Assessment of Cumulative Incidence and Severity of Primary Open-Angle Glaucoma Among Participants in the Ocular Hypertension Treatment Study After 20 Years of Follow-up

Key Points

Question

Do 20-year follow-up data from the Ocular Hypertension Treatment Study inform the management of patients with ocular hypertension?

Findings

In this cohort study of 1636 participants with ocular hypertension who participated in the Ocular Hypertension Treatment Study, the 20-year cumulative incidence of primary open-angle glaucoma was 46% in 1 or both eyes, and the cumulative incidence of visual field loss was 25% after adjusting for exposure time.

Meaning

This study’s findings, together with a predictive model, may help clinicians and patients make informed personalized decisions about the management of ocular hypertension.

Abstract

Importance

Ocular hypertension is an important risk factor for the development of primary open-angle glaucoma (POAG). Data from long-term follow-up can be used to inform the management of patients with ocular hypertension.

Objective

To determine the cumulative incidence and severity of POAG after 20 years of follow-up among participants in the Ocular Hypertension Treatment Study.

Design, Setting, and Participants

Participants in the Ocular Hypertension Treatment Study were followed up from February 1994 to December 2008 in 22 clinics. Data were collected after 20 years of follow-up (from January 2016 to April 2019) or within 2 years of death. Analyses were performed from July 2019 to December 2020.

Interventions

From February 28, 1994, to June 2, 2002 (phase 1), participants were randomized to receive either topical ocular hypotensive medication (medication group) or close observation (observation group). From June 3, 2002, to December 30, 2008 (phase 2), both randomization groups received medication. Beginning in 2009, treatment was no longer determined by study protocol. From January 7, 2016, to April 15, 2019 (phase 3), participants received ophthalmic examinations and visual function assessments.

Main Outcomes and Measures

Twenty-year cumulative incidence and severity of POAG in 1 or both eyes after adjustment for exposure time.

Results

A total of 1636 individuals (mean [SD] age, 55.4 [9.6] years; 931 women [56.9%]; 1138 White participants [69.6%]; 407 Black/African American participants [24.9%]) were randomized in phase 1 of the clinical trial. Of those, 483 participants (29.5%) developed POAG in 1 or both eyes (unadjusted incidence). After adjusting for exposure time, the 20-year cumulative incidence of POAG in 1 or both eyes was 45.6% (95% CI, 42.3%-48.8%) among all participants, 49.3% (95% CI, 44.5%-53.8%) among participants in the observation group, and 41.9% (95% CI, 37.2%-46.3%) among participants in the medication group. The 20-year cumulative incidence of POAG was 55.2% (95% CI, 47.9%-61.5%) among Black/African American participants and 42.7% (95% CI, 38.9%-46.3%) among participants of other races. The 20-year cumulative incidence for visual field loss was 25.2% (95% CI, 22.5%-27.8%). Using a 5-factor baseline model, the cumulative incidence of POAG among participants in the low-, medium-, and high-risk tertiles was 31.7% (95% CI, 26.4%-36.6%), 47.6% (95% CI, 41.6%-53.0%), and 59.8% (95% CI, 53.1%-65.5%), respectively.

Conclusions and Relevance

In this study, only one-fourth of participants in the Ocular Hypertension Treatment Study developed visual field loss in either eye over long-term follow-up. This information, together with a prediction model, may help clinicians and patients make informed personalized decisions about the management of ocular hypertension.

Trial Registration

ClinicalTrials.gov Identifier: NCT00000125.

This cohort study comprises phase 3 of the Ocular Hypertension Treatment Study and assesses the 20-year cumulative incidence and severity of primary open-angle glaucoma among the original cohort of participants with ocular hypertension.

Introduction

Ocular hypertension (OHT), a common condition, is a leading risk factor for the development of primary open-angle glaucoma (POAG). The Ocular Hypertension Treatment Study (OHTS) was designed to address the debate regarding the management of patients with OHT. The OHTS was completed in 3 phases (OHTS 1, OHTS 2, and OHTS 3). The OHTS 1 was a randomized clinical trial conducted from February 28, 1994, to June 2, 2002, with the aim of answering 2 questions: (1) Does early treatment reduce the incidence of POAG in patients with OHT? (2) Do baseline demographic and clinical factors predict which patients with OHT are at low, medium, or high risk of developing POAG? Between February 1994 and October 1996, 1636 participants with OHT were randomized to receive either topical ocular hypotensive medication (medication group) or close observation (observation group). Treatment with medication was associated with a reduction in the 5-year cumulative incidence of POAG from 9.5% in the observation group to 4.4% in the medication group (hazard ratio [HR], 0.40; 95% CI, 0.27-0.59).¹ A baseline prediction model that included participant age, intraocular pressure (IOP), central corneal thickness, vertical cup-disc ratio (VCDR), and visual field pattern SD (PSD) accurately predicted which participants would have a low, medium, or high risk of developing POAG.² This prediction model was confirmed in the European Glaucoma Prevention Study and the Diagnostic Innovations in Glaucoma Study.^3,4

