultaneously evaluating stereoscopic images of the disc with a slide stereoviewer. Parapapillary atrophy was differentiated in 2 zones (α and β) as previously described.17 Zone α was defined as irregular hypopigmentation and hyperpigmentation and zone β was defined by the presence of visible sclera and the large choroidal vessels that were typically located closer to the optic disc border. The scleral ring was not included in the definition of zone β. To describe the extension of parapapillary atrophy, the ratio of the circumferential length of parapapillary atrophy along the optic disc border to the optic disc circumference was used.
Demographic and clinical data collected from patient charts included previous treatment modalities (medical and/or laser and/or surgical). We used mean IOP levels that were calculated using all recorded measurements during the study period. We did not study the effect of specific medical antiglaucoma treatments on progression since they were not constant during the study period. Some of the patients were taking systemic β-adrenergic antagonists and/or calcium channel blockers during the study period. However, we did not examine the effect of these systemic agents on progression since their use was inconstant during the study period. In addition, we did not study the effect of systemic diseases since our collection of this information was incomplete.
Changes in the morphometric optic disc parameters during the study period and associations between progressive optic disc damage and patient- and eye-specific characteristics were statistically analyzed. Optic disc damage was defined as progressive when a 5% decrease occurred in the neural rim area-to-disc area ratio that was beyond the 95% confidence interval of intraexaminer reproducibility.14 We applied a logistic regression model using Proc GENMOD in SAS (SAS Institute Inc, Cary, NC) to obtain odds ratios (ORs) for the associations of the progressive optic disc damage with patient-specific and eye-specific variables (ie; OR >1.0 implies a positive association). Data from all glaucomatous eyes were evaluated together. In addition, we tested for heterogeneity of ORs between primary open-angle glaucoma and normal-pressure glaucoma groups. Data from both eyes of each individual were analyzed to get the most generalizable and precise estimates possible. We used generalized estimating equations and obtained robust variance estimates under an independence working correlation matrix.18,19 These variance estimates accounted for the relationship between eyes within subjects. Multivariable models were developed using a backward stepwise selection procedure. Statistical testing was conducted at the .05 level. Both adjusted and unadjusted (for covariables) regression models are presented.
During the 5-year study period, 141 (43.5%) of 324 eyes exhibited progressive optic disc damage. Forty-six (28.4%) of 162 patients exhibited progressive optic disc damage in both eyes; 49 patients (30.2%) progressed in only 1 eye; and 67 patients (41.4%) did not progress in either eye. The progression rate during the 5-year study period was 37.6% (70/186) in eyes with primary open-angle glaucoma and 51.4% (71/138) in eyes with normal-pressure glaucoma. Baseline patient- and eye-specific characteristics are presented in Table 1 and Table 2.
Representative optic disc photographs taken at baseline and follow-up visits are shown in Figure 1. Table 3 presents the distribution of eyes exhibiting progressive optic disc damage or parapapillary atrophy in 6 categories determined by the baseline level of neural rim damage. While optic disc measures were defined as progressive when a 5% decrease occurred in the neural rim area-disc area ratio, parapapillary atrophy measures were defined as progressive when a 5% increase occurred in the parapapillary atrophy parameters.
In a univariate model using generalized estimating equations, several variables were associated with progressive optic disc damage. These included a diagnosis of normal-pressure glaucoma, a history of surgical treatment, a mean treated level of IOP, and baseline damage in the optic disc and parapapillary tissue. No association was found between progressive optic disc damage and demographic characteristics of patients, including age and sex. In addition, we examined whether progression of optic disc damage differed between black and white patients. However, the confidence interval was wide and inconclusive, largely owing to the small number of black patients in our cohort. An analysis of the relationships between progressive optic disc damage and different variables is presented in Table 4.
