CLINICAL SCIENCES. Risk Factors for Visual Field Progression in Treated Glaucoma

Transcription

1 CLINICAL SCIENCES Risk Factors for Visual Field Progression in Treated Glaucoma Carlos Gustavo V. De Moraes, MD; Viral J. Juthani, MD; Jeffrey M. Liebmann, MD; Christopher C. Teng, MD; Celso Tello, MD; Remo Susanna Jr, MD; Robert Ritch, MD Objective: To determine intraocular pressure (IOP) dependent and IOP-independent variables associated with visual field (VF) progression in treated glaucoma. Design: Retrospective cohort of the Glaucoma Progression Study. Methods: Consecutive, treated glaucoma patients with repeatable VF loss who had 8 or more VF examinations of either eye, using the Swedish Interactive Threshold Algorithm (24-2 SITA-Standard, Humphrey Field Analyzer II; Carl Zeiss Meditec, Inc, Dublin, California), during the period between January 1999 and September 2009 were included. Visual field progression was evaluated using automated pointwise linear regression. Evaluated data included age, sex, race, central corneal thickness, baseline VF mean deviation, mean follow-up IOP, peak IOP, IOP fluctuation, a detected disc hemorrhage, and presence of beta-zone parapapillary atrophy. Results: We selected 587 eyes of 587 patients (mean [SD] age, 64.9 [13.0] years). The mean (SD) number of VFs was 11.1 (3.0), spanning a mean (SD) of 6.4 (1.7) years. In the univariable model, older age (odds ratio [OR], 1.19 per decade; P=.01), baseline diagnosis of exfoliation syndrome (OR, 1.79; P=.01), decreased central corneal thickness (OR, 1.38 per 40 µm thinner; P.01), a detected disc hemorrhage (OR, 2.31; P.01), presence of betazone parapapillary atrophy (OR, 2.17; P.01), and all IOP parameters (mean follow-up, peak, and fluctuation; P.01) were associated with increased risk of VF progression. In the multivariable model, peak IOP (OR, 1.13; P.01), thinner central corneal thickness (OR, 1.45 per 40 µm thinner; P.01), a detected disc hemorrhage (OR, 2.59; P.01), and presence of beta-zone parapapillary atrophy (OR, 2.38; P.01) were associated with VF progression. Conclusions: IOP-dependent and IOP-independent risk factors affect disease progression in treated glaucoma. Peak IOP is a better predictor of progression than is IOP mean or fluctuation. Arch Ophthalmol. 2011;129(5): Author Affiliations: Einhorn Clinical Research Center, New York Eye and Ear Infirmary (Drs De Moraes, Liebmann, Teng, Tello, and Ritch) and New York University School of Medicine (Drs De Moraes, Juthani, Liebmann, and Teng), New York, and New York Medical College, Valhalla (Drs Teng, Tello, and Ritch); and University of São Paulo School of Medicine, São Paulo, Brazil (Dr Susanna). THE NATIONAL INSTITUTES OF Health randomized clinical trials (RCTs) of glaucoma treatment have helped elucidate the main risk factors for the development and progression of the disease, one of the major causes of irreversible blindness worldwide. 1-7 Yet, the patient populations in these RCTs may not resemble the majority of patients, and their protocol designs may not resemble the clinical conduct typically seen in clinical practice CME available online at and questions on page 543 Retrospective cohort studies have inherent disadvantages compared with RCTs, mostly due to nonstandardized therapy, the presence of various glaucoma phenotypes, lack of stringent reliability criteria for structural or functional testing, and decreased frequency of testing. Nonetheless, large, retrospective cohorts have the potential of providing clinicians with information on the extent to which these prospective, welldesigned trials can be applied to a realword population. The purpose of our study is to verify whether the main risk factors identified in populations enrolled in the major RCTs can also be applied to populations seen in scenarios that more closely resemble a typical clinical practice. METHODS Our study was approved by the New York Eye and Ear Infirmary institutional review board and followed the tenets of the Declaration of Helsinki. We included subjects from the New York Glaucoma Progression Study who were evaluated in the glaucoma referral practice of three of us (J.M.L., R.R., and C.T.) between January 1999 and September After an initial visit consisting of a complete ophthalmologic examination, standard achromatic perimetry (24-2 SITA-Standard, Humphrey Field Analyzer II; Carl Zeiss Meditec, Inc, Dublin, California), and optic disc stereophotographs, patients were reexamined, usually at 3- to 6-month intervals, and the same tests repeated within 6 to 12 months. 562

2 We included patients with glaucomatous optic neuropathy and repeatable visual field (VF) loss who had 8 or more VF examinations of either eye, using the Swedish Interactive Threshold Algorithm (24-2 SITA-Standard, Humphrey Field Analyzer II) because a greater number of VF tests increases the sensitivity and specificity of pointwise linear regression analysis in detecting progression. 13,14 All eligible eyes were required to have bestcorrected visual acuity of 20/40 or better at baseline and a spherical equivalent of less than 6 diopters. If both eyes of the same patient were eligible, the eye with the greater number of VF tests was selected. We excluded subjects with optic disc photographs of poor quality, subjects who were unfamiliar with static perimetry, subjects without reliable baseline results of standard automated perimetry, and subjects whose conditions (other than glaucoma and mild cataract) would likely affect VF testing. DISC PHOTOGRAPH REVIEW Disc photographs of patients in this cohort were reviewed by 2 masked glaucoma specialists searching for disc hemorrhage (DH) and beta-zone parapapillary atrophy ( PPA) using a slide projector. A DH was defined as a splinter-like or flame-shaped hemorrhage on or within the retinal nerve fiber layer or neuroretinal rim. 15,16 Parapapillary atrophy was defined as an inner crescent of chorioretinal atrophy with visible sclera and choroidal vessels ( PPA) and an outer irregular area of hypopigmentation and hyperpigmentation (alpha-zone PPA). 