The OHTS 2 (conducted from June 3, 2002, to December 30, 2008) was designed to determine whether delaying treatment with medication was detrimental to participants in the observation group (ie, whether delaying treatment initiated a process of glaucomatous deterioration that was difficult to reverse). Both of the original randomization groups received medication during the OHTS 2. The incidence of POAG in the OHTS 2 was found to be generally equal in the 2 randomization groups (HR, 1.06; 95% CI, 0.74-1.50), indicating that delaying treatment in the observation group during the OHTS 1 did not initiate a process of irreversible deterioration in most participants.⁵ Treatment was no longer determined by study protocol beginning in 2009.

This article reports data from the OHTS 3 (conducted from January 7, 2016, to April 15, 2019) regarding the incidence and severity of POAG after 20 years of follow-up or over the lifetime of participants (ie, within 2 years of death). This information may be useful to clinicians and patients in determining the frequency of examinations and tests and the potential benefits of preventive treatment.

Methods

Protocols for the core battery of tests and measures and the determination of POAG end points remained the same in the OHTS 1, OHTS 2, and OHTS 3. The schedule of tests and measures used during each study phase is summarized in eTable 1 in Supplement 1. The OHTS 1 and OHTS 2 study protocols are described in detail elsewhere^5,6 and are summarized in eMethods 1 in Supplement 1; the OHTS 3 study protocol is described in detail in eMethods 2 in Supplement 1. In the OHTS 1, 1636 participants with OHT were enrolled in 22 clinics. Participants signed separate informed consent forms for the OHTS 1, OHTS 2, and OHTS 3; these forms were approved by the local institutional review board of each participating clinic.

A POAG end point required 2 steps: (1) confirmation of optic disc deterioration and/or visual field abnormality that was identified by masked readers at reading centers through repeat testing, and (2) determination that this deterioration and/or abnormality was associated with POAG based on the assessment of a masked end point committee.⁷ In the OHTS 1 and the OHTS 2, participants who developed POAG continued to participate in the study, but treatment was not determined by study protocol. Visual field testing results and photographs obtained from outside sources underwent the same review process. In the OHTS 3, optical coherence tomographic (OCT) measurements of the retinal nerve fiber layer and macula were performed but were not used to determine an end point to ensure that the end point determination was consistent across the OHTS 1, OHTS 2, and OHTS 3.

Data collection for the 20-year follow-up of participants in the OHTS 3 was performed from January 7, 2016, to April 15, 2019. The Battelle Memorial Institute Tracing Department assisted with retrieval of participant vital status and contact information using multiple proprietary databases. We assumed participants were alive unless death was confirmed by obituary or by registration in the National Death Index or the Social Security Death Index. Assessment of visual function over the participants’ lifetimes or after 20 years required (1) OHTS examination, interim clinical data, or medical records within 2 years of death or (2) data after January 2013 that included OHTS 3 examination or medical records released by participants or their legally authorized representatives. Quality-of-life surveys were completed at OHTS 3 visits or by telephone, and survey findings will be reported in a future article.

Clinical status was assessed by visual field mean deviation (MD), visual field PSD, foveal sensitivity (using the Humphrey Field Analyzer; Carl Zeiss Meditech), Early Treatment Diabetic Retinopathy Study and Snellen visual acuity measures, Pelli-Robson contrast sensitivity testing, VCDR from optic disc images, and OCT assessment of peripapillary retinal nerve fiber layer. To accommodate the change from the Humphrey full-threshold test algorithm to the Swedish interactive threshold algorithm, 1.2 dB was added to the full-threshold test points to make them comparable with the Swedish interactive threshold algorithm standard test points.^8,9

