Year : 2000 | Volume
: 48 | Issue : 2 | Page : 107--11
The role of central corneal thickness in the diagnosis of glaucoma
R Thomas, S Korah, J Muliyil
Schell Eye Hospital, Department of Ophthalmology, Vellore, India
Schell Eye Hospital, Department of Ophthalmology, Vellore
Purpose: To determine the effect of central corneal thickness (CCT) on applanation tonometry and any resultant misclassification of normals as ocular hypertension.
Method: The central corneal thickness was measured using the ultrasound pachometer in 50 normals, 25 glaucoma and 23 ocular hypertensive patients. The student«SQ»s «DQ»t«DQ» test was used to determine any significant difference in CCT between the three groups.
Results: There was a statistically significant difference in the mean CCT of the ocular hypertensives ( 0.574 + 0.033mm) as compared to the glaucomas (0.534 ± 0.030mm) and normals (0.537 ± 0.034mm). Applying the described correction factor for corneal thickness, 39% of eyes with ocular hypertension were found to have a corrected IOP of 21mmHg or less.
Conclusions: Increased corneal thickness in ocular hypertension may lead to an overestimation of IOP in 39% of cases. Measurement of central corneal thickness is advisable when the clinical findings do not correlate with the applanation IOP.
|How to cite this article:|
Thomas R, Korah S, Muliyil J. The role of central corneal thickness in the diagnosis of glaucoma.Indian J Ophthalmol 2000;48:107-11
|How to cite this URL:|
Thomas R, Korah S, Muliyil J. The role of central corneal thickness in the diagnosis of glaucoma. Indian J Ophthalmol [serial online] 2000 [cited 2021 Jun 22 ];48:107-11
Available from: https://www.ijo.in/text.asp?2000/48/2/107/14894
Glaucoma is defined as an optic neuropathy characterized by a typical appearance of the optic nerve head and characteristic visual field loss. The diagnosis of glaucoma is based on a combination of factors including intraocular pressure (IOP), optic disc (and nerve fibre layer) damage and specific field defects. Raised IOP is the only known causal risk factor; certainly it is the only risk factor that can be therapeutically manipulated. Goldmann applanation tonometry, the current gold standard for the measurement of IOP, is based on the Imbert-Fick law. Goldmann observed that when the area applanated was 7.35mm2, the surface tension due to the tear film counterbalanced the resistance to indentation of the cornea, thus making it unnecessary to consider the rigidity of the globe and the surface tension of the tear film in applanation tonometry. More recent evidence indicates that these, as well as a number of other factors (e.g. significant astigmatism, corneal curvature) do affect the accuracy of applanation tonometry.
Variations in corneal thickness change the resistance of the cornea to indentation so that this is no longer balanced exactly by the tear film surface tension. This may affect the accuracy of the measurement of IOP. A thinner cornea may require less force to applanate it, leading to underestimation of the true IOP, while a thicker cornea would need more force thus giving an artifactually high IOP reading.
Goldmann himself discussed the influence of variations of central corneal thickness (CCT) on IOP measured by applanation. However, he felt that significant variations in CCT occurred only rarely and hence assumed a "normal" CCT of 520 μm for his instrument.
A positive correlation between increased corneal thickness and IOP has been reported earlier.[6-8] Studies in eyes with manometrically controlled IOP have demonstrated a significant disparity between the actual IOP and simultaneous applanation tonometry readings. This disparity was related to the CCT. The underestimation of IOP was as much as 4.9mmHg in thin corneas, while thick corneas produced an overestimation of about 6.8mmHg.[9-11] Accordingly it has been suggested that measurement of corneal thickness is necessary for the accurate interpretation of applanation tonometry.
It has been calculated that applanation tonometry over/underestimated IOP by 5mmHg for every 70 μm corneal thickness. A correction factor for IOP, to adjust for CCT measurements that differ from "normal CCT" was proposed as follows:
Since a cornea can only be so thin before it becomes pathological, the normal variation of CCT in the population (like the IOP), is skewed to the right. The possibility exists that the measured IOP in patients at either end of the spectrum could be an under/over estimation of the actual hydrostatic IOP and this could be a confounding factor in the terminology of glaucoma based only on the cut-off value of 21mmHg.