17,18 The presence of DH and PPA at any time in the photographs reviewed was reported. In cases in which the investigators disagreed, a third investigator was used for adjudication. VISUAL FIELD ANALYSIS A glaucomatous VF was defined if a patient had a glaucoma hemifield test result that was outside the normal limits or if the pattern standard deviation was triggered at P.05 on at least 2 consecutive baseline VF tests. The 2 baseline tests required reliability indices better than 25% in order to be included. Automated pointwise linear regression analysis was performed using Progressor version 3.3 software (Medisoft, Ltd, Leeds, England), providing slopes (in units of decibels per year) of progression both globally and locally for each point based on threshold sensitivity maps, as well as its level of significance (P values). Details of the software have been described elsewhere. 19 All patients were familiar with 24-2 achromatic perimetry prior to enrollment. Progression was defined as the presence of a test point with a slope of sensitivity over time of a greater than 1.0-dB loss per year, with P.01. For edge points (nasal-most points of the 24-2), a stricter slope criterion of a greater than 2.0-dB loss per year (also with P.01) was used. 19,20 Because using a single progressing point meeting the aforementioned criteria could result in high falsepositive rates, 20 we increased the specificity of our analysis by requiring that at least 2 progressing points had to be adjacent and within the same hemifield in order to denote the eye as progressing. Global rates of VF change are provided automatically by Progressor and represent the averaged rates for each test point of the sequence of VF tests analyzed. The mean deviation values of the baseline VF were recorded. CLINICAL DATA Peak intraocular pressure (IOP) was the highest measured IOP during the entire follow-up time. The mean IOP during the entire follow-up period was calculated by averaging all pressure measurements obtained after the date of the first VF test entered in the regression. To avoid the undesired effect that numerous sequential IOP measurements during a short period of time would have on the final average, we used the average IOP for each 6-month period following enrollment to calculate the mean follow-up IOP. Intraocular pressure fluctuation was defined as the standard deviation of this value. Intraocular pressure covariance was calculated by dividing the standard deviation IOP by the mean IOP and then multiplying by 100. We excluded all IOP measurements occurring 4 weeks after any type of incisional surgery or laser procedure to avoid the effect of transitory IOP changes that often occur during this period. Central corneal thickness (CCT) was measured using ultrasonic pachymetry (DGH-550; DGH Technology Inc, Exton, Pennsylvania). Exfoliation syndrome was defined as the presence of characteristic exfoliation material on the pupil border or on the lens capsule following pupil dilation. STATISTICAL ANALYSIS Categorical variables were compared using the 2 test. Independent-samples t tests were used for comparisons of continuous variables between groups. Pearson correlation coefficients (r) were calculated to assess the relationship between different IOP parameters. First, a logistic regression was used to evaluate the effect of each IOP parameter on the predefined progression outcome. Because greater follow-up time increases the likelihood of detecting progression, 13,14 all analyses were time adjusted. Also, because IOP-lowering interventions may affect the IOP variability, 21,22 the model was adjusted for the occurrence of any type of glaucoma-filtering procedure during follow-up. Then, the IOP parameter that best correlated with a progression end point was entered in the model that included all evaluated factors. Each variable was first tested in a univariable model. Those with P.25 were then entered in the multivariable analysis. The multivariable model was performed using a stepwise approach; that is, significant variables (P.05) were entered sequentially. Statistical significance was defined at P.05 in the final model. RESULTS We selected 587 eyes of 587 patients. Their mean (SD) age at baseline assessment was 64.9 (13.0) years. The mean (SD) number of evaluated VFs was 11.1 (3.0) (range, 8-24), spanning a mean (SD) of 6.4 (1.7) years (range, years). More often they were women (58%) and of European ancestry (90%). The most common diagnosis was primary open-angle glaucoma (46%) (Table 1). The mean (SD) rate of global VF change for the entire population was 0.45 (0.7) db/y. One hundred seventy eyes (29%) reached a progression end point. Progressing eyes and stable eyes had a mean (SD) global rate of VF change of 1.0 (0.8) db/y and 0.20 (0.4) db/y, respectively (P.01). The Figure shows the distribution of global and pointwise slopes between progressing and stable eyes. Fiftythree eyes (9%) had at least 1 detected DH, and 370 eyes (63%) had PPA. Other clinical data are shown in Table 1. Two hundred six eyes (35%) underwent at least 1 incisional glaucoma surgery during the follow-up period. Peak and mean IOP showed the strongest correlation coefficient (Pearson r=0.74; P.01). The correlation coefficient for peak IOP and IOP standard deviation was also strong and significant (r=0.69; P.01). A significant but still weaker correlation coefficient was observed for mean IOP and standard deviation IOP (r=0.18; P.01). In the time-adjusted logistic regression, all IOP parameters were significantly associated with progression 563