Statistical Analysis

Statistical methods are detailed in eMethods 3 in Supplement 1. In brief, the conversion rate to POAG was calculated using 2 methods: (1) percentage of the original cohort of 1636 participants without adjustment for exposure time and (2) incidence of POAG with adjustment for person-years of exposure time. The 20-year cumulative incidences of POAG at 5, 10, 15, and 20 years were estimated using Kaplan-Meier product limit methods and were reported by baseline risk tertile (low, medium, or high), randomization group (medication or observation), and participant race (Black/African American or other race, including American Indian or Alaska Native, Asian or Pacific Islander, and Hispanic). Kaplan-Meier product limit methods with interval censoring time were used to estimate the 20-year incidence of POAG.¹⁰ Statistical analysis was performed using SAS software, version 9.4 (SAS Institute Inc) and the R survfit package, version 2.11-4 (R Foundation for Statistical Computing). The significance threshold was P < .05. Data were analyzed from July 2019 to December 2020.

Results

Follow-up

A total of 1636 participants (mean [SD] age, 55.4 [9.6] years; 931 women [56.9%]; 1138 White participants [69.6%]; 407 Black/African American participants [24.9%]) were enrolled in the original OHTS 1 cohort. A total of 515 participants (31.5%) died over the course of the OHTS. An assessment of clinical status after January 1, 2013, or within 2 years of death was available for 971 of 1636 participants from the original cohort (59.4%), representing 718 of 1121 surviving participants (64.0%) and 253 of 515 deceased participants (49.1%) (Figure 1). The median duration of follow-up was 20.2 years (range, 0-24.5 years). Sources of data for assessing the clinical status of the 971 participants included in phase 3 (59.4% of the original cohort) included an OHTS 3 examination (661 participants [40.4%]), interim clinical data or medical records (85 participants [5.2%]), and an OHTS 1 or OHTS 2 examination within 2 years of death (225 participants [13.8%]) (eTable 2 in Supplement 1). The mean (SD) age of participants in the OHTS 3 was 73.8 (8.4) years; 427 participants (44.0%) were male, and 220 participants (22.7%) were Black/African American. Clinically meaningful differences between participants who did and did not receive ophthalmic examinations and visual function assessments were not detected, with the exception of age at enrollment and race (eTable 3 in Supplement 1).

Figure 1. Participant Flowchart for Ocular Hypertension Treatment Study (OHTS) Phases 1, 2, and 3.

POAG indicates primary open-angle glaucoma.

^aData collection period.

^bAssessment of visual function required an OHTS examination or ocular and medical history after January 1, 2013, or within 2 years of death.

^cThe cumulative number of deaths from February 1994 to April 2019 was 515, which includes 225 participants with OHTS 1 and OHTS 2 visits or medical history within 2 years of death, 19 participants who died after completing an OHTS 3 visit, and 9 participants who died but provided medical history after January 2013.

The mean (SD) of right and left eyes for eye-specific measures were as follows: for IOP, 18.4 (4.5) mm Hg; for VCDR, 0.52 (0.20); for visual field MD, −2.53 (4.64) dB; for visual field PSD, 2.73 (1.96) dB; for foveal sensitivity, 30.8 (10.2) dB; and for mean logMAR visual acuity, 0.09 (0.17; Snellen equivalent, 20/25). Change in clinical status from baseline to 20 years or within 2 years of death did not differ between participants who were originally randomized to the observation or medication groups (eTable 4 in Supplement 1).

20-Year Cumulative Incidence

Among the original cohort of 1636 participants in the OHTS 1, 483 participants (29.5%) developed POAG in 1 or both eyes without adjustment for exposure time; 199 participants (12.2%) developed optic disc deterioration in 1 or both eyes without visual field loss (based on study criteria), and 284 participants (17.4%) developed visual field loss in 1 or both eyes with or without optic disc deterioration (Table 1).

Table 1. Participants in the Original Cohort Who Developed POAG Diagnosed by Optic Disc Deterioration or VF Abnormality.

POAG status	Eyes affected, No.	Participants, No. (%) (N = 1636)a
Any POAG	1 or 2	483 (29.5)
No POAG in either eye	0	1141 (69.7)
POAG optic disc deterioration and no VF POAG in either eye	1	126 (7.7)
	2	73 (4.5)
VF POAG and no POAG optic disc deterioration in either eye	1	55 (3.4)
	2	15 (0.9)
VF POAG and POAG optic disc deterioration in same eye	1	126 (7.7)
	2	78 (4.8)
VF POAG in one eye and POAG optic disc deterioration in fellow eye	1	10 (0.6)
Other forms of glaucomab	1	4 (0.2)
	2	8 (0.5)

Abbreviation: POAG, primary open-angle glaucoma; VF, visual field.