Table 5 presents the ORs using a multivariate model. Using the multiple logistic regression model, a smaller baseline neural rim area-disc area ratio, a larger zone β area-disc area ratio, a larger parapapillary atrophy length-disc circumference ratio, a diagnosis of normal-pressure glaucoma, and combined medical and surgical treatment were strongly associated with progression in neural rim damage. After adjusting for other factors using a backward stepwise variable selection procedure, some of the previously detected variables, including mean IOP level, were no longer significant.
When we examined the heterogeneity of ORs between the groups of primary open-angle glaucoma and normal-pressure glaucoma, the effects of risk factors on progressive optic disc damage were the same in all groups of patients.
Despite treatment of IOP, progressive optic disc damage continued in 44% of glaucomatous eyes during a follow-up period of 5 years. We observed that glaucomatous optic disc damage was more likely to progress in eyes with advanced damage to the neural rim and parapapillary tissue. The finding that advanced neural rim loss and advanced parapapillary tissue damage are both associated with progressive optic disc damage is not surprising since these variables are associated with each other.20 The relationship between the baseline level and progression of the optic disc damage was evident in both primary open-angle glaucoma and normal-pressure glaucoma. Combining medical and surgical treatment was found to be another indicator of poor prognosis. This finding may reflect a selection bias since patients thought to be at higher risk for progression or those with higher IOP may have been treated differently by the clinicians from those thought to be at lower risk. For example, filtering surgery is a treatment option that is commonly applied in cases that are unresponsive to other treatment modalities. Moreover, surgical treatment is often used without prior laser treatment in severe cases with advanced damage, rapid progression, or high IOP. The eyes of patients in whom surgery was performed because of clinical judgment of increased risk of progression or greater IOP naturally had more advanced damage at the onset of our study. Therefore, the relationship of the combined treatment modality for progression is a reflection of the relationship we found between advanced status at baseline and progression of glaucomatous damage.
Our finding indicating an association between the level of glaucomatous optic disc damage and its progression is in agreement with previous reports finding that patients with more advanced damage need to be maintained at lower IOP levels to decrease or halt progression. Kronfeld and McGarry21 found that later stages of glaucoma showed a strong tendency toward further progression despite normalization of IOP. Chandler22 and Shaffer23 reported that in patients with advanced glaucomatous changes, IOP should be kept in a normal range to prevent progressive visual field damage. Harbin et al24 asserted that an optic nerve already damaged by glaucoma is more susceptible to further damage. Wilson et al25 noted that apart from initial IOP and a family history of glaucoma, an advanced glaucomatous visual field defect is associated with progression. Other studies have reinforced the importance of IOP reduction in patients with advanced visual field defects and have implied that the disease does not remain stable in patients with advanced visual field loss.26 Chauhan and Drance5 reported that visual fields of patients with established glaucomatous damage can progress, with mean IOP levels of 12 to 16 mm Hg. It has long been presumed that patients with limited glaucomatous damage may tolerate higher IOP levels than patients with advanced disease.4,21 In the current study, the relationship of different factors with progressive optic disc damage and their combined effects on progression were examined using a multivariate analysis. Our findings revealed that glaucomatous damage is more likely to progress in eyes with advanced optic disc damage, independent of IOP.
The IOP in patients with glaucoma may encompass a broad range of values; the significance of pressure-dependent vs pressure-independent pathogenic factors likely varies among patients. Therefore, we studied all patients with glaucoma as a single cohort since both types of glaucoma manifest similar signs of neurodegeneration and are not entirely different diseases.14,27 We found that progressive optic disc damage was closely associated with the baseline level of glaucomatous damage. In addition, we found that a diagnosis of normal-pressure glaucoma was associated with an increased risk of progressive optic disc damage. Our findings demonstrated an increased risk of progressive optic disc damage in normal-pressure glaucoma, despite a 20% reduction in IOP. The disease progression in patients with normal-pressure glaucoma is thought to be markedly reduced by the traditional intervention of IOP reduction.1,2,28,29 Despite measured IOP values within the statistically normal range, many patients with normal-pressure glaucoma show a more marked deterioration of the visual field in the eye with the higher IOP,30- 33 although such a relationship was not confirmed in all studies.34 This finding implies that lowering of IOP might remain a mainstay in the treatment of all patients with glaucoma, regardless of the absolute level of IOP. On the other hand, evaluation of intraocular or systemic factors associated with progressive visual field damage in normal-pressure glaucoma revealed that risk factors unrelated to IOP are associated with progression of visual field loss. While treatment with calcium channel blockers was found to be protective, peripapillary atrophy and disc hemorrhage were found to be significant indicators of progression in Japanese patients with normal-pressure glaucoma.35,36 Our findings also indicate that pressure-independent factors may be more critical in the prognosis of optic nerve damage in normal-pressure glaucoma.
