Corneal Thickness and Glaucoma

Whether a patient possesses a corneal thickness of greater than 600 µm or less than 500 µm dramatically affects his IOP. The true measurement of IOP may be lower or higher, respectively, than that calculated by applanation tonometry. This article explores research on central corneal thickness and its implications for patient care.

APPLANATION TONOMETRY
Hans Goldmann, MD, conceptualized the cornea as a sheath covered by two membranes between which stable water is located. When first describing the Goldmann Applanation Tonometer (HAAG-STREIT AG, Köniz, Switzerland), Dr. Goldmann and a colleague discussed the effect of central corneal thickness on IOP as measured by the device.¹ They felt that variations in corneal thickness rarely occurred in the absence of corneal disease, and they assumed a central corneal thickness of 520 µm. Nevertheless, they acknowledged that central corneal thickness might, theoretically, influence applanation readings.

It has since become apparent that the central corneal thickness varies to a greater degree among clinically normal patients than Dr. Goldmann realized. Niels Ehlers, MD, PhD, and his colleagues^2-4 performed a number of studies in the 1970s that assessed the effect of central corneal thickness on IOP. In a manometric study,⁴ they demonstrated that the average error of tonometry was 0.71 mm Hg for a 10-µm deviation from the normal central corneal thickness of 520 µm. In another study,⁵ investigators reported on a patient who possessed a central corneal thickness of 900 µm and a manometric IOP of 11 mm Hg. When measured with applanation tonometry, however, this patient's IOP had ranged from 30 to 40 mm Hg while she was receiving maximum medical therapy.

Similarly, a manometric study⁶ with the Perkins tonometer (Veatch Ophthalmic Instruments, Tempe, AZ) demonstrated that the device underestimated IOP by as much as 4.9 mm Hg in thin corneas and as much as 6.8 mm Hg in thick corneas. These findings corresponded to a calculated range of 0.18 to 0.49 mm Hg of change in IOP for a 10-µm change in the central corneal thickness from the mean central corneal thickness.

The reason for these study findings becomes clear when we consider the mechanism of applanation tonometry. Goldmann tonometry measures the force needed to applanate the eye to 3.06 mm diameter. The force required is a combination of that necessary to counter the IOP plus that needed to bend the cornea (minus a small attraction due to surface tension). The thicker the cornea, the greater the force required to bend it. For that reason, a deviation from the normal central corneal thickness results in a potentially incorrect indication of IOP using this measurement technique.

DIFFERENCES AMONG PATIENT GROUPS
A number of studies have evaluated the distribution of central corneal thickness among patients diagnosed with primary open-angle glaucoma (POAG), normal-tension glaucoma (NTG), and ocular hypertension (OHT). This research determined that a significant difference exists between the mean central corneal thicknesses of these three groups. In one study,⁷ the mean central corneal thickness was 554 µm in the control eyes, 550 µm in the eyes with POAG, 514 µm in those with NTG, and 580 µm in those with OHT. Similarly, a study⁸ measuring the central corneal thickness among patients classified as having POAG, NTG, and OHT both confirmed that those with OHT possessed thicker corneas than their control and POAG counterparts and found that patients classified as having NTG had thinner corneas versus POAG and OHT eyes.

CORNEAL THICKNESS AND POAG DEVELOPMENT
The recently published results of the Ocular Hypertension Treatment Study showed that the risk of developing POAG was greatest for study participants with the thinnest central corneas.⁹ Subjects who had central corneal thicknesses of 555 µm or less were at three times greater risk of developing POAG than participants with central corneal thicknesses of higher than 588 µm. The Ocular Hypertension Treatment Study's results have been the most compelling evidence to date that glaucoma practitioners should incorporate pachymetry into their evaluations of patients with OHT.

CORNEAL THICKNESS AND ADVANCED PROGRESSION
Study Objective and Methodology
At the recent ARVO meeting, my colleagues and I presented a paper that examined the effect of central corneal thickness on the degree of POAG severity evident upon patients' initial presentation to a glaucoma specialist.¹⁰ Our hypothesis was that individuals with thinner central corneas would present with more advanced disease compared with those who had thicker corneas. All patients had been evaluated at the Duke University Eye Center, and we retrospectively selected subjects from our research database. We reviewed the initial visit of each consecutive new POAG patient who presented to me from 1997 to 2002. Each subject had to have undergone pachymetry within 1 month of his initial visit. Stereo disc photographs, a detailed optic nerve drawing, or automated visual field testing were also required within 1 month of the initial visit.

For each patient, we recorded age, sex, race, a family history of glaucoma in a first-degree relative, and the presence of diabetes mellitus or systemic hypertension as reported by the patient. For each eye, we recorded Snellen visual acuity, number of glaucoma medications prescribed, spherical equivalent, IOP according to Goldmann applanation tonometry, average central corneal thickness, visual field data where available, and vertical and horizontal cup-to-disc ratios. Visual field data included the Advanced Glaucoma Intervention Study (AGIS) score, mean deviation of visual field, fixation losses, false negative responses, and false positive responses. The AGIS score has been described in detail previously.¹¹ In brief, we graded the visual fields on a scale of 0 to 20 based on the degree of damage on the Total Deviation printout. A score of 0 is a normal visual field; 1 to 5 represents mild disease; 6 to 11 represents moderate disease; 12 to 17 represents severe disease; and 18 to 20 represents end-stage glaucoma.

Results
Of the 434 patient charts reviewed, 350 eyes of 190 patients met the inclusion and exclusion criteria for this study. The mean central corneal thickness was 544 µm in the right eyes and 546 µm in the left eyes. The average AGIS score was approximately 6.2 in both eyes, while the average mean deviation of visual field was -8.3 dB in the right eyes and -8.1 dB in the left eyes. The mean vertical cup-to-disc ratio was approximately 0.80 in both eyes, and the mean horizontal cup-to-disc ratio was 0.76 in both eyes.

When considered alone, the central corneal thickness was a significant predictor of every outcome variable (P<.001 to P=.043). An increase in the corneal thickness was associated with (1) an improved AGIS score, (2) an improved mean deviation of visual field, (3) a decrease in both the vertical and horizontal cup-to-disc ratios, and (4) a decrease in the number of medications used for glaucoma (Table 1). The results of multivariable modeling are shown in Table 2. When we considered all variables simultaneously, the only significant predictor of the AGIS score was the central corneal thickness (P=.001). For an increase of 10 µm of corneal thickness, the AGIS score improves by 0.31 points.

---

CONCLUSION
The implications of the research described in this article are significant. It would appear that large numbers of patients may have been misclassified based on their measured IOP. Many may not be treated as effectively as they could be. Those with OHT may be overtreated, a problem that could be avoided by physicians' performing pachymetry. The appropriate target IOPs of patients with NTG or POAG, meanwhile, could be much lower than what glaucoma specialists have until now considered safe.

The final algorithm needed to convert the measured applanating pressure to the true IOP has yet to be calculated. We now have adequate information, however, to suggest that routinely measuring patients' central corneal thicknesses is critical in cases of suspected glaucoma in order to obtain the correct diagnosis and to ensure these individuals' optimal management.