OHTS
Goals
OHTS1,2 was designed to answer two key questions: (1) Does reducing IOP in ocular hypertensive eyes reduce their risk of developing primary open-angle glaucoma? and (2) Are there identifiable features in ocular hypertensives that predict who is at an increased risk of developing primary open-angle glaucoma?
To answer these questions, the OHTS investigators recruited 1,636 patients with ocular hypertension for the study. All subjects had an elevated IOP in both eyes (at least 24 mm Hg in one eye and at least 21 mm Hg in the other), normal white-on-white visual fields, and optic nerves without any evidence of glaucomatous damage. The investigators randomized subjects into two groups: (1) observation (those patients whose IOPs were left untreated) and (2) treatment (with a 20% IOP-reduction target). By study design, investigators could use any available medication to reduce patients' IOP in order to determine if decreasing eye pressure would prevent glaucomatous development.
Once enrolled, patients were examined every 6 months for at least 5 years. Investigators obtained white-on-white visual fields at every visit and stereoscopic disc photographs annually. The visual fields and photographs were sent to reading centers for masked review in order to determine whether any changes from baseline had occurred. Study endpoints were either new and reproducible visual field defects or clinically significant and reproducible changes in the appearance of the optic nerve, as judged by the masked reading center personnel.
Results
OHTS demonstrated that lowering the IOP in eyes with ocular hypertension reduced subjects' risk of developing primary open-angle glaucoma over time. Data on the observation group gave us insight into the natural history of untreated ocular hypertension in this well-defined sample population: 9.5% of patients developed glaucoma over the course of 5 years. Data from the treatment group demonstrated that a modest 20% reduction in IOP reduced the 5-year risk of developing glaucoma to 4.4%.
A risk factor analysis yielded both expected and unanticipated results. Not surprisingly, traditional risk factors such as a higher IOP and older age were associated with higher rates of developing glaucoma. Also, patients with larger cup-to-disc ratios (greater than 0.5) were at a higher risk for developing glaucoma. (This might actually represent patients with early [preperimetric] glaucoma at baseline who were enrolled in the study with very early glaucoma.)
A finding that has generated great interest in the field is the important role of central corneal thickness in determining a given patient's 5-year risk of developing glaucoma; patients with thinner corneas were at greater risk. For example, a patient with a central corneal thickness of less than 555 µm and a baseline IOP of greater than 25.75 mm Hg has a 5-year risk of 36%, while a similar patient whose baseline IOP is also greater than 25.75 mm Hg but whose central corneal thickness is greater than 588 µm has a 6% 5-year risk of developing glaucoma.
Why should central corneal thickness be related to the risk of glaucoma? Goldmann applanation tonometry underestimates the true IOP in thinner corneas and overestimates it in thicker corneas. Subjects with thinner corneas may therefore have had an IOP higher than that measured by applanation. Corneal thickness was an independent risk factor in OHTS, however, and errors in IOP measurement cannot fully explain the finding. Another intriguing explanation for an association between central corneal thickness and the development of glaucoma is that eyes with thin corneas may also have less structural support from connective tissue at crucial places such as the lamina cribrosa. A weaker lamina could create greater susceptibility to mechanical deformation and damage from elevated IOP. The relationship between central corneal thickness and optic nerve susceptibility to IOP remains to be investigated.
Interestingly, OHTS did not identify African American race or a positive family history of glaucoma as risk factors for converting to glaucoma in a multivariate analysis. African-Americans were more likely than whites, however, to have thinner corneas and higher initial IOP levels. We cannot, therefore, reject race or family history as risk factors based on the OHTS results.
Implications
Many ophthalmologists have dismissed the notion of treating ocular hypertension. They prefer instead to conduct careful, ongoing surveillance of these patients and to treat only those who demonstrate a conversion to glaucoma, an approach they feel will avoid unnecessary treatment. Moreover, most astute clinicians believe that they can detect the early signs of developing glaucoma before it affects visual function to the extent of impairing patients' quality of life.
Glaucoma experts, however, generally hold the belief (which has not been scientifically established) that an optic nerve's susceptibility to further damage increases once it experiences some degree of glaucomatous damage. In other words, perhaps cupping begets cupping. For that reason, the more advanced the damage, the lower we set the target IOP.
OHTS does not tell us whom to treat, but it does demonstrate that treatment can decrease the incidence of conversion to glaucoma. Furthermore, the study has pinpointed the risk factors that identify the patients who are likely at the highest risk for developing glaucoma. These factors include age, untreated IOP, vertical cup:disc diameter, and central corneal thickness. The chart of every patient with ocular hypertension should prominently feature the values for each of these categories, and clinicians not now obtaining pachymetry readings on patients with ocular hypertension probably soon will be.
Although OHTS has not generated a risk calculator, we can formulate some general principles based on its results. We should treat patients who are at high risk for developing glaucoma and young enough that they may develop a noticeable visual defect during their lifetime. We may observe patients at low risk and those unlikely to survive long enough to develop a visual disability but treat them at the earliest signs of glaucomatous damage. It is more difficult to decide how to manage patients at moderate risk, but it is appropriate to engage them in an informed discussion about the risks and benefits of therapy.