This study was undertaken to determine whether, in our population, patients classified as "ocular hypertensive" did indeed have thicker corneas compared to normal subjects or true glaucoma patients.
Materials and Methods
Ninety eight patients seen in the outpatient department from June 1996 to October 1998 were included in this masked, controlled, prospective study. Based on criteria given below, three groups of patients were selected: normal (n=50), diagnosed glaucomas currently on treatment (n=25), and ocular hypertensives (n=23).
Group I. Normals: IOP ≤ 21mmHg; normal optic discs; angles open on gonioscopy; and no suspicion of any form of glaucoma.
Group 2, Glaucomas: IOP prior to treatment > 21mmHg; current IOP on treatment ≤ 21mmHg; glaucomatous optic disc ± nerve fibre layer defects; glaucomatous field defects fulfilling at least two of Anderson's criteria on white on white automated static perimetry and correlating with the disc changes; and open angles on gonioscopy.
Patients with appositional chronic angle closure who had undergone laser peripheral iridotomies at least 3 months prior to the study with currently open angles were also included in the study.
Group 3. Ocular hypertensives: IOP >21mmHg on at least two occasions; healthy optic discs with no glaucomatous features; no nerve fibre layer defects; no glaucomatous field defects (none of Anderson's criteria fulfilled) with white-on-white automated perimetry; and open angles on gonioscopy.
- Evidence of other anterior segment pathology including corneal opacities
- Previous intraocular or corneal surgery
- Diabetes mellitus, use of contact lens and any other condition which might affect corneal thickness
- Any other optic nerve or intracranial disease
- Corneal oedema
- Corneal astigmatism > 2D
Sample size calculations
The sample size was calculated from the expected difference in the CCTs in these three populations as determined in the literature.[9,11] In these studies, the standard deviations between the groups studied ranged between 0.02 and 0.04. In order to guarantee the calculated power (80%), we used the larger standard deviation of 0.04 for our calculation.
Based on previous studies, we expected a clinically significant difference of 0.05mm in the CCT between the study groups and decided to settle for 80% power to detect this difference. Accordingly, a Type I error of 5% (α = 0.05) and a Type II error of 20% (β = 0.2) was used.
Sample size was calculated using the following formula:
n = 2 (Zα + Zβ)2 S2 / d2
n = sample size
Zα = Normal deviate of α = 1.96
Zβ = Normal deviate of β = 0.842
S = standard deviation = 0.04
d = expected clinically significant difference = 0.05mm
n = 2(1.96 + 0.842)2 x 0.042 / 0.052 = 10 (to the nearest whole number)
The minimum sample size was 10 eyes in each group.
To increase the power, we included 50 eyes in the normal group and 25 in the glaucoma group. As ocular hypertension (according to our definition) was rarer, we decided to include a minimum of 10 patients, but to aim for as many as would agree to participate in the study. Ultimately 23 ocular hypertensive patients were examined.
After determination of the best corrected visual acuity, a complete ocular examination was performed. This included slitlamp biomicroscopy, applanation tonometry (on at least two occasions), gonioscopy, indirect ophthalmoscopy and stereoscopic examination of the disc and nerve fibre layer using a +60 D or a +78D condensing lens with the slitlamp. Automated perimetry was performed on all glaucomatous and ocular hypertensive patients prior to dilatation using the 30-2 program of the Humphrey Field Analyzer if it had not been done within 3 months of the study.
At the following visit, all patients in the normal category were randomized applying the technique of. permutated block randomization (computer generated blocks of four), to measurements in their right or left eye. The glaucomatous and ocular hypertensive patients had measurements taken for both eyes; for analysis only one eye from each patient was randomly selected. If, however, only one eye of a patient fit the selection criteria, that eye was selected.
Topical 4% lidocaine hydrochloride was instilled in the lower cul-de-sac and the central corneal thickness was taken using the Tomey (model AL 1000; Tomey Corp, Nagoya Aichi, Japan) pachometer. (We have used the original term, pachometer, in preference to the later modified term, pachymeter.14) Five consecutive readings were recorded to the nearest thousandth of a millimeter and an average value obtained.
