- Email: [email protected]
Corneal biomechanical properties: Precision and influence on tonometry
ARTICLE IN PRESS
G Model CLAE-658; No. of Pages 8
Contact Lens & Anterior Eye xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Contact Lens & Anterior Eye journal homepage: www.elsevier.com/locate/clae
Corneal biomechanical properties: Precision and influence on tonometry Kelechi C. Ogbuehi ∗ , Uchechukwu L. Osuagwu 1 Cornea Research Chair, Department of Optometry and Vision Sciences College of Applied Medical Sciences, King Saud University, P.O. Box 10219, Riyadh 11433, Saudi Arabia
a r t i c l e
i n f o
Article history: Received 31 October 2012 Received in revised form 19 August 2013 Accepted 16 September 2013 Keywords: Goldmann tonometer Ocular Response Analyzer ORA Noncontact tonometer Repeatability Tonometry.
a b s t r a c t Purpose: To assess the precision and reproducibility of the corneal biomechanical parameters, and their relationships with the intraocular pressure (IOP) measured with the Goldmann tonometer and a noncontact tonometer. Methods: Readings for biomechanical properties and for IOP measured with the Goldmann and noncontact tonometers, were taken on one randomly selected eye of 106 normal subjects, on each one of two measurement sessions. Measurements with the ocular response analyzer (ORA) and the noncontact tonometer were randomized, followed by the measurement of central corneal thickness and with the Goldmann tonometer. Results: Repeatability coefficients for CCT, corneal hysteresis (CH) and corneal resistance factor (CRF) in Session 1 were ±0.01 m, ±3.05 mmHg and ±2.62 mmHg, respectively. The mean CCT, CH, CRF, Goldmann and noncontact tonometry did not vary significantly between sessions. Reproducibility coefficients for CCT, CH and CRF were ±0.02 m, ±2.19 mmHg and ±1.97 mmHg, respectively. Univariate regression analysis showed that CCT, CH and CRF significantly (P < 0.0001) correlated with the IOP measured with the Goldmann and noncontact tonometers (and with the differences between tonometers) in Session 1. There were no significant correlations with the differences between tonometers in Session 2. Multivariate analysis revealed a minimal effect of CCT on Goldmann measurements but a significant effect on those of the noncontact tonometer. Conclusions: Measurement of the biomechanical properties of the cornea, using the ORA, are repeatable and reproducible, affect Goldmann tonometry less than noncontact tonometry, and have a minimal influence on the difference in measured intraocular pressure between tonometers. © 2013 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
1. Introduction Accurate measurement of intraocular pressure (IOP) is vital for the diagnosis and management of glaucoma because, as has been shown by all the major glaucoma trials, it is the only variable which can be altered to prevent or delay the onset and/or progression of glaucoma [1–7]. The Goldmann applanation tonometer (GAT) is the clinical gold standard for IOP assessment, but as with most tonometers, its measurements are influenced by the biomechanical properties of the cornea [8–12]. In contrast the true IOP, as measured intracamerally, is not subject to the biomechanical properties of the cornea [13,14].
∗ Corresponding author. Tel.: +966 1 4693530; fax: +966 1 4693536. E-mail addresses: [email protected], [email protected] (K.C. Ogbuehi), [email protected], [email protected] (U.L. Osuagwu). 1 Fax: +966 1 4693536.