3 Table 1. Demographic Characteristics of the Study Population Patient Population Characteristic (n=587) Age, mean (SD), y 64.9 (13.0) Women, No. (%) 345 (58.7) Ethnicity, No. White 524 African 24 Latino 23 Asian 16 Type of glaucoma, No. POAG 275 NTG 81 XFG 84 JOAG 37 ACG 76 PG 34 CCT, µm (37.3) Follow-up IOP, mean (SD), mm Hg 15.2 (3.1) Peak IOP, mean (SD), mm Hg 19.9 (4.5) Baseline (MD), db 7.1 (5.1) Global rate of VF change, mean (SD), db/y 0.45 (0.7) Eyes reaching progression end point, No. (%) 170 (28.9) Abbreviations: ACG, angle-closure glaucoma; CCT, central corneal thickness; IOP, intraocular pressure; JOAG, juvenile open-angle glaucoma; MD, mean deviation; NTG, normal-tension glaucoma; PG, pigmentary glaucoma; POAG, primary open-angle glaucoma; VF, visual field; XFG, exfoliative glaucoma. A Proportion of Eyes B Proportion of Eyes < 2.0 > 2.0 to < 1.5 < 5.0 > 5.0 to < 4.0 > 1.5 to < 1.0 > 1.0 to > 0.5 to >0.0 to < 0.5 <0.0 <0.5 Rate of Global Sensitivity Change, db/y > 4.0 to < 3.5 > 3.5 to < 3.0 > 3.0 to < 2.5 Rates of Pointwise Sensitivity Change, db/y >0.5 to <1.0 Progressing Stable >1.0 > 2.5 to > 2.0 to > 1.5 to < 2.0 < 1.5 < 1.0 Figure. A, Distribution of global slopes in the entire study population (ie, 587 eyes of 587 patients). B, Distribution of localized slopes (representing progressing eyes only). (Table 2). However, in the multivariable model, only peak IOP remained significantly associated with progression. Eyes that underwent surgery had greater standard deviation IOP than did eyes that did not undergo surgery (3.3 [1.5] mm Hg vs 2.1 [1.0] mm Hg; P.01). Using a predefined cutoff value, eyes with a peak IOP higher than 18 mm Hg had an odds ratio (OR) of 1.81 (95% confidence interval, ) (P.01) associated with a progression outcome. The results of the logistic regression looking at factors associated with VF progression in the entire population are shown in Table 3. In the univariable model, older age, baseline diagnosis of exfoliation syndrome, decreased CCT, detected DH, presence of PPA, and IOP peak were associated with increased risk of progression. In the multivariable model, we found that higher peak IOP (OR, 1.13; P.01), thinner CCT (OR, 1.45 per 40 µm thinner; P.01), DH detection (OR, 2.59; P.01), and presence of PPA (OR, 2.38; P.01) remained significant factors associated with progression. Older age (OR, 1.14; P=.09) and presence of exfoliation syndrome (OR, 1.64; P=.07) reached borderline significance. Eyes with peak IOP measurements of 18 mm Hg or greater progressed at a mean (SD) global rate of 0.52 (0.8) db/y, and eyes with peak IOP measurements of less than 18 mm Hg progressed at a mean (SD) global rate of 0.28 (0.5) db/y (P.01). We then separated the entire population into 2 similarly sized groups based on a median split of the mean follow-up IOP (median, mm Hg). Group A consisted of those with a mean follow-up IOP of mm Hg or greater, and group B consisted of those with a mean follow-up IOP of less than mm Hg. Univariable and multivariable models were also created for these groups. In group A, peak IOP (OR, 1.13; P.01), decreased CCT (OR, 1.41; P=.02), and presence of PPA (OR, 2.45; P.01) were associated with progression in the multivariable model (Table 4). In group B, female sex was protective (OR, 0.54; P=.04), whereas older age (OR, 1.28; P=.04), peak IOP (OR, 1.10; P=.02), decreased CCT (OR, 1.53; P.01), DH detection (OR, 4.53; P.01), and presence of PPA (OR, 2.54; P=.01) were associated with progression. Presence of exfoliation syndrome reached borderline. significance (OR, 2.40; P=.07) (Table 5) COMMENT This retrospective cohort resembles a pragmatic clinical trial that may be useful in ophthalmology. 27 We tested the hypothesis that, in clinical practice, risk factors for VF progression may not always resemble those reported by major glaucoma RCTs owing to stringent inclusion and exclusion criteria and standardized treatment plans and goals. We confirmed the effect of many of these parameters 1-7 and also found significant risk associated with peak IOP and presence of PPA. Our results complement and add to the high level of evidence generated by these RCTs, provide information about the behavior of glaucoma patients receiving therapy, and may help clinicians decide how aggressively to treat specific patients to slow the rate of glaucoma progression. We found a significant effect of IOP in all subgroup analyses, consistent with the majority of clinical trials, 1-6 and we demonstrated that for each increase in millimeters of mercury in IOP, there is a significant increase in the risk of progression for treated glaucoma patients. Even though all IOP parameters were related to progression, peak IOP 564

4 Table 2. Logistic Regression Testing for Association Between Each Intraocular Pressure Parameter and a Progression End Point a Univariable Model Multivariable Model b Parameter OR (95% CI) P Value OR (95% CI) P Value Mean IOP, per mm Hg higher 1.10 ( ) Peak IOP, per mm Hg higher 1.10 ( ) ( ).01 SD IOP, per 0.1 mm Hg higher 1.03 ( ) Covariance IOP, per mm Hg higher 1.02 ( ) Abbreviations: CI, confidence interval; IOP, intraocular pressure; OR, odds ratio; SD, standard deviation. a All patients were included in the analysis. b Adjusted for follow-up time and surgical IOP-lowering interventions only. A stepwise multivariable model was used. Table 3. Logistic Regression Testing the Association Between All Baseline and Intercurrent Factors With a Progression Outcome a Univariable Model Multivariable Model b Variable OR (95% CI) P Value OR (95% CI) P Value Female sex 0.87 ( ).45 African or Latino 0.77 ( ).46 Presence of exfoliation syndrome 1.79 ( ) ( ).07 Age, per decade older 1.19 ( ) ( ).09 CCT, per 40 µm thinner 1.38 ( ) ( ).01 Peak IOP, per mm Hg higher 1.10 ( ) ( ).01 Baseline MD, per db better 1.01 ( ).86 Detection of disc hemorrhage 2.31 ( ) ( ).01 Presence of PPA 2.17 ( ) ( ).01 Abbreviations: PPA, beta-zone parapapillary atrophy; CCT, central corneal thickness; CI, confidence interval; IOP, intraocular pressure; MD, mean deviation; OR, odds ratio. a Only peak IOP was entered in the model because it showed the best association (Table 2). All 587 patients were included in the analysis. b Adjusted for all variables with P.25 in the univariable model. Table 4. Logistic Regression Testing for Association Between All Baseline