^aPercentages were not adjusted for exposure time.

^bA total of 5 participants had pseudoexfoliation, 3 participants had angle-closure glaucoma, 3 participants had combined-mechanism glaucoma, and 1 participant had pigmentary glaucoma.

The 20-year cumulative incidence of POAG in 1 or both eyes, adjusted per 21 864 person-years of exposure, was 45.6% (95% CI, 42.3%-48.8%; annual incidence, 2.2% based on Kaplan-Meier estimation) among all participants, 49.3% (95% CI, 44.5%-53.8%) among participants in the observation group, and 41.9% (95% CI, 37.2%-46.3%) among participants in the medication group. The incidence of POAG was higher in the observation group during the OHTS 1, but this difference remained relatively constant thereafter (Figure 2A and Table 2). The 20-year cumulative incidence among Black/African American participants was higher (55.2% [95% CI, 47.9%-61.5%]) compared with participants of other races (42.7% [95% CI, 38.9%-46.3%]); P < .001) (Table 2; Figure 2B). The 20-year incidence of visual field POAG was 25.2% (95% CI, 22.5%-27.8%).

Figure 2. Kaplan-Meier Curves of Cumulative Incidence of Primary Open-Angle Glaucoma (POAG) at 5, 10, 15, and 20 Years.

A, Cumulative incidence of POAG by treatment group. B, Cumulative incidence of POAG by race. C, Cumulative incidence of POAG by risk tertile. The whiskers indicate 95% CIs.

Table 2. Cumulative Incidence of Primary Open-Angle Glaucoma Adjusted for Exposure Time^a .

Variable	Cumulative incidence, % (95% CI)
	5 y	10 y	15 y	20 y
Observation group	7.7 (5.8-9.5)	17.5 (14.7-20.2)	26.3 (22.9-29.7)	49.3 (44.5-53.8)
Medication group	4.0 (2.6-5.4)	9.4 (7.3-11.5)	20.2 (17.0-23.3)	41.9 (37.2-46.3)
Black/African American	8.1 (5.3-10.8)	19.9 (15.6-24.0)	31.6 (26.2-36.7)	55.2 (47.9-61.5)
Other raceb	5.1 (3.8-6.4)	11.4 (9.5-13.2)	20.7 (18.1-23.2)	42.7 (38.9-46.3)
Low-risk tertile	1.3 (0.3-2.3)	3.6 (1.9-5.2)	11.4 (8.3-14.4)	31.7 (26.4-36.6)
Medium-risk tertile	4.3 (2.5-6.1)	10.9 (8.1-13.7)	21.9 (17.7-25.8)	47.6 (41.6-53.0)
High-risk tertile	11.9 (9.1-14.7)	26.1 (22.0-29.9)	37.3 (32.5-41.7)	59.8 (53.1-65.5)

^aAdjusted per 21 864 person-years of exposure.

^bOther race includes American Indian or Alaska Native, Asian or Pacific Islander, and Hispanic.

A baseline prediction model was used to categorize participants into low-, medium-, and high-risk tertiles (Table 2; Figure 2C). The 20-year cumulative incidences of POAG, adjusted for exposure time, were 31.7% (95% CI, 26.4%-36.6%) for participants in the low-risk tertile, 47.6% (95% CI, 41.6%-53.0%) for participants in the medium-risk tertile, and 59.8% (95% CI, 53.1%-65.5%) for participants in the high-risk tertile (Table 2). These 20-year cumulative incidences represented significant differences between low-risk vs medium-risk tertiles (P < .001), low-risk vs high-risk tertiles (P < .001), and medium-risk vs high-risk tertiles (P < .001). Cumulative incidences of POAG among the risk tertiles were different at 5, 10, 15, and 20 years.

During the entire study, 11 of 1636 participants (0.7%) developed clinical evidence of pseudoexfoliation in 1 or both eyes (an exclusion criterion at baseline); of those, 5 participants (45.5%) developed pseudoexfoliation glaucoma. Although 89 of 1636 participants (5.4%) had laser iridotomy, the treating clinicians and end point committee diagnosed only 3 of 89 participants (3.4%) with angle-closure glaucoma and 3 of 89 participants (3.4%) with combined-mechanism glaucoma. The other 83 of 89 participants (93.3%) had laser iridotomy for narrow angles only. One of 1636 participants (0.1%) developed pigmentary glaucoma. Data from these 12 participants were censored at the time of diagnosis and were not included in the cumulative incidence of POAG or in other analyses of POAG.