However, the exact relationship between the corneal biomechanical properties and the IOP measured through the cornea is still unclear. The central corneal thickness (CCT) is supposed to influence the IOP measured through the cornea with an overestimation of IOP in thicker corneas and an underestimation in thinner corneas. A number of formulae [8,11,15,16] have been computed to correct trans-corneally measured IOP for the effect of CCT but their usage has failed to gain wide acceptance to date. Some of the studies have even come to contradictory conclusions. Foster et al. [16] found no relationship between corneal thickness and the difference in the measured IOP (between applanation tonometry and intracameral cannulation), leading them to provide a correction formula for IOP measured by applanation tonometry which did not take corneal thickness into consideration. This result was in contrast to studies by Ehlers et al. [8] and Whitacre et al. [15]. Both latter studies reported significant relationships between the corneal thickness and the measurement error between the IOP measured by applanation tonometry and the true IOP (as measured by intracameral cannulation). Data from the Ehlers et al. [8] and Whitacre et al. [15] studies suggest that a 10% increase in corneal thickness would
1367-0484/$ – see front matter © 2013 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clae.2013.09.006
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8 2
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
result in a 3.5 mmHg [8] and 1.1 mmHg [15] increase in the IOP measured by applanation tonometry. The recent introduction of the Ocular Response Analyzer® (ORA) – Reichert Ophthalmic Instruments, Depew, NY, USA – means that corneal biomechanical properties other than CCT may be measured, indirectly, in vivo. The ORA fires a metered air pulse at the cornea, causing it indent and thus pass through two (inward and outward) applanation events. From the two IOP readings, the ORA computes two corneal biomechanical properties. Corneal hysteresis (CH) is supposed to represent the viscoelastic properties of the cornea. The corneal resistance factor (CRF) is thought to predominantly reflect the elastic resistance of the cornea, and could reflect the overall resistance of the eye [17,18]. Previous studies have demonstrated statistically significant reductions in CH and CRF in post-lasik eyes [19,20], pseudophakic eyes [21], keratoconus [18,19,22], in eyes with glaucoma [23–27], and in Fuchs corneal dystrophy [17]. It seems that CH and CRF are significantly lower in normal relatives of keratoconus patients [28] and in patients who showed a more rapid progression of glaucomatous visual field loss [23,29]. However, one recent study [30] reported only a weak relationship between CH and CRF on the one hand, and structural (retinal nerve fibre layer thickness) and functional (visual field) oculovisual damage in glaucoma, on the other. Since both parameters decrease along with CCT in post-LASIK corneas, and have also been shown to be higher in thicker corneas [31], it would appear that CH and CRF are functional surrogates for CCT, which itself has been shown to be a risk factor for the onset and/or progression of primary open-angle glaucoma (POAG) and normal tension glaucoma (NTG) [5,7,32–34]. Corneal hysteresis and CRF tend to be positively correlated with CCT in a number of anterior segment diseases notably keratoconus and glaucoma. In some of these cases, CH and CRF appear to be risk factors for the development and/or progression of the disease process independent of CCT [23,35,36]. The increasing relevance of the corneal biomechanical properties makes it important to thoroughly assess repeatability, as well as long- and short-term reproducibility, for each of these biomechanical properties. Also, since most forms of glaucoma are chronic disease processes, multiple, longitudinal comparisons of IOP measurements are a necessary and crucial element of patient care [37]. A number of studies have reported on the repeatability and reproducibility of CH and CRF [22,25,38] and on the influence of these parameters on the IOP measured by indentation and applanation methods [14,24,36,39,40] but none has compared the consistency of the effect (on the measured IOP) of CH and CRF (measured on separate days), on the one hand, with that of CCT on the other. Also there are no reports in the literature of the comparative influence of these properties on the IOP measured with the GAT and that measured with a non-ORA noncontact tonometer (NCT). Only one study to date has reported on the influence of CH and CRF on the difference in measured IOP between an NCT and the GAT [41]. The goals of this study were to assess the repeatability and reproducibility of the biomechanical properties (CH and CRF) measured by the ORA, determine their effects on the IOP measured with the GAT and a NCT in comparison to the effect of CCT on those same parameters, and to study the influence of CCT, CH and CRF on the difference between the IOPs measured by the GAT and a NCT. With specific regard to the effect on the measured IOP in two separate sessions, we sought to test the null hypotheses of:
(1) No significant difference between sessions for IOP readings with the GAT and noncontact tonometer, and for CH, CRF and CCT.