and Intercurrent Factors With a Progression Outcome for Group A Eyes Only a Univariable Model Multivariable Model b Variable OR (95% CI) P Value OR (95% CI) P Value Female sex 1.02 ( ).92 African or Latino 0.87 ( ).79 Presence of exfoliation syndrome 1.43 ( ) ( ).31 Age, per decade older 1.10 ( ) ( ).47 CCT, per 40 µm thinner 1.48 ( ) ( ).02 Peak IOP, per mm Hg higher 1.12 ( ) ( ).01 Baseline MD, per db better 1.00 ( ).97 Detection of disc hemorrhage 1.41 ( ).42 Presence of PPA 2.23 ( ) ( ).01 Abbreviations: PPA, beta-zone parapapillary atrophy; CCT, central corneal thickness; CI, confidence interval; IOP, intraocular pressure; MD, mean deviation; OR, odds ratio. a Group A consisted of those with mean follow-up IOP of mm Hg or higher. b The only IOP parameter entered in the multivariable model was peak IOP (as described in the fourth paragraph of the Results section). The multivariable model was adjusted for all variables with P.25 in the univariable model. was more important than the mean follow-up IOP. To our knowledge, few trials investigating the effect of IOP parameters (mean, fluctuation, and range) on glaucoma progression 1-7,21,22 have focused on peak IOP. Most RCTs have established target or peak IOP rather than trying to limit means or fluctuation. Zeimer et al 28 suggested the importance of peak IOP on progression in In a post hoc analysis of the Advanced Glaucoma Intervention Study, 3 eyes that consistently had a peak IOP of less than 18 mm Hg had more stable VF throughout follow-up. In our study, 21% of eyes with a peak IOP of 18 mm Hg or greater progressed, with a mean rate of VF change almost twice that of eyes with peak IOP of less than 18 mm Hg. Moreover, a peak IOP of greater than 18 mm Hg increased the risk of progression by 81%. During water-drinking IOP stress tests, eyes with higher IOP peaks have been reported to be more likely to have worse VF damage and an increased likelihood of progression using event analysis. 29,30 565

5 Table 5. Logistic Regression Testing for Association Between All Baseline and Intercurrent Factors With a Progression Outcome for Group B Eyes Only a Univariable Model Multivariable Model b Variable OR (95% CI) P Value OR (95% CI) P Value Female sex 0.65 ( ) ( ).04 Nonwhite 0.71 ( ).52 Presence of exfoliation syndrome 2.18 ( ) ( ).07 Age, per decade older 1.31 ( ) ( ).04 CCT, per 40 µm thinner 1.44 ( ) ( ).01 Peak IOP, per mm Hg higher 1.09 ( ) ( ).02 Baseline MD, per db better 0.98 ( ).53 Detection of disc hemorrhage 3.80 ( ) ( ).01 Presence of PPA 2.59 ( ) ( ).01 Abbreviations: PPA, beta-zone parapapillary atrophy; CCT, central corneal thickness; CI, confidence interval; IOP, intraocular pressure; MD, mean deviation; OR, odds ratio. a Group B consisted of those with mean follow-up IOP of less than mm Hg. b The only IOP parameter entered in the multivariable model was peak IOP (as described in the fourth paragraph of the Results section). The multivariable model was adjusted for all variables with P.25 in the univariable model. There is disagreement on the effects of IOP variation and fluctuation and whether IOP variation and fluctuation are risk factors for progression 21,22,31,32 or due largely to the intrinsic relationship between IOP variation and the effect of intervention. For instance, if progression is suspected over time, additional procedures (incisional surgery, laser surgery, or medication) will result in lowering of the IOP with a consequent treatment-induced increase in the IOP fluctuation. Some RCTs established target peak IOP, even though this parameter was not incorporated into their risk models. For instance, in the Ocular Hypertension Treatment Study, 1 the treatment goals for subjects randomly assigned to medication were IOPs of 24 mm Hg or less and a 20% reduction in IOP from the average of the qualifying and baseline IOP. Medical therapy was advanced to achieve the targeted IOP reduction. For the Early Manifest Glaucoma Trial, 2 a standard IOP-lowering protocol was initially applied to all patients in the treatment arm. Our data demonstrated a significant correlation among all IOP parameters. Peak and mean IOP showed the strongest association, suggesting that patients with a higher mean IOP during follow-up also tend to be those having the highest peaks. One advantage of using peak IOP in clinical practice is its ease of assessment because of its intrinsic relationship to target IOP. Peak IOP can be reassessed at each visit, whereas mean IOP determination requires longitudinal data collection and may be affected by the interval between visits. For instance, if a patient presents with a high IOP at one visit, IOP checks may occur more frequently, falsely elevating the mean. Also, establishing a target peak IOP is clinically easier than seeking to establish a target mean IOP or target IOP fluctuation. Of note, peak IOP remained a significant factor for progression even among eyes with lower mean follow-up IOP in our study. Three major RCTs confirmed a strong association between DH and future VF development and progression. 7,33,34 However, the etiology of DH remains unclear. Disc hemorrhage may be a primary vascular event, 35,36 or it may result from an ongoing degenerative process at the optic nerve head Supporting the latter hypothesis, we have reported that sustained, localized, spatially consistent VF loss precedes the onset of DH and continues in the same region. 39 The development of DH could therefore be an indirect sign that the optic nerve complex (optic nerve head and parapapillary region) is particularly susceptible to damage. Disc hemorrhage more often occurs adjacent to areas of neuroretinal rim thinning and retinal nerve fiber layer loss 37,38 and within areas of greater PPA width. 40 This also suggests that DH occurs in areas of existing structural damage and increases the likelihood of subsequent VF progression. 41,42 The PPA has been associated with subsequent VF loss in patients with ocular hypertension 43 and in patients with established glaucoma. 