Severity

Clinical measures after 20 years of follow-up or within 2 years of death for the eye with the better and worse visual field MD grouped by 5 categories of POAG status are shown in Table 3. Reliable visual field data were available for 933 of 1636 participants (57.0%). Among the 565 participants who did not develop POAG, the mean (SD) visual field MD was −0.6 (2.7) dB in the better eye and −2.2 (4.4) dB in the worse eye (Table 3). Among the 62 participants who developed bilateral POAG optic disc deterioration without detectable POAG visual field loss (based on study criteria), the mean (SD) visual field MD was −1.5 (4.1) dB in the better eye and −2.9 (5.0) dB in the worse eye. Among the 91 participants who developed unilateral POAG visual field loss and POAG disc deterioration in the same eye, the mean (SD) visual field MD was −2.6 (4.8) dB in the better eye and −7.4 (7.2) dB in the worse eye. Among the 63 participants who developed both bilateral POAG visual field loss and bilateral POAG disc deterioration, the mean (SD) visual field MD was −6.8 (5.8) dB in the better eye and −12.6 (9.0) dB in the worse eye. Distribution of visual field loss severity using a modified Hodapp-Parrish-Anderson scale is shown in eTable 5 in Supplement 1.¹¹ Other measures of clinical status by better and worse eye by visual field MD are shown in Table 3.

Table 3. Visual Function Assessed After 20 Years or Within 2 Years of Death.