(2) No difference in the effect of CCT or CH or CRF, between sessions, on the GAT-measured IOP and on the noncontact IOP. (3) No difference in session 1 and in session 2 between the effects on measured IOP, of: CCT versus CH; CCT versus CRF; and CH versus CRF.
2. Methods The study cohort included one randomly selected eye of 106 routine Optometry patients, who met the criteria for participation. Randomization was carried out by a designated researcher, using a sequence of random numbers generated on a Microsoft Excel spreadsheet. The purpose of the investigation, and the rights of the participants (before, during and after the study) was explained to each subject before his/her participation, and each subject gave informed consent to participate in the study, in accordance with the 1975 Helsinki Declaration, as modified in Edinburgh 2000. The study protocol received local ethical committee approval. Prior to inclusion in the study, each subject underwent a comprehensive ophthalmic examination which included slit-lamp biomicroscopy of the external eye and anterior segment, monocular direct ophthalmoscopy, central visual field assessment with automated static perimetry, objective & subjective refraction and pupil evaluation. Subjects with a positive history for one of the following were excluded: participation in one of the previous studies by our research group; a positive history for (or objective evidence of) anterior segment disease or surgery; a history of contact lens wear; a history of ocular hypertension or glaucoma. Therefore all subjects were oculovisual normals in good general health. The advantage of selecting a young sample of oculovisual normals is that the parameters measured in this study would be expected to vary within narrow limits compared with those in patients with anterior segment disease, for example. Therefore, any significant variation from one week to the next would suggest a poor reproducibility of the technique in question, within this time period. Of the one hundred and ten subjects recruited for this study, four (three men) were lost to follow up. Each subject was required to visit the clinic for two separate measurement sessions separated by approximately one week. In both sessions, IOP measurements were made between 14:00 h and 16:00 h to ensure that the IOP was assessed at the period of the day when it is known to be most stable [42]. The measurements made in the second session were to confirm the results of the first session, to establish reproducibility indices for the biomechanical properties measured with the ORA, and to judge the consistency of the effects of CH and CRF (on the IOP measured with the GAT and a NCT) compared with that of CCT. In each session, the order of IOP measurements with the Ocular Response Analyzer (Reichert Ophthalmic Instruments, Depew, New York, USA) and Topcon CT80 noncontact tonometer (Topcon Medical Systems Inc., Oakland, New Jersey) was randomized by a designated researcher, using a sequence of random numbers generated on a Microsoft Excel spreadsheet. Following the measurements with the noncontact tonometers, the cornea was anaesthetized with one drop of 0.4% oxybuprocaine for assessment of the corneal thickness with the ultrasound pachymeter attached to the ORA followed by Goldmann applanation tonometry. As there are no studies in the published literature showing that ultrasound pachymetry causes an ocular massage effect, we deemed that the potential effect of performing pachymetry before Goldmann tonometry would be less than that if the order of measurements were reversed (i.e. if Goldmann tonometry was performed before pachymetry, we believed that the effect on the corneal thickness measured would have been greater). Our reasoning is supported by the conclusion, by AlMubrad and Ogbuehi
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
[43], that even the (approximately 1 mmHg) ocular massage effect attributed to the Goldmann tonometer in their study was almost identical to the effect caused by topical anaesthesia alone in an earlier study [44]. This suggested to us that the applanation of the Goldmann tonometer by itself is not enough to cause a significant ocular massage effect without topical anaesthesia. Therefore we reasoned that the much lighter touch of the much smaller ultrasound probe would cause an even smaller effect that would be both statistically and clinically insignificant. To preclude the possible effect of an ocular massage, Goldmann tonometry was performed after IOP assessment with the noncontact tonometers [45–48]. Contact of the applanation probe with the cornea was kept to between 4 