44 Using a different cohort, we have also shown that both the presence and the size of PPA affect the risk for progression. 45 Park et al 46 also reported that PPA is a significant risk factor in young patients having normal-tension glaucoma with moderate to severe VF loss. Nevertheless, to our knowledge, no RCT has examined the effect of PPA on glaucoma development or progression. One disadvantage of PPA assessment is the difficulty in differentiating glaucomatous PPA from myopic parapapillary degeneration. We minimized this by excluding eyes with high myopia and including only eyes with confirmed glaucomatous optic neuropathy. One reason PPA may have had a strong association with progression in our sample is that we defined its presence by looking at the entire set of disc photographs. Previous studies have suggested that PPA enlarges as glaucoma progresses. 47,48 Structural progression increases the likelihood of future VF progression. 41,42 Therefore, PPA could be a surrogate for previous structural progression, which dramatically increases its association with subsequent VF progression. Also, the microstructural changes within the optic nerve complex 49 may predispose to a faster rate of future structural loss. This is partly corroborated by the fact that DHs are more commonly seen in eyes with PPA and within areas where PPA is the widest. 40 Although a thinner central cornea increases the risk of VF progression, 1,2,5,6 the precise pathophysiology of this effect remains unclear. Proposed mechanisms range from 566

6 tonometric artifact to altered ocular biomechanics. To support the latter hypothesis, some reports have suggested an association between corneal biomechanical markers and optic nerve compliance and susceptibility. 50,51 We confirmed that a thinner cornea is a risk factor for progression of glaucoma in treated patients. In the Early Manifest Glaucoma Trial, 2 exfoliation syndrome increased the risk of progression by more than 200%, and other studies have reported similar findings We have previously demonstrated that eyes with exfoliative glaucoma progress almost twice as fast as eyes with normaltension glaucoma. 55 Even though exfoliation syndrome reached only borderline significance in most of our multivariable analyses, there was a trend for these eyes to progress faster than eyes without exfoliation syndrome (mean [SD] global rate of VF change, 0.65 [0.7] db/yr vs 0.41 [0.7] db/y; P=.008). However, exfoliation syndrome lost its significance once IOP and other parameters were added to the model (Tables 3, 4, and 5). TheOcularHypertensionTreatmentStudy 1 andtheearly Manifest Glaucoma Trial 2 showed a significant association between baseline mean deviation and an increased risk of progression. The Advanced Glaucoma Intervention Study found that nonprogressing eyes had better VF scores at 4 years than did progressing eyes. 21 Our study found no association between baseline mean deviation and subsequent progression. This could also be due, at least in part, to the intent-to-treat approach used in RCTs It is also possible that the baseline mean deviation was not significantly associated with VF progression in our study because we includedapopulationthatrangedfrommildtoseverevfloss. For those with more negative mean deviation values, a floor effect may have affected pointwise linear regression. Once the VF sensitivities reach very low values, the slopes tend to be flat even though progression continues to occur. Age as a risk factor for glaucoma progression has been confirmed by many clinical trials 1,2,4,5,21 but refuted by others. 6,7 In our study, univariable analysis revealed the significance of older age when the entire population was analyzed (19% increased risk per decade older), but older age lost significance after adjusting for other factors, such as IOP and CCT. However, in the group with lower mean follow-up IOP (group B), older age became significant even in the multivariable model, confirming results of previous studies. 1,2,4,5,21 The Early Manifest Glaucoma Trial also found that older age played a more significant role in the group of patients with lower baseline IOP. 2 Once IOP is substantially lowered to a certain level, 56,57 the effect of other IOP-independent factors (such as age) may become more apparent. Aging may increase the susceptibility of the optic nerve complex to IOP stress. 58,59 Despite population-based studies stating that African ancestry confers susceptibility to glaucoma development 60,61 and progression, 62 none of the RCTs that investigated the effect of race on progression has found any significant association in multivariable analyses. In our study, nonwhites were underrepresented, preventing us from any interpretation of these results. Most RCTs have failed to show any association between sex and glaucoma development or progression. 1-6 In the Ocular Hypertension Treatment Study, 1 men were more likely to convert to glaucoma than were women in the univariable model. In our study, male sex was a risk factor among patients with lower mean follow-up IOP. Our study is limited by its retrospective nature and by the biases inherent in a tertiary-care referral center; therefore, we suggest that a pragmatic clinical trial 25,26 be evaluated with care. These limitations, however, are mitigated by the large sample size, the long follow-up period, the large number of VF tests, and a cohort that closely resembles those seen in clinical practice. Our findings may not be applicable to other clinical settings because they reflect a specific population for which certain risk factors may play a more important role than others. The significant biases and limitations of nonstandardized treatment and follow-up protocols may slant the assessment of risk. For example, patients may be treated more or less aggressively depending on the ophthalmologist s perceived risk for further deterioration. This may not be the same for all ophthalmologists or even the same ophthalmologist at another time. Our main purpose, however, was to confirm whether risk factors identified in RCTs are the same as those in clinical practice. Confirmation from other clinical databases could provide even greater clinical relevance for RCTs and enhance their importance. 