POAG status	No.		Mean (SD)
	Eyes affected	Participants	Eye with better MD	Eye with worse MD
MD, dB
No POAG in either eyea	0	565	−0.6 (2.7)	−2.2 (4.4)
POAG optic disc deterioration and no POAG VF loss in either eye	1	103	−0.9 (2.8)	−2.2 (3.7)
	2	62	−1.5 (4.1)	−2.9 (5.0)
VF POAG and no POAG optic disc deterioration in either eye	1	33	−1.8 (3.3)	−4.2 (4.7)
	2	9	−5.9 (4.9)	−10.6 (6.5)
VF POAG and POAG optic disc deterioration in same eye	1	91	−2.6 (4.8)	−7.4 (7.2)
	2	63	−6.8 (5.8)	−12.6 (9.0)
VF POAG in 1 eye and POAG optic disc deterioration in other eye	1	7	−1.1 (0.8)	−5.2 (4.2)
PSD, dB
No POAG in either eyea	0	565	2.0 (1.3)	2.4 (1.9)
POAG optic disc deterioration and no POAG VF loss in either eye	1	103	1.8 (0.8)	2.2 (1.5)
	2	62	2.2 (2.2)	2.7 (2.7)
VF POAG and no POAG optic disc deterioration in either eye	1	33	2.7 (1.9)	4.0 (2.6)
	2	9	5.6 (2.8)	7.2 (2.5)
VF POAG and POAG optic disc deterioration in same eye	1	91	2.4 (1.8)	5.0 (3.0)
	2	63	5.3 (2.8)	6.4 (2.8)
VF POAG in 1 eye and POAG optic disc deterioration in other eye	1	7	2.0 (0.3)	5.8 (3.9)
Foveal sensitivity, dB
No POAG in either eyea	0	562	32.2 (9.8)	30.7 (10.8)
POAG optic disc deterioration and no POAG VF loss in either eye	1	103	32.1 (9.9)	30.2 (11.6)
	2	62	31.8 (10.8)	30.9 (10.5)
VF POAG and no POAG optic disc deterioration in either eye	1	33	29.7 (12.6)	28.2 (12.8)
	2	9	29.2 (12.1)	31.1 (4.6)
VF POAG and POAG optic disc deterioration in same eye	1	91	32.0 (9.7)	29.1 (12.2)
	2	63	28.7 (12.2)	25.3 (13.4)
VF POAG in 1 eye and POAG optic disc deterioration in other eye	1	7	28.9 (13.5)	27.1 (14.7)
Visual acuity
No POAG in either eyea
logMAR	0	419	0.06 (0.12)	0.10 (0.19)
Snellen equivalent	NA	NA	20/23	20/25
POAG optic disc deterioration and no POAG VF loss in either eye
logMAR	1	94	0.05 (0.12)	0.07 (0.14)
Snellen equivalent	NA	NA	20/22	20/32
logMAR	2	59	0.02 (0.08)	0.04 (0.14)
Snellen equivalent	NA	NA	20/21	20/22
VF POAG and no POAG optic disc deterioration in either eye
logMAR	1	22	0.08 (0.16)	0.09 (0.13)
Snellen equivalent	NA	NA	20/24	20/25
logMAR	2	4	0.37 (0.45)	0.35 (0.41)
Snellen equivalent	NA	NA	20/47	20/45
VF POAG and POAG optic disc deterioration in same eye
logMAR	1	80	0.11 (0.24)	0.15 (0.27)
Snellen equivalent	NA	NA	20/26	20/28
logMAR	2	61	0.11 (0.17)	0.17 (0.25)
Snellen equivalent	NA	NA	20/26	20/30
VF POAG in 1 eye and POAG optic disc deterioration in other eye
logMAR	1	6	0.06 (0.07)	0.06 (0.10)
Snellen equivalent	NA	NA	20/23	20/23
Contrast sensitivity (logCS)
No POAG in either eyea	0	336	1.6 (0.3)	1.5 (0.3)
POAG optic disc deterioration and no POAG VF loss in either eye	1	88	1.6 (0.2)	1.5 (0.3)
	2	60	1.6 (0.2)	1.6 (0.2)
VF POAG and no POAG optic disc deterioration in either eye	1	21	1.6 (0.3)	1.6 (0.3)
	2	5	1.3 (0.2)	1.3 (0.3)
VF POAG and POAG optic disc deterioration in same eye	1	78	1.6 (0.3)	1.4 (0.4)
	2	59	1.4 (0.4)	1.3 (0.5)
VF POAG in 1 eye and POAG optic disc deterioration in other eye	1	6	1.5 (0.2)	1.5 (0.3)
Spectralis RNFL, μm
No POAG in either eyea	0	102	88.9 (12.6)	88.5 (13.1)
POAG optic disc deterioration and no POAG VF loss in either eye	1	31	81.6 (13.4)	81.9 (14.6)
	2	22	86.0 (23.5)	76.5 (11.4)
VF POAG and no POAG optic disc deterioration in either eye	1	4	97.8 (4.5)	93.3 (2.9)
	2	1	115.0	81.0
VF POAG and POAG optic disc deterioration in same eye	1	27	78.8 (15.1)	69.4 (13.0)
	2	24	64.6 (15.6)	56.1 (11.7)
VF POAG in 1 eye and POAG optic disc deterioration in other eye	1	3	70.0 (20.0)	57.7 (13.6)
Cirrus RNFL, μm
No POAG in either eyea	0	217	85.8 (10.3)	84.6 (10.6)
POAG optic disc deterioration and no POAG VF loss in either eye	1	56	82.6 (9.7)	78.6 (11.7)
	2	34	75.9 (10.3)	73.7 (11.5)
VF POAG and no POAG optic disc deterioration in either eye	1	16	79.3 (13.6)	74.9 (13.2)
	2	3	64.0 (6.0)	61.7 (9.5)
VF POAG and POAG optic disc deterioration in same eye	1	44	75.7 (9.1)	67.1 (8.2)
	2	42	66.8 (9.8)	62.8 (8.6)
VF POAG in 1 eye and POAG optic disc deterioration in other eye	1	4	82.8 (11.6)	79.6 (13.7)

Abbreviations: logCS, logarithm of contrast sensitivity; MD, mean deviation; NA, not applicable; POAG, primary open-angle glaucoma; PSD, pattern standard deviation; RNFL, retinal nerve fiber layer; VF, visual field.

^aIncludes participants who did not develop POAG in either eye. These participants may have had other ocular and systemic conditions that could have affected visual function.

Visual Impairment and Glaucoma Therapy

At 20 years of follow-up or within 2 years of death, best-corrected visual acuity that was worse than 20/40 and associated with any factor occurred in 85 of 775 participants (11.0%); of those, 67 participants (8.6%) had visual acuity worse than 20/40 in 1 eye, and 18 participants (2.3%) had visual acuity worse than 20/40 in both eyes. Visual acuity worse than 20/200 in 1 eye was noted in 9 of 775 participants (1.2%). Visual field MD that was worse than or equal to −22 dB and associated with any factor occurred in 31 of 957 participants (3.2%); 24 participants (2.5%) and 7 participants (0.7%) had MD that was worse than or equal to −22 dB in 1 eye and both eyes, respectively.