and 5 s to simultaneously minimize the effects of aqueous massage on repeated applanation readings [45,46] and the possible errors caused by oscillatory fluctuations in IOP due mainly to the ocular pulse and respiratory cycles [49–51]. To further minimize the aforementioned errors, triplicate readings were made (with the Goldmann tonometer) and averaged to get the IOP reading for an eye. Quadruplicate measurements of IOP were taken with the Topcon CT80 non-contact tonometer but only the last three of measurements were used to compute the average IOP for statistical analyses. This procedure was adopted, as recommended by the manufacturer, to suit the principle of IOP measurement used by the Topcon CT80 non-contact tonometer in which, after the first pulse is fired at the cornea, subsequent pulses are automatically adjusted to the IOP of the subject to minimize the risk of excessive air pressure. Four measurements were also taken with the ORA in each session. The four measurements made with the ORA (software version 1) had to be of good quality as described by the inventor – ‘a good bell-shaped pressure signal with tall, relatively equal applanation peaks and relatively smooth raw and filtered applanation signals’ [17]. For some subjects, up to seven measurements had to be taken to obtain four good quality readings. Repeatability was computed using all ORA biomechanical measurements, which were also used to calculate the average corneal hysteresis and corneal resistance factor per patient per session. The averages so determined were used for further statistical analyses. The GAT readings were obtained in a masked fashion so that the clinician who performed GAT did not know the IOP of a subject before performing GAT on him/her and the IOP data records were maintained by another clinician. Each time a Goldmann IOP measurement was made, it was read off the scale by the randomizing researcher who, subsequently, reset the scale to an arbitrary value between 7 and 10 mmHg. This procedure was adopted to avoid tactile clues which might aid the tonometrist to determine the endpoint of a reading [44]. In each session, prior to IOP measurement with the GAT, CCT was assessed with the ultrasound pachymeter attached to the ORA. 2.1. Statistical analyses Statistical analyses were conducted with the Graphpad Instat for Windows programme, version 3.00 (Graphpad Software Inc., San Diego, California, USA, www.graphpad.com), and with Microsoft Excel 2007. For sample size determination, the statistical freeware G*Power (version 3.0.10) was downloaded from the web and employed as described elsewhere [52]. A sample size power calculation was based on a pilot sample of 20 subjects with a mean (±SD) CH of 11.9 mmHg (±1.9) and CRF of 11.5 mmHg (±2.0). The sample size necessary to achieve a statistical power of 95% at an ˛ level of 0.05 was 96 subjects. 2.2. Assessment of repeatability and reproducibility To assess repeatability, the differences (between triplicate readings, or quadruplicate readings in the case of the ORA) for each
3
subject were averaged. These averaged differences and the associated standard deviations were used to plot the 95% limits of repeatability for CCT, CH and CRF in each session. To assess the consistency of the within-session differences for each corneal biomechanical property, paired t-test comparisons were performed for the column of within-session differences derived from session one with that derived from session two, for CCT, CH and for CRF. Repeatability coefficients were computed by taking the square root of the average within-subject variance, to get the average within-subject standard deviation (Sw). The repeatability coefficient was calculated as 2.77Sw [53] for CH and CRF in each session, and one reproducibility coefficient each was plotted for CH and for CRF. To assess reproducibility for CCT, CH and CRF (for the same observer on a different day), a reproducibility coefficient was calculated using the mean measurement in session 1 and that in session 2 to compute a within-subject standard deviation, between sessions. 