All pachometry readings were taken in similar manner by a single observer who was masked to the diagnosis. In order to minimize effects of diurnal variation on thickness, all measurements were taken between 2 PM and 5 PM. The CCT values of the three groups were compared using the student's t-test.
The mean CCT was highest in the ocular hypertensive group [Table:1][Figure:1]. The CCT values of the ocular hypertensives were significantly different from those of normals and the glaucomas (p <0.001). The difference in the CCT of the glaucomas as compared to the normals was not statistically significant (p=0.7).The mean age was highest in the glaucoma group (56.84 years); and the ocular hypertensives had the lowest mean age (44.20 years). The mean IOP as expected, was highest in the ocular hypertensive group.
The "adjusted" IOP (using the adjustment factor for corneal thickness described by Ehlers) plotted against the measured IOP in the ocular hypertensive group, is shown in [Figure:2]. Of the ocular hypertensives, 39% had corrected IOP measurements of 21 mmHg or less.
Intraocular pressure is the only causal risk factor for glaucoma and the only one that can be manipulated to alter the course of the disease. It is therefore important that it is measured accurately. Direct manometric measurement of IOP is possible but not practical; for clinical use we have to rely on indirect measurements. Unfortunately, the indirect estimation of any physiological parameter by extrapolation from the measurement of another more easily accessible value is based on assumptions about the relationship between the two. This is bound to be fraught with errors.
Till recently it was not entirely clear whether the relationship between the IOP and corneal thickness was artifactual or real. It may certainly be caused by a measurement error in applanation tonometry due to the differences in corneal thickness. Alternatively, an actual physiological effect of the raised IOP on the cornea may result in, for example, an increase in collagen fibres and a consequent increase in the thickness of the cornea. Manometric studies, however, disproved this argument and found that the higher pressures obtained by applanation tonometry were in fact due to a measurement error induced by the presence of the thicker corneas, and not accurate representations of the true IOP.[1,6]
It may also be argued that the higher IOP challenges the corneal endothelial deturgescense mechanism secondarily causing corneal thickening (oedema). However, simultaneous measurement of IOP by the manometer and applanation tonometry does not appear to support this hypothesis, unless the corneal thickness changes rapidly in response to short-term variation in IOP. Modest increase in IOP (< 40mmHg), even for prolonged periods of time seem to have little effect on the corneal endothelium and corneal thickness. Patients with a healthy endothelium rarely develop corneal oedema unless the pressure elevation is acute or lasts for prolonged periods. In addition, several studies done on the effect of corneal oedema on IOP measurements have shown an artifactually low measurement of IOP in the oedematous cornea. This is different from the increased IOP in the thickened non-oedematous cornea.
The diagnosis of ocular hypertension, primary glaucoma and normal tension glaucoma is made on the basis of an arbitrary IOP cutoff point of 21 mmHg. This cutoff is based on statistical grounds, primarily for screening purposes rather than as a diagnostic criterion. It is nevertheless in clinical use and any factor that alters the value of the IOP can lead to a misclassification of the patient.
The fact that a relatively minor change in CCT can produce a statistically significant change in mean IOP measurement suggests that CCT may be more important in the overall management of glaucoma than previously suspected. This may have an impact on values around the "magic" 21 mmHg. It is obvious from our results that many patients with increased IOP without other glaucomatous features might merely have thicker corneas, but not be at higher risk for glaucoma. That is, there is a certain proportion of perfectly normal people who are labeled as glaucoma suspects or ocular hypertensives.
The values generated for CCT in this study are consistent with the findings of earlier studies.[9-11] A definite direct relationship of the CCT with IOP values was found. The corneas in ocular hypertensives were significantly thicker than those of glaucomas and normals (0.574mm compared to 0.534mm and 0.537mm respectively).