After 20 years of follow-up or within 2 years of death, 696 of 967 participants (72.0%) were receiving ocular hypotensive medication, as documented by an OHTS visit, medical history, or telephone survey. Among those, 397 of 488 participants (81.4%) were originally randomized to the medication group, and 317 of 479 participants (66.2%; P = .001) were originally randomized to the observation group. Medication receipt was higher among Black/African American participants (169 of 219 participants [77.2%]) than among participants of other races (527 of 748 participants [70.5%]; P = .05). Medication receipt was also higher among participants who developed POAG (304 of 372 participants [81.7%]) compared with those who did not develop POAG (392 of 595 participants [65.9%]; P < .001).

Glaucoma surgery was not specified in the OHTS protocol. However, 296 of 1636 participants (18.1%) underwent 1 or more of the following surgical procedures for the treatment of glaucoma during the OHTS: laser trabeculoplasty (157 participants [9.6%]), laser iridotomy (89 participants [5.4%]), filtering surgery (55 participants [3.4%]), combined cataract and filtering surgery (55 participants [3.4%]), and tube shunt surgery (16 participants [1.0%]). The incidence of glaucoma surgery among participants originally randomized to the medication group was 146 of 817 participants (17.9%) vs 150 of 819 participants (18.3%; χ² P = .81) originally randomized to the observation group. The incidence of glaucoma surgery among Black/African American participants was slightly higher (86 of 407 participants [21.1%]) compared with participants of other races (210 of 1229 participants [17.1%]; P = .07).

Discussion

The OHTS 3 was a cohort study designed to determine the incidence and severity of POAG after 20 years of follow-up or within 2 years of death among a cohort of participants with OHT with a median age of 55 years at baseline. Of the 1636 participants enrolled, 483 participants (29.5%) developed POAG in 1 or both eyes over a median follow-up of 20.2 years (range, 0-24.5 years). The cumulative incidence of POAG at 20 years, adjusted for exposure time, was 45.6%. The Kaplan-Meier estimator was used to censor participants at the time of death because they were no longer at risk of developing POAG (if they had not developed POAG before death), which explains much of the difference between the unadjusted incidence of 29.5% and the adjusted incidence of 45.6%. Over the course of the OHTS, 515 participants died. It is concerning that the number of participants who died and the number who developed POAG were approximately the same during the OHTS. The incidence of POAG appeared to be generally linear over 20 years, with a possible modest increase in the rate of conversion after 15 years. We cannot determine whether this increase was a true change associated with increasing age or whether it was associated with missing data from loss to follow-up and death.

One should be cautious in generalizing the incidence of POAG reported in the OHTS cohort to patients seen in standard clinical settings. Several factors have implications for the generalizability of the results. The OHTS participants had a higher risk of developing POAG because the mean baseline IOP of participants in the OHTS was 24.9 mm Hg, and the OHTS intentionally recruited a higher percentage of Black/African American participants (25.0%) than is representative of the general population.¹ Participants in the OHTS may have a lower risk of developing POAG because volunteers in any study may be more likely to return for follow-up visits and adhere to medication. In addition, the OHTS used strict criteria for conversion to POAG. If other diagnostic criteria, such as OCT results, had been included or if less stringent reproducibility criteria had been used, a higher incidence of POAG would have been observed.

Black/African American participants had a higher 20-year cumulative incidence of POAG than participants of other races (55.2% vs 42.7%). This discrepancy occurred despite similar baseline and follow-up IOP and a higher percentage of Black/African American participants in the OHTS 3 reporting receipt of treatment. Higher incidences of POAG among Black/African American participants were also observed in the OHTS 1 and OHTS 2.^1,5 One might conclude that Black/African American individuals should receive more aggressive treatment than others; however, when patients are stratified by baseline risk, Black/African American individuals and others in the same risk category have similar outcomes.⁵ Therefore, the decision about treatment may be better determined based on a predictive model and patient age, health status, life expectancy, and preference rather than race.