2.3. Influence of corneal parameters on clinical IOP measurements Regression analysis was used to examine the one-to-one relationship between the independent variables (CCT, CH and CRF) measured with the ORA, and the dependent variables (IOP measured with the Goldmann applanation tonometer – GATIOP –, the IOP measured with the CT80 noncontact tonometer – NCTIOP – and the difference between them – NCTIOP − GATIOP ). Regression Analysis was used to evaluate the associations between combined independent variables GATIOP , NCTIOP , and GATIOP − NCTIOP . The aim was to determine if a combination of biomechanical properties was better than any one biomechanical property in explaining the variation in the IOP measured with the Goldmann tonometer, with the noncontact tonometer, as well as the variation in the difference between contact and noncontact tonometers. In combining the corneal biomechanical properties, there were problems with multicollinearity caused by the high correlation (R = 0.9) between CH and CRF, but the variance inflation factors (VIFs) for CH and CRF (5.29 and 6.62, respectively) were not large enough to necessitate an attempt to reduce multicollinearity, by, for example, eliminating either CH or CRF. The presence of multicollinearity means that the results relating to the individual coefficients of determination (for CH and CRF) should be interpreted with caution. This is especially true if the null hypothesis is not rejected. The reason is that moderately large VIFs (as are present with multicollinearity) would make it more likely that the null hypothesis will not be rejected. In addition, O’Brien [54] cautioned against the inappropriate application of rules of thumb (the rule of 4; the rule of 10, etc.) concerning VIFs, stating that several factors affect the variance of a regression coefficient, and that sometimes, these factors can reduce the variance of regression coefficients far more than they are inflated by the VIF. The correlations of CH and CRF with CCT (as the independent variable) were also assessed using Pearson’s product–moment correlation coefficient. The level of statistical significance for this study was set at 0.05. 3. Results Table 1 presents the demographic data for the subject cohort, and descriptive statistics for measurements of intraocular pressure and the corneal biomechanical properties. Mean values for IOP with both tonometers as well as the mean values for CCT, CH and CRF did not vary significantly (P > 0.05) between sessions.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
4 Table 1 Biometric and demographic data of the subjects. Numbers and parameters (106 subjects) Female gender (no.) Right eyes (no.) Age (yrs)
Mean
Standard deviation
Range
49 61 22.5
2.1
Session 1 GATIOP (mmHg) CT80IOP (mmHg) CRF (mmHg) CH (mmHg) CCT (m)
14.2 14.5 10.6 11.0 560
3.0 3.3 2.2 1.8 37
7.7–19.7 7.3–22.7 6.7–17.1 7.1–16.2 480–665
Session 2 GATIOP (mm/Hg) CT80IOP (mm/Hg) CRF (mm/Hg) CH (mm/Hg) CCT (m)
14.1 14.4 10.5 11.1 559
2.8 3.0 2.1 1.9 36
8.3–22.7 7.3–24.7 6.7–16.1 7.6–16.6 483–678
19–30
GATIOP , intraocular pressure measured with the Goldmann applanation tonometer; CT80IOP , intraocular pressure measured with the Topcon CT80 noncontact tonometer; CRF, corneal resistance factor; CH, corneal hysteresis; CCT, central corneal thickness.
Repeatability indices did not differ significantly (P > 0.05) between sessions for the corneal biomechanical properties and the repeatability coefficients for CCT, CH and CRF (Fig. 1) in Session 1 were ±0.01 m, ±3.05 mmHg and ±2.62 mmHg, respectively. The corresponding repeatability coefficients for Session 2 were: ±0.02 m; ±2.74 mmHg; and ±2.52 mmHg, respectively. In comparison to the repeatability coefficients for CH and CRF, those for the GAT (measured in the same units) were 2.28 mmHg and 2.34 mmHg for Sessions 1 and 2 respectively. Neither the averages nor the columns of differences (computed by averaging the differences between the triplicate or quadruplicate readings for each subject) for CCT, CH or CRF varied significantly (P > 0.05) between sessions. The reproducibility coefficients for CCT, CH and CRF were ±0.02 m, ±2.19 mmHg and ±1.97 mmHg, respectively. In comparison to CH and CRF, the reproducibility coefficient for the GAT was ±4.17 mmHg. Univariate analysis between the independent variables (CCT, CH, and CRF) and GATIOP in Session 1, returned coefficients of determination of: R2 = 0.178 (F1,104 = 22.46, P < 0.0001), R2 = 0.104 (F1,104 = 12.09, P < 0.0001), and R2 = 0.335 (F1,104 = 52.40, P < 0.0001), for CCT, CH and CRF (Fig. 2) respectively. The corresponding coefficients of determination for Session 2, were: R2 = 0.309 (F1,104 = 46.50, P < 0.0001), R2 = 0.149 (F1,104 = 18.22, P < 0.0001), and R2 = 0.384 (F1,104 = 64.85, P < 0.0001). The regression of CT80IOp on CCT, CH and CRF (Fig. 3), respectively, returned values of: R2 = 0.464 (F1,104 = 89.96, P < 0.0001), R2 = 0.319 (F1,104 = 48.70, P < 0.0001), and R2 = 0.595 (F1,104 = 153.04, P < 0.0001) in Session 1. For Session 2, the corresponding coefficients of determination were: R2 = 0.405 (F1,104 = 70.86, P < 0.0001); R2 = 0.222 (F1,104 = 29.64, P < 0.0001); and R2 = 0.473 (F1,104 = 93.26, P < 0.0001).