27 One advantage of using trend analysis over event analysis was our ability to provide rates of VF change in this population. Depending on the number and frequency of VF tests, trend analysis may provide better sensitivity and specificity to detect progression compared with event analysis. 13,14 Our population underwent a large number of VF tests, increasing the power of our analyses. We chose a more specific definition of progression using pointwise linear regression, which could have been at the expense of lower sensitivity. Yet our definition of progression, requiring 2 adjacent points in the same hemifield, increased our likelihood of detecting true VF progression. In conclusion, we confirmed that, in a population of patients treated for glaucoma, well-known and reported risk factors are indeed relevant for risk assessment. Moreover, additional information provided by simple diagnostic tools (tonometry and ophthalmoscopy) suggest the importance of peak IOP and the presence of PPA to assess the future risk of glaucomatous vision loss. Clinicians should be aware of the continued roles played by both IOP-dependent and IOP-independent factors in treated patients. Submitted for Publication: February 22, 2010; final revision received July 13, 2010; accepted August 1, Correspondence: Jeffrey M. Liebmann, MD, Einhorn Clinical Research Center, New York Eye and Ear Infirmary, 310 E 14th St, New York, NY Financial Disclosure: None reported. Funding/Support: This study was supported by the Joseph and Geraldine LaMotta Research Fund of the New York Glaucoma Research Institute and by the Glaucoma Research and Education Fund of Lenox Hill Hospital (Dr De Moraes), both of which are located in New York. REFERENCES 1. Gordon MO, Beiser JA, Brandt JD, et al. The Ocular Hypertension Treatment Study: baseline factors that predict the onset of primary open-angle glaucoma. Arch Ophthalmol. 2002;120(6): ,

7 2. Leske MC, Heijl A, Hyman L, Bengtsson B, Dong L, Yang Z; EMGT Group. Predictors of long-term progression in the Early Manifest Glaucoma Trial. Ophthalmology. 2007;114(11): AGIS Investigators. The Advanced Glaucoma Intervention Study (AGIS): 7, the relationship between control of intraocular pressure and visual field deterioration. Am J Ophthalmol. 2000;130: Musch DC, Gillespie BW, Lichter PR, Niziol LM, Janz NK; CIGTS Study Investigators. Visual field progression in the Collaborative Initial Glaucoma Treatment Study: the impact of treatment and other baseline factors. Ophthalmology. 2009; 116(2): Miglior S, Torri V, Zeyen T, Pfeiffer N, Vaz JC, Adamsons I; EGPS Group. Intercurrent factors associated with the development of open-angle glaucoma in the European glaucoma prevention study. Am J Ophthalmol. 2007;144(2): Medeiros FA, Sample PA, Zangwill LM, Bowd C, Aihara M, Weinreb RN. Corneal thickness as a risk factor for visual field loss in patients with preperimetric glaucomatous optic neuropathy. Am J Ophthalmol. 2003;136(5): Drance S, Anderson DR, Schulzer M; Collaborative Normal-Tension Glaucoma Study Group. Risk factors for progression of visual field abnormalities in normaltension glaucoma. Am J Ophthalmol. 2001;131(6): Cioffi GA, Liebmann JM. Translating the OHTS results into clinical practice. J Glaucoma. 2002;11(5): Singh K. The randomized clinical trial: beware of limitations. J Glaucoma. 2004; 13(2): Jampel HD. We should treat fewer patients with elevated intraocular pressure now that we know the results of the Ocular Hypertension Treatment Study. Arch Ophthalmol. 2004;122(3): Wishart PK. Interpretation of the glaucoma landmark studies. Br J Ophthalmol. 2009;93(5): Kass MA, Gordon MO, Kymes SM. Incorporating the results of the Ocular Hypertension Treatment Study into clinical practice. Arch Ophthalmol. 2005;123 (7): Chauhan BC, Garway-Heath DF, Goñi FJ, et al. Practical recommendations for measuring rates of visual field change in glaucoma. Br J Ophthalmol. 2008; 92(4): Spry PG, Bates AB, Johnson CA, Chauhan BC. Simulation of longitudinal threshold visual field data. Invest Ophthalmol Vis Sci. 2000;41(8): Airaksinen PJ, Mustonen E, Alanko HI. Optic disc hemorrhages: analysis of stereophotographs and clinical data of 112 patients. Arch Ophthalmol. 1981;99 (10): Jonas JB, Iester M. Disc hemorrhage and glaucoma. Ophthalmology. 1995;102 (3): Jonas JB, Nguyen XN, Gusek GC, Naumann GO. Parapapillary chorioretinal atrophy in normal and glaucoma eyes: I, morphometric data. Invest Ophthalmol Vis Sci. 1989;30(5): Jonas JB, Naumann GO. Parapapillary chorioretinal atrophy in normal and glaucoma eyes: II, correlations. Invest Ophthalmol Vis Sci. 1989;30(5): Fitzke FW, Hitchings RA, Poinoosawmy D, McNaught AI, Crabb DP. Analysis of visual field progression in glaucoma. Br J Ophthalmol. 1996;80(1): Gardiner SK, Crabb DP. Examination of different pointwise linear regression methods for determining visual field progression. Invest Ophthalmol Vis Sci. 2002; 43(5): Nouri-Mahdavi K, Hoffman D, Coleman AL, et al; Advanced Glaucoma Intervention Study. Predictive factors for glaucomatous visual field progression in the Advanced Glaucoma Intervention Study. Ophthalmology. 2004;111(9): Bengtsson B, Leske MC, Hyman L, Heijl A; Early Manifest Glaucoma Trial Group. Fluctuation of intraocular pressure and glaucoma progression in the Early Manifest Glaucoma Trial. Ophthalmology. 2007;114(2): Möller HJ. Do effectiveness ( real world ) studies on antipsychotics tell us the real truth? Eur Arch Psychiatry Clin Neurosci. 2008;258(5): Kent DM, Kitsios G. Against pragmatism: on efficacy, effectiveness and the real world. Trials. 