Few studies of patients with OHT have reported 20-year outcomes. Lundberg et al¹² found that 11 of 33 patients with OHT (33.3%) developed POAG over 20 years. Hovding and Aasved¹³ reported that 20 of 80 patients with OHT (25.0%) and no pseudoexfoliation developed POAG after 20 years. Oskarsdottir et al¹⁴ followed up 77 high-risk patients with OHT until death. After a mean follow-up of 17.6 years, 7 patients (9.1%) developed bilateral visual impairment, and 2 patients (2.6%) developed bilateral blindness associated with glaucoma. The unadjusted rates of conversion to POAG reported in these studies are similar to those reported in the OHTS 3.

In total, 199 participants developed optic disc POAG deterioration in 1 or both eyes without visual field abnormality based on study criteria. As a group, these participants had few differences from participants who did not develop POAG when comparing the last assessed visual field MD (<1 dB difference) or visual field PSD, visual acuity, contrast sensitivity, or foveal sensitivity. Greater functional differences might have been detected with more rigorous psychophysical, performance, or electrophysiologic tests that were not performed in the OHTS.

A baseline prediction model that included age, IOP, central corneal thickness, VCDR, and visual field PSD differentiated participants with low, medium, or high risk of developing POAG at 20 years (31.7%, 47.6%, and 59.8%, respectively). This model has been useful to clinicians and patients in determining the frequency of examinations and the potential benefits of preventive treatment. The baseline prediction model included VCDR from optic disc photographs, which has been replaced in most practices by OCT measurements of the optic disc, peripapillary nerve fiber layer, and macula. Visual field testing and evaluation have also evolved over 20 years. Future prediction models will be more accurate as they incorporate genetic, environmental, and clinical factors using artificial intelligence and deep learning algorithms.

A major challenge in interpreting the results of this study is determining how the association of treatment with long-term incidence and severity of POAG should be evaluated. The OHTS 1 and OHTS 2 were performed as randomized clinical trials; however, treatment was continued or stopped thereafter at the discretion of clinicians and participants. In the OHTS 3, more than 70% of patients were receiving topical ocular hypotensive treatment, and some patients had undergone glaucoma surgery. Treatment in the OHTS 1 was associated with a greater than 50% reduction in the 5-year incidence of POAG.¹ A similar 50% reduction in the incidence of POAG was projected in a 10-year model.¹⁵ The OHTS 3 did not provide data to extrapolate the 10-year model of treatment vs observation to 20 years, although it seems likely that long-term decreases in IOP would continue to be associated with reductions in the incidence of POAG.

Strengths and Limitations

This study has several strengths. These include a large sample, a diverse patient cohort, careful follow-up procedures, an onset date of glaucoma diagnosis within 6 months in the OHTS 1 and OHTS 2, use of standardized tests, and masked determination of end points by reading centers and an end point committee.

This study also has limitations. These include a limited baseline IOP range of 21 to 32 mm Hg, collection of medical history and family history of glaucoma only by self-report, loss to follow-up over the 20-year study period, missing data on participants (particularly between the OHTS 2 and the OHTS 3), nonrandomized treatment after the OHTS 2, and lack of OCT measurements until the OHTS 3.

Conclusions

The cumulative 20-year incidence of POAG was 45.6%, with 25.2% of participants developing visual field loss in 1 or both eyes after adjustment for exposure time. These findings, combined with the use of a prediction model, may help to inform clinicians and patients about the prognosis of OHT and facilitate patient-specific management that takes into consideration age, health status, life expectancy, and personal preference.

Supplement 1.

eMethods 1. Synopsis of OHTS 1 and OHTS 2 Protocols

eMethods 2. OHTS 3 Protocol: Chapter 2 Manual of Procedures

eMethods 3. OHTS 3 Statistical Methods

eTable 1. Frequency of Tests and Measures in OHTS 1, 2, and 3

eTable 2. Participants in Original Cohort With Visual Function Assessed After January 1, 2013, or Within 2 Years of Death

eTable 3. Comparison of Participants With and Without Visual Function Assessed After January 1, 2013, or Within 2 Years of Death

eTable 4. Demographic Characteristics and Visual Function Measures by Randomization Group at Baseline and at Assessment After January 1, 2013, or Within 2 Years of Death

eTable 5. Distribution of Severity of Visual Field Loss Among Participants Who Developed Visual Field Primary Open-Angle Glaucoma Assessed on Visual Field Test After January 1, 2013, or Within 2 Years of Death

Supplement 2.

Nonauthor Collaborators. The Ocular Hypertension Study Group Investigators, Coordinators, and Executive Committe