Therefore CRF best explained the variation of both GATIOP and CT80IOP . Of the three biomechanical properties, it also had the most consistent effect on GATIOP between sessions. The coefficient of determination for the regression of GATIOP on CCT showed the most (two-fold) variation between sessions suggesting that the effect of CCT on GATIOP (but not CT80IOP ) is very variable. All three biomechanical properties had a greater effect on CT80IOP than on GATIOP in both sessions. Regression of ‘CT80IOp – GATIOP ’ on CCT, CH and CRF (Fig. 4), respectively, returned values of: R2 = 0.207 (F1,104 = 27.15, P < 0.0001), R2 = 0.171 (F1,104 = 21.50, P < 0.0001), and R2 = 0.142 (F1,104 = 17.21, P < 0.0001) in Session 1. For Session 2, the corresponding coefficients of determination were: R2 = 0.021 (F1,104 = 2.19, P = 0.14); R2 = 0.019 (F1,104 = 1.98, P = 0.16); and R2 = 0.017 (F1,104 = 1.83, P = 0.18). The effect of the individual biomechanical properties on the difference between the Goldmann IOP and the CT80IOP can therefore be surmised as ‘small and inconsistent’. The adjusted multiple linear regression coefficients of determination, for Sessions 1 and 2, are detailed in Table 2. In the multiple linear regression models, the corneal biomechanical properties also showed a consistently greater effect on the noncontact tonometer than on the Goldmann tonometer. When the biomechanical properties were paired, CCT and CH together had the least predictive ability of the variation in GATIOP readings and CT80IOP readings. The combination of CH and CRF best explained the variations in both the Goldman-measured IOP and CT80-measured IOP in both sessions, and, in the first session, the combination of CH and CRF better explained the variation of GATIOP than the combination of CH, CCT and CRF. Generally, once CH and CRF were taken
Fig. 1. Repeatability of corneal biomechanical properties – depicted as the mean of the differences between triplicate readings, and the 95% limits of repeatability (±1.96 SD) – in Session 1 and Session 2: (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF).
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
5
Fig. 2. Regression of the Goldmann-measured IOP on the corneal biomechanical properties in Session 1 and Session 2 (R2 , multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF).
Fig. 3. Regression of the CT80-measured IOP on the corneal biomechanical properties in Session 1 and Session 2 (R2 , multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF).