The IOP in the ocular hypertensives corrected for the thicker CCT resulted in corrected IOPs of 21mmHg or less in 39% of ocular hypertensives. If the cutoff value was raised to 22mmHg or less, the number of ocular hypertensives found to have "normal" pressures increased to 12 (52%). Hence at least 9 of our 23 ocular hypertensives (12 if a cutoff of 22 mmHg is used) were probably otherwise normal but labeled as ocular hypertensive because of a CCT-induced error in IOP.
In order to determine any clinical difference between the group of patients in whom the corrected IOP was below 21mmHg and the true ocular hypertensives, we reviewed the records of all ocular hypertensives included in the study. Of the 23 patients, 9 had undergone short-wave automated perimetry. Of these 9, only 3 had significant field defects. Two of the 3 patients with defects remained in the ocular hypertensive category after the correction factor for CCT was applied. A careful disc and nerve fibre layer examination was repeated. However, there did not seem to be any other clinical predictive factor that differentiated the patients in whom the adjusted IOP decreased below 22mmHg from the group in which it remained high. The numbers are, however, too small to draw any meaningful inferences.
The ocular hypertensives in this study were significantly younger than the other two groups, while the glaucoma cohort was the oldest (Mean age: normals 49.54yrs, glaucomas 56.84yrs, ocular hypertensives 44.20yrs). It is known that corneal thickness decreases with age. It is possible therefore that some of the "ocular hypertensives" might become normotensive with age; the ones that continue to have a raised IOP might be the ones more susceptible to glaucomatous damage. This is a testable hypothesis.
The glaucoma group had a corneal thickness no different from the normal group (0.534mm and 0.537mm respectively). Hence the higher pressures recorded in this group were probably accurate. The cohort of glaucoma patients chosen for this study were patients with IOP controlled on medication, in most cases topical timolol maleate 0.5%. The effect of topical drops on corneal thickness is a genuine concern. However, it has been shown that topical drops ( timolol maleate 0.5% twice daily, betaxolol hydrochloride 0.5% twice daily) do not have any significant effect on corneal thickness after 12 months of therapy in patients with a baseline endothelial cell density greater than 1500 cells/ mm2 and a CCT less than 0.68mm.
Patients with normal tension glaucoma were too few to be included in this study. However, the mean corneal thickness in the five patients that we were able to examine was 0.517mm (SD 0.001). Even though the numbers are small, this does suggest that in this group of patients the IOP recording may actually be an underestimation of the true IOP. Others have reported similar findings for low tension glaucoma.[10,18] In the editorial "Tonometry and clinical thickness" Brubaker has suggested that while adjustment of IOP for corneal thickness could result in reclassification of borderline cases, it would rarely alter the decision to treat or not to treat a given case.
Diagnosed glaucoma patients were included in this study if their IOP was stable and controlled with medications. This was done to avoid any physiological changes in CCT as a response to the widely fluctuating pressures known to occur in this population of patients. For the same reason, patients with angle closure glaucoma were excluded, as they are prone to intermittent closure with its attendant acute rise in IOP We, however, included those patients with appositional angle closure in whom the angle was completely open following a YAG laser peripheral iridectomy done at least 3 months prior to inclusion in the study. It has been demonstrated that uncomplicated YAG laser peripheral iridectomies produce no significant change in the cornea! thickness when preoperative measurements are compared to measurements 12 weeks postoperatively.
We used ultrasound pachometry for our study as it has been shown to have better inter-observer and intra-observer variability than the optical method in several studies.[20,21] The inter and intra observer variation of the optical and ultrasound pachometer in our set up will be reported in a subsequent communication. Additionally, readings taken from the central 2-3mm of the cornea have been shown to be more replicable than from paracentral or peripheral locations in the cornea.
The final common pathway in the management of glaucoma is the decision to treat or reduce IOP to a certain level in a given patient. Using Ehler's correction factor, 39% of our ocular hypertensives were found to have a corrected IOP of 21mmHg or less. On applying a more recently described correction, this decreases to 13%. Our study thus confirms that CCT can be a confounding factor while recording IOP. A patient may be labeled an ocular hypertensive just because of the error in measuring his applanation IOP, leading to unnecessary prolonged treatment and/ or follow up. The CCT measurement would go a long way in helping us make a clinically relevant decision.