2009;10: Tunis SR, Stryer DB, Clancy CM. Practical clinical trials: increasing the value of clinical research for decision making in clinical and health policy. JAMA. 2003; 290(12): Glasgow RE, Magid DJ, Beck A, Ritzwoller D, Estabrooks PA. Practical clinical trials for translating research to practice: design and measurement recommendations. Med Care. 2005;43(6): De Moraes CG, Ritch R, Liebmann JM. Bridging the major prospective National Eye Institute-sponsored glaucoma clinical trials and clinical practice. J Glaucoma. 2011;20(1): Zeimer RC, Wilensky JT, Gieser DK, Viana MA. Association between intraocular pressure peaks and progression of visual field loss. Ophthalmology. 1991; 98(1): Susanna R Jr, Hatanaka M, Vessani RM, Pinheiro A, Morita C. Correlation of asymmetric glaucomatous visual field damage and water-drinking test response. Invest Ophthalmol Vis Sci. 2006;47(2): Susanna R Jr, Vessani RM, Sakata L, Zacarias LC, Hatanaka M. The relation between intraocular pressure peak in the water drinking test and visual field progression in glaucoma. Br J Ophthalmol. 2005;89(10): Medeiros FA, Weinreb RN, Zangwill LM, et al. Long-term intraocular pressure fluctuations and risk of conversion from ocular hypertension to glaucoma. Ophthalmology. 2008;115(6): Asrani S, Zeimer R, Wilensky J, Gieser D, Vitale S, Lindenmuth K. Large diurnal fluctuations in intraocular pressure are an independent risk factor in patients with glaucoma. J Glaucoma. 2000;9(2): Budenz DL, Anderson DR, Feuer WJ, et al; Ocular Hypertension Treatment Study Group. Detection and prognostic significance of optic disc hemorrhages during the Ocular Hypertension Treatment Study. Ophthalmology. 2006;113(12): Bengtsson B, Leske MC, Yang Z, Heijl A; EMGT Group. Disc hemorrhages and treatment in the Early Manifest Glaucoma Trial. Ophthalmology. 2008;115(11): Begg IS, Drance SM, Sweeney VP. Ischaemic optic neuropathy in chronic simple glaucoma. Br J Ophthalmol. 1971;55(2): Sonnsjö B, Krakau CE. Arguments for a vascular glaucoma etiology. Acta Ophthalmol (Copenh). 1993;71(4): Law SK, Choe R, Caprioli J. Optic disk characteristics before the occurrence of disk hemorrhage in glaucoma patients. Am J Ophthalmol. 2001;132(3): Jonas JB, Martus P, Budde WM, Hayler J. Morphologic predictive factors for development of optic disc hemorrhages in glaucoma. Invest Ophthalmol Vis Sci. 2002;43(9): De Moraes CG, Prata TS, Liebmann CA, Tello C, Ritch R, Liebmann JM. Spatially consistent, localized visual field loss before and after disc hemorrhage. Invest Ophthalmol Vis Sci. 2009;50(10): Radcliffe NM, Liebmann JM, Rozenbaum I, et al. Anatomic relationships between disc hemorrhage and parapapillary atrophy. Am J Ophthalmol. 2008; 146(5): Chauhan BC, Nicolela MT, Artes PH. Incidence and rates of visual field progression after longitudinally measured optic disc change in glaucoma. Ophthalmology. 2009;116(11): Medeiros FA, Alencar LM, Zangwill LM, Bowd C, Sample PA, Weinreb RN. Prediction of functional loss in glaucoma from progressive optic disc damage. Arch Ophthalmol. 2009;127(10): Quigley HA, Enger C, Katz J, Sommer A, Scott R, Gilbert D. Risk factors for the development of glaucomatous visual field loss in ocular hypertension. Arch Ophthalmol. 1994;112(5): Jonas JB, Martus P, Horn FK, Jünemann A, Korth M, Budde WM. Predictive factors of the optic nerve head for development or progression of glaucomatous visual field loss. Invest Ophthalmol Vis Sci. 2004;45(8): Teng CC, De Moraes CG, Prata TS, Tello C, Ritch R, Liebmann JM. Beta-Zone parapapillary atrophy and the velocity of glaucoma progression. Ophthalmology. 2010;117(5): Park SC, Lee DH, Lee HJ, Kee C. Risk factors for normal-tension glaucoma among subgroups of patients. Arch Ophthalmol. 2009;127(10): Budde WM, Jonas JB. Enlargement of parapapillary atrophy in follow-up of chronic open-angle glaucoma. Am J Ophthalmol. 2004;137(4): Uchida H, Ugurlu S, Caprioli J. Increasing peripapillary atrophy is associated with progressive glaucoma. Ophthalmology. 1998;105(8): Curcio CA, Saunders PL, Younger PW, Malek G. Peripapillary chorioretinal atrophy: Bruch s membrane changes and photoreceptor loss. Ophthalmology. 2000; 107(2): Wells AP, Garway-Heath DF, Poostchi A, Wong T, Chan KC, Sachdev N. Corneal hysteresis but not corneal thickness correlates with optic nerve surface compliance in glaucoma patients. Invest Ophthalmol Vis Sci. 2008;49(8): Bochmann F, Ang GS, Azuara-Blanco A. Lower corneal hysteresis in glaucoma patients with acquired pit of the optic nerve (APON). Graefes Arch Clin Exp Ophthalmol. 2008;246(5): Konstas AG, Stewart WC, Stroman GA, Sine CS. Clinical presentation and initial treatment patterns in patients with exfoliation glaucoma versus primary openangle glaucoma. Ophthalmic Surg Lasers. 1997;28(2): Puska PM. Unilateral exfoliation syndrome: conversion to bilateral exfoliation and to glaucoma: a prospective 10-year follow-up study. J Glaucoma. 2002;11 (6): Ritch R. The management of exfoliative glaucoma. Prog Brain Res. 2008;173: Ahrlich KG, De Moraes CG, Teng CC, et al. Visual field progression differences between normal-tension and exfoliative high-tension glaucoma. Invest Ophthalmol Vis Sci. 2010;51(3): Folgar FA, de Moraes CG, Prata TS, et al. Glaucoma surgery decreases the rates of localized and global visual field progression. Am J Ophthalmol. 2010;149 (2): Medeiros FA, Pinheiro A, Moura FC, Leal BC, Susanna R Jr. Intraocular pressure fluctuations in medical versus surgically treated glaucomatous patients. J Ocul Pharmacol Ther. 2002;18(6): Burgoyne CF, Downs JC. Premise and prediction-how optic nerve head biomechanics underlies the susceptibility and clinical behavior of the aged optic nerve head. J Glaucoma. 2008;17(4): Girard MJ, Suh JK, Bottlang M, Burgoyne CF, Downs JC. Scleral biomechanics in the aging monkey eye. Invest Ophthalmol Vis Sci. 2009;50(11): Sommer A, Tielsch JM, Katz J, et al. Racial differences in the cause-specific prevalence of blindness in east Baltimore. N Engl J Med. 1991;325(20): Congdon N, O Colmain B, Klaver CC, et al; Eye Diseases Prevalence Research Group. Causes and prevalence of visual impairment among adults in the United States. Arch Ophthalmol. 2004;122(4): Wilson R, Richardson TM, Hertzmark E, Grant WM. Race as a risk factor for progressive glaucomatous damage. Ann Ophthalmol. 1985;17(10):