Fig. 4. Regression of the difference between the CT80-measured IOP and the Goldmann-measured IOP (CT80IOP − GATIOP ) on the corneal biomechanical properties in Session 1 and Session 2. (R2 , Multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF). Table 2 Adjusted multiple regression Coefficients of determination. Dependent variables
Independent variables Session 1 CCT + CH CCT + CRF CH + CRF CCT + CH + CRF Session 2 CCT + CH CCT + CRF CH + CRF CCT + CH + CRF
GATIOP
CT80IOP
CT80IOP − GATIOP
Adjusted R2 = 0.165, F2 ,103 = 11.38, P < 0.0001 Adjusted R2 = 0.322, F2,103 = 25.97, P < 0.0001 Adjusted R2 = 0.540, F2,103 = 62.61, P < 0.0001 Adjusted R2 = 0.536, F3,102 = 41.46, P < 0.0001
Adjusted R2 = 0.477, F2,103 = 48.86, P < 0.0001 Adjusted R2 = 0.614, F2,103 = 84.44, P < 0.0001 Adjusted R2 = 0.682, F2,103 = 113.59, P < 0.0001 Adjusted R2 = 0.703, F3,102 = 83.66, P < 0.0001
Adjusted R2 = 0.215, F2,103 = 15.35, P < 0.0001 Adjusted R2 = 0.195, F2,103 = 13.72, P < 0.0001 Adjusted R2 = 0.155, F2,103 = 10.65, P < 0.0001 Adjusted R2 = 0.220, F3,102 = 10.88, P < 0.0001
Adjusted R2 = 0.301, F2,103 = 23.65, P < 0.0001 Adjusted R2 = 0.404, F2,103 = 36.61, P < 0.0001 Adjusted R2 = 0.522, F2,103 = 58.33, P < 0.0001 Adjusted R2 = 0.537, F3,102 = 41.52, P < 0.0001
Adjusted R2 = 0.409, F2,103 = 37.32, P < 0.0001 Adjusted R2 = 0.514, F2,103 = 56.47, P < 0.0001 Adjusted R2 = 0.573, F2,103 = 71.47, P < 0.0001 Adjusted R2 = 0.606, F3,102 = 54.81, P < 0.0001
Adjusted R2 = 0.006, F2,103 = 1.32, P = 0.27 Adjusted R2 = 0.004, F2,103 = 1.19, P = 0.31 Adjusted R2 = 0.000, F2,103 = 1.00, P = 0.37 Adjusted R2 = −0.003, F3,102 = 0.88, P = 0.45
GATIOP , intraocular pressure measured with the Goldmann applanation tonometer; CT80IOP , intraocular pressure measured with the Topcon CT80 Noncontact Tonometer; CT80IOP − GATIOP , difference between the intraocular pressure measured with the Topcon CT80 noncontact tonometer and that measured with the Goldmann applanation tonometer; CRF, corneal resistance factor; CH, corneal hysteresis; CCT, central corneal thickness; R2 , multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
6
Fig. 5. Regression of the ORA-measured biomechanical properties of the cornea on central corneal thickness in Session 1 and Session 2 (R2 , Multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) corneal hysteresis (CH) and (B) corneal resistance factor (CRF).
into account, the addition of CCT to the multiple regression model brought slight but statistically significant increases in the predictive ability of the model, except with the GATIOP for Session 1 in which the contribution of CCT to the model was not statistically significant. In Session 2, also with the GATIOP , the contribution of CCT just reached the level of significance (t = 2.06, P = 0.04) set for this study. The multiple linear regressions of ‘CT80IOp − GATIOP ’ on CCT, CH and CRF followed the same trend as the simple linear regressions, showing statistically significant coefficients of determination in the first session but not in the second session. The multiple linear regressions also showed that CCT alone best explained the variation in the differences between tonometers and that once CCT had been accounted for, adding CH and CRF to the model only minimally influenced the predictive ability of the model (the contribution of CRF was not statistically significant – t = −1.31, P = 0.19 – and the contribution of CH just reached statistical significance – t = 2.08, P = 0.04), with the combination of CCT and CRF even having a reduced predictive ability than with CCT alone. All coefficients of determination in the second session were not statistically significant and thus these effects can also be surmised as: ‘small and inconsistent’. Corneal hysteresis and corneal resistance factor both demonstrated consistent, high and statistically significant correlations with CCT, with CRF demonstrating a higher correlation with CCT than CH (Fig. 5). 4. Discussion The ORA is a relatively new noncontact tonometer designed to measure functional biomechanical properties of the cornea in vivo and incorporate these properties to adjust, and possibly improve the accuracy of, the IOP measured by tonometry. The ORA offers the clinician a means to assess biomechanical properties of patients other than CCT. It is therefore imperative to ensure adequate repeatability and reproducibility of these new ORA biomechanical properties as they become more widely incorporated into routine clinical practice. Though earlier studies have assessed repeatability of CH and CRF, none has assessed reproducibility in the manner in which it was assessed in this study. Further, no study has compared the effect on the measured IOP of CCT on the one hand, and CH and CRF on the other, on the same group of subjects and on two separate occasions. The results from this study showed adequate repeatability indices for CCT, CH and CRF. The repeatability indices for CH and