A greater reduction of IOP might have been beneficial in either primary open-angle glaucoma or normal-pressure glaucoma. However, analyses of our data in which only treated levels of IOP were included did not reveal any evidence to support such a conclusion. While no relationship was found between the progression and IOP levels using a multivariable analysis with an univariable model, lower mean levels of treated IOP were found to be associated with progressive optic disc damage. This seemingly paradoxical finding may reflect that the lower treated levels of IOP were found mostly in patients with normal-pressure glaucoma who exhibited the most significant progression in our cohort during the study period. A limitation of our study design might be the potential inability of using mean IOP to correctly identify the stabilization of optic disc damage in a patient with aggressive lowering of IOP during the 5-year study period. Alternatively, variability of the IOP may be an additional factor associated with progressive glaucomatous damage. The evaluation of such a relationship using diurnal IOP readings obtained during a follow-up period might have been informative.
Although elevated IOP has been proposed to be an important risk factor for the development of glaucomatous optic nerve damage,37,38 during advanced stages of the glaucomatous damage, as well as during the course of neural damage seen in normal-pressure glaucoma, additional death pathways may be activated by elevated IOP itself or by concurrent events. Contemporary views of glaucomatous optic neuropathy now recognize that noxious events such as neurotrophin withdrawal, ischemia, excitotoxicity, cytotoxicity by reactive oxygen species, nitric oxide, and tumor necrosis factor-α, glial reactivation, neuronal tissue damage, or autoimmunity, may all potentially contribute to primary and/or secondary degeneration of retinal ganglion cells, which results in disease progression. The sensitivity of already alive but injured neuronal cells to these damaging factors may be increased during the advanced stages of disease. Certain genetic and systemic factors may also contribute to further optic disc damage in the presence of normal IOP. Therefore, although for decades the conventional treatment of glaucoma has been aimed to reduce IOP, it is not always sufficient to halt disease progression. It is for this reason that recent research in glaucoma is focused on alternative treatment strategies for neuroprotection.
Our study was retrospective and therefore has some limitations intrinsic to such a study. Nevertheless, the information obtained from this study provides important clues for the management of patients with glaucoma. First, our findings indicate that continuous and careful examination of the glaucomatous damage is important in all patients with glaucoma regardless of the response of pressure-lowering treatment. Second, our findings suggest that patients with advanced glaucomatous optic nerve damage, as well as patients with normal-pressure glaucoma who already have IOP levels within the statistically normal range, may be good candidates for anticipated neuroprotective interventions.
Accepted for publication December 28, 2000.
This study was supported in part by the Glaucoma Foundation, New York, NY (Dr Tezel), Glaucoma Research Foundation, San Francisco, Calif (Dr Tezel), grant EY012314 from the National Eye Institute, Bethesda, Md (Dr Wax), and an unrestricted grant to Washington University School of Medicine, Department of Ophthalmology and Visual Sciences from Research to Prevent Blindness Inc, New York, NY.
We thank Mae Gordon, PhD, for her helpful advice during the statistical analysis.
Corresponding author and reprints: Gülgün Tezel, MD, Department of Ophthalmology and Visual Sciences, Washington University School of Medicine, Box 8096, 660 S Euclid Ave, St Louis, MO 63110(e-mail: email@example.com).