8 Additional Contributions: We thank Rita Popat, PhD, at Stanford University; Bennie Jeng, MD, and Dandan Wang, MD, at the University of California, San Francisco; and the San Francisco General Hospital staff for their invaluable help with this project. REFERENCES 1. Giangiacomo AC. The epidemiology of glaucoma. In: Grehn F, Stamper R, eds. Glaucoma. Berlin, Germany: Springer; 2009: Kass MA, Heuer DK, Higginbotham EJ, et al. The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. Arch Opthalmol. 2002;120(6): Leske MC, Heijl A, Hussein M, Bengtsson B, Hyman L, Komaroff E; Early Manifest Glaucoma Trial Group. Factors for glaucoma progression and the effect of treatment: the Early Manifest Glaucoma Trial. Arch Ophthalmol. 2003;121(1): Leske MC, Heijl A, Hyman L, Bengtsson B, Dong L, Yang Z; EMGT Group. Predictors of long-term progression in the Early Manifest Glaucoma Trial. Ophthalmology. 2007;114(11): Epstein DL, Krug JH Jr, Hertzmark E, Remis LL, Edelstein DJ. A long-term clinical trial of timolol therapy versus no treatment in the management of glaucoma suspects. Ophthalmology. 1989;96(10): Kass MA. Timolol treatment prevents or delays glaucomatous visual field loss in individuals with ocular hypertension: a five-year, randomized, double-masked, clinical trial. Trans Am Ophthalmol Soc. 1989;87: Schulzer M, Drance SM, Douglas GR. A comparison of treated and untreated glaucoma suspects. Ophthalmology. 1991;98(3): DiMatteo MR. Variations in patients adherence to medical recommendations: a quantitative review of 50 years of research. Med Care. 2004;42(3): Friedman DS, Hahn SR, Quigley HA, et al. Doctor-patient communication in glaucoma care: analysis of videotaped encounters in community-based office practice. Ophthalmology. 2009;116(12): American Academy of Ophthalmology Glaucoma Panel. Preferred Practice Patterns: Primary Open-Angle Glaucoma, Primary Open-Angle Glaucoma Suspect, and Primary Angel Closure. San Francisco, CA: American Academy of Ophthalmology; Kosoko O, Quigley HA, Vitale S, Enger C, Kerrigan L, Tielsch JM. Risk factors for noncompliance with glaucoma follow-up visits in a residents eye clinic. Ophthalmology. 1998;105(11): Lee BW, Sathyan P, John RK, Singh K, Robin AL. Predictors of and barriers associated with poor follow-up in patients with glaucoma in South India. Arch Ophthalmol. 2008;126(10): Quigley HA, West SK, Rodriguez J, Munoz B, Klein R, Snyder R. The prevalence of glaucoma in a population-based study of Hispanic subjects: Proyecto VER. Arch Ophthalmol. 2001;119(12): Shimmyo M, Ross AJ, Moy A, Mostafavi R. Intraocular pressure, Goldmann applanation tension, corneal thickness, and corneal curvature in Caucasians, Asians, Hispanics, and African Americans. Am J Ophthalmol. 2003;136(4): Tielsch JM, Sommer A, Katz J, Royall RM, Quigley HA, Javitt J. Racial variations in the prevalence of primary open-angle glaucoma: the Baltimore Eye Survey. JAMA. 1991;266(3): Kim E, Varma R. Glaucoma in Latinos/Hispanics. Curr Opin Ophthalmol. 2009;21 (2): Varma R, Ying-Lai M, Francis BA, et al; Los Angeles Latino Eye Study Group. Prevalence of open-angle glaucoma and ocular hypertension in Latinos: the Los Angeles Latino Eye Study. Ophthalmology. 2004;111(8): Martin MJ, Sommer A, Gold EB, Diamond EL. Race and primary open-angle glaucoma. Am J Ophthalmol. 1985;99(4): Sommer A, Tielsch JM, Katz J, et al. Racial differences in the cause-specific prevalence of blindness in east Baltimore. N Engl J Med. 1991;325(20): Wormald RP, Basauri E, Wright LA, Evans JR. The African Caribbean Eye Survey: risk factors for glaucoma in a sample of African Caribbean people living in London. Eye (Lond). 1994;8(pt 3): Gwira JA, Vistamehr S, Shelsta H, et al. Factors associated with failure to follow up after glaucoma screening: a study in an African American population. Ophthalmology. 2006;113(8): Richardson JL, Langholz B, Bernstein L, Burciaga C, Danley K, Ross RK. Stage and delay in breast cancer diagnosis by race, socioeconomic status, age and year. Br J Cancer. 1992;65(6): Madison T, Schottenfeld D, James SA, Schwartz AG, Gruber SB. Endometrial cancer: socioeconomic status and racial/ethnic differences in stage at diagnosis, treatment, and survival. Am J Public Health. 2004;94(12): Rauscher GH, Ferrans CE, Kaiser K, Campbell RT, Calhoun EE, Warnecke RB. Misconceptions about breast lumps and delayed medical presentation in urban breast cancer patients. Cancer Epidemiol Biomarkers Prev. 2010;19(3): Institute of Medicine. The healthcare environment and its relation to disparities. In: Smedley BD, Stith AY, Nelson AR, eds. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: Institute of Medicine; 2003: Institute of Medicine. Assessing potential sources of racial and ethnic disparities in care: patient- and system-level factors. In: Smedley BD, Stith AY, Nelson AR, eds. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: Institute of Medicine; 2003: Institute of Medicine. Assessing potential sources of racial and ethnic disparities in care in the clinical encounter. In: Smedley BD, Stith AY, Nelson AR, eds. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: Institute of Medicine; 2003: US Department of Health & Human Services. National Healthcare Disparities Report. Washington, DC: US Dept of Health and Human Services; Correction Error in Text. In the Clinical Sciences article titled Risk Factors for Visual Field Progression in Treated Glaucoma, by De Moraes et al, published in the May issue of the Archives (2011;129[5]: ), the last sentence of the last paragraph in the right column of page 566 should have appeared as follows: This is partly corroborated by the fact that DHs are more commonly seen in eyes with PPA and within areas where PPA is the widest. 40 This article was corrected online. ARCH OPHTHALMOL / VOL 129 (NO. 7), JULY