CRF were comparable to the index (measured in the same units) for the Goldmann tonometer. The average CCT did not vary significantly between sessions and the column of differences (generated from the triplicate readings for each subject) did not vary between sessions. Therefore the reproducibility for CCT between sessions was good. The same was true for CH and CRF between sessions. The reproducibility indices for CH and CRF were twice as good for the index with the Goldmann tonometer in this study. The averages, repeatability and reproducibility indices for the corneal biomechanical properties did not vary significantly between men and women. Univariate analysis showed that CRF explained the variation in GATIOP and CT80IOP better than either CCT or CH, but CCT explained the differences between tonometers better than CRF or CH. The effect of the biomechanical properties of the differences between tonometers was not repeatable between sessions. In the univariate analysis (with respect to GATIOP , CT80IOP , and CT80IOP − GATIOP ) CCT, CH and CRF (the independent variables) were all able to explain a significant amount of the variation in the dependent variables, except in session 2 in which none of the independent variables was able to explain a statistically significant portion of the variation of CT80IOP − GATIOP . These results are in agreement with results from studies that have demonstrated CRF to be the most important factor influencing GAT measurements [24,55], with studies showing a significant correlation between GATIOP and CCT [14,24,56], and with studies which reported that CCT is the primary biomechanical parameter influencing the measurement differences between two tonometers [24]. Other studies have reported results in conflict with the findings from this study. Notably, Yu et al. [14] reported that the difference in measured IOP between the tonopen and intracamerally measured IOP was only associated with CRF. Tranchina et al. [41] showed that while CCT, CH and CRF all significantly influenced the difference between a NCT and the GAT, CRF was the best predictor of this difference. In the first measurement session of this study, all three biomechanical parameters influenced the differences between the CT80IOp and the GATIOP , but CCT (not CRF) was the best predictor of these differences. Multivariate analyses of the data from the first session this study showed that CCT, CH and CRF together accounted for 22% (R2 = 0.22, F3,102 = 10.88, P < 0.0001) of the variance between tonometers, similar to the 25% in the Tranchina et al. [41] study. However, these results were not repeated in the second session suggesting that the small effects of the corneal biomechanical properties on the differences between the CT80IOP and the GATIOP are rather variable.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
The results from this study also showed that, in univariate analyses, the GATIOP significantly correlated with CCT in both sessions. In multivariate analyses, once CH and CRF had been taken into account, the addition of CCT to the multivariate model either, did not significantly increase the correlation with GATIOP (in Session 1), or the increased correlation with the GATIOP just reached statistical significance (t = 2.06, P = 0.04, in Session 2). These findings concur with those of other studies [14,53] and suggest that CH and CRF are not completely independent of CCT and may even be functional surrogates of CCT and the tissue material properties of the cornea. This conclusion is supported by the high correlation (in both sessions) of CH and CRF with CCT. The biomechanical properties of the cornea all had a markedly greater influence on the CT80IOP than on the GATIOP . This effect could be explained in part by the greater influence that CH would be expected to have on noncontact tonometers. The corneal viscoelasticity (which CH is supposed to represent) is a dynamic resistance component for which corneal deformation is proportional to the force