EMGT
Goals
EMGT3 is a randomized, controlled clinical trial that compared the effect on the progression of newly detected open-angle glaucoma of immediately lowering the IOP versus no treatment or later treatment. The study was designed in the early 1990s when controversy existed regarding the timing and appropriate degree of aggressiveness of treatment and when practitioners were openly discussing the uncertainty of treatment effect.4
The study enrolled 255 patients aged 50 to 80 years (median=68 years). All had early glaucoma, visual field defects (median mean deviation= -4 dB), and a median IOP of 20 mm Hg. Patients with an IOP greater than 30 mm Hg or advanced visual field loss were excluded.
Instead of setting a target IOP level and advancing therapy to achieve that goal, investigators performed 360º laser trabeculoplasty on and administered topical betaxolol hydrochloride to all treated patients (n=129). If the IOP in treated eyes exceeded 25 mm Hg at two consecutive follow-up visits or 35 mm Hg in control eyes, the investigators prescribed latanoprost eye drops dosed once daily. The EMGT differs from other recent glaucoma studies in that it did not set a target IOP. Despite the fact that standard treatment was administered rather than treatment to a target, the treatment group achieved and sustained an effective IOP (mean reduction was 5.1 mm Hg or 25%).
Investigators followed all subjects with Humphrey Full Threshold 30-2 visual field tests (Zeiss Humphrey Systems, Dublin, CA) and tonometry every 3 months and optic disc photography every 6 months. Once they detected progression, investigators could initiate treatment in the control group or advance it in subjects of either group already receiving treatment.
The endpoints for progression were visual field progression or optic nerve progression. Investigators determined the former using glaucoma change probability maps. Progression was defined as at least three points showing progression at the same location on three consecutive glaucoma change probability maps compared with the average of two baseline fields. The visual field analysis using glaucoma change probability maps has not been employed in other clinical trials, but it might be more sensitive to subtle change than other visual field grading systems. Masked readers evaluated the optic nerves for progression.
Results
The baseline untreated IOP was not particularly high at 20.6 mm Hg in the treatment group and 20.9 mm Hg in the control group. After 3 months, the IOP was reduced 25% in the treatment group (15.5 mm Hg) and remained essentially unchanged in the control group (20.8 mm Hg).
In the control group, progression was detected and occurred more frequently (78 of 126 subjects; 62%) than in the initial treatment group (58 of 129 subjects; 45%). This reduction in risk was statistically significant (P=.007). Virtually all patients who reached the endpoint did so by visual field criteria; only one endpoint was reached based on optic nerve photographs.
Implications
The EMGT is the first large, randomized, prospective clinical trial that includes an untreated control arm in order to evaluate the effects of IOP reduction in patients with open-angle glaucoma. Lowering IOP decreased the proportion of subjects who progressed and delayed the progression, but treatment was associated with an increase in nuclear lens opacity (cataract).
One particular strength of this study was its remarkable retention rate. Only 2.4% of the subjects were lost to causes other than death. The study population was not racially diverse, however; most of the subjects were white. This group probably fared better than the general glaucoma population in the US, and they did not fare very well overall: Nearly half of the treatment group progressed.
We still need to understand better the high rate of progression. The visual field endpoint in this study was a new criterion for defining progression. Although glaucoma change probability maps may be highly sensitive and specific indicators of visual field progression, it will be interesting and important to see more published information about this approach. Only then will we know how to interpret the surprisingly high rates of progression detected in both the treatment and control groups. We need to know whether the results are a function of an overly sensitive test, if current techniques for identifying progression are insensitive to small changes, and whether something was wrong with the treatment algorithm.
Although there was a rationale for the treatment algorithm, I doubt that laser trabeculoplasty and betaxolol will become the standard of care. Perhaps most important is the disturbing news that, despite a 25% mean IOP reduction, 45% of the treated patients experienced visual field progression over a median follow-up period of 6 years. If glaucoma change probability maps are more sensitive and specific for glaucoma, we may be able to evaluate better the success or failure of treatment in our patients.
CONCLUSION
In summary, OHTS is a well-designed and conducted study, but the most important question it leaves unanswered is, “Whom should I treat?” Although this decision must be based on clinical judgment, OHTS has highlighted identifying risk factors for us to consider in our ocular hypertension patients. For its part, the EMGT clearly supports the idea that patients with newly diagnosed, relatively early glaucoma benefit from a reduction in IOP, but the study does not tell us how to treat these patients. Future reports from these studies may provide us with further guidance.
Robert D. Fechtner, MD, is Professor of Ophthalmology at the New Jersey Medical School in Newark, New Jersey. Dr. Fechtner may be reached at (973) 972-2030; fechtner@umdnj.edu.
Paul J. Lama, MD, is Assistant Professor of Ophthalmology at the New Jersey Medical School in Newark, New Jersey. Dr. Lama may be reached at (973) 972-9467; lamapj@umdnj.edu.
1. 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 Ophthalmol. 2002;120:701-713.
2. 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;120:714-720.
3. Heijl A, Leske MC, Bengtsson B, et al. Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial. Arch Ophthalmol. 2002;120:10:1268-1279.
4. Eddy DM, Billings J. The quality of medical evidence: implications for quality of care. Health Aff (Millwood). 1988;7:1:19-32.