In conclusion, the measurement of central corneal thickness though perhaps not necessary in all suspected glaucoma patients, may be of value in selected cases in order to improve clinical decision making, especially if the other clinical findings do not seem to correlate with the IOP. This will help prevent the erroneous labeling of normal patients as "ocular hypertensive" and primary open angle patients as "normal tension glaucoma".
|1||Ritch R, Shields MB, Krupin T. The Glaucomas, 2nd edition. St.Louis:C.V. Mosby; 1996. Vol. 1, 2. pp 407-29, 615-57, 859-74.|
|2||Duke-Elders S. System of Ophthalmology. St. Louis:1976. C.V. Mosby; Vol.XI. pp 379-560.|
|3||Schmidt TAF. The clinical application of the Goldmann applanation tonometer. Am J Ophthalmol 1960;49:967-78.|
|4||Whitacre MM, Stein R. Sources of error with use of Goldmann-type tonometers. Surv Ophthalmol 1993;38:l-30.|
|5||Moses RA. The Goldman applanation tonometer. Am J Ophthalmol 1958;46:865-69.|
|6||Johnson M, Kass MA, Moses RA, Grodzki WJ. Increased corneal thickness simulating elevated intraocular pressure. Arch Ophthalmol 1978;96:664-65.|
|7||Hansen FK. A clinical study of the normal human corneal thickness. /Acta Ophthalmol 1971;49:82-88.|
|8||Ehlers N, Bramsen T, Sperling S. Applanation tonometry and central corneal thickness. Acta Ophthalmol 1975;53:34-43.|
|9||Herdon LW, Choudhri S A, Cox T, Damji K T, Shields MR, Allingham RR. Central corneal thickness in normal, glaucomatous and ocular hypertensive eyes. Arch Ophthalmol 1997; 115:1137-41|
|10||Copt RP, Thomas R, Mermoud A. Corneal thickeness in ocular hypertension, primary open angle and normal pressure glancoma. Arch Ophthalmol 1999;117:14-16.|
|11||Wolfs RCW, Klaver CCW, Vingerling JR, Grobbee DE, Hofman A, de Jong PT. Distribution of central corneal thickness and its association with intraocular pressure:The Rotterdam Study. Am J Ophthalmol 1997;123:767-72.|
|12||Anderson DR, Pattela VM. Automated Static Perimetry. 2nd edition. St. Louis:C V Mosby, 1999. pp 152-53.|
|13||Florey CDV. Sample size for beginners. Br Med J 1993; 306:1181-84.|
|14||Whictacre MM, Stein RA, Hassanein K. The effect of corneal thickness on applanation tonometry. Am J Ophthalmol 1993;ll5:592-96.|
|15||Korey M, Gieser D, Kass MA. Central corneal endothelial density and central corneal thickness in ocular hypertension and primary open angle glaucoma. Am J Ophthalmol 1982;94:610-16.|
|16||Ytteborg J, Dohlman CH. Corneal edema and intraocular pressure. Acta Ophthalmol 1965;74:477-84.|
|17||Lass JH, Khosrof SA, Laurence JK, Horwitz B, Ghosh K. A Double-masked, randomized, 1-year study comparing the cornea] effects of dorzolamide, timolol and betaxolol. Arch Ophthalmol 1998;ll6:1003-10.|
|18||Morad Y, Sharon E, Hefetz L, Nemet P. Corneal thickness and curvature in normal tension glaucoma. Am J Ophthalmol 1998;125:164-68.|
|19||Brubaker RF. Tonometry and corneal thickness (editorial). Arch Ophthalmol 1999;117:104-5.|
|20||Giasson C, Forthomme D. Comparision of central thickness measurements between optical and ultrasound pachymeters. Optom Vis Sci 1992;69:236-41.|
|21||Nissen J, Hjortdal JO, Ehlers N. A clinical comparison of optical and ultrasonic pachometry. Acta Ophthalmol 1991;69:659-63.|
|22||Higgins S E, Fishbaugh JA, Strike DJ, Rapuano CJ. Reproducibility and variation of corneal thickness in different locations in the cornea as measured by an ultrasonic pachymeter. Insight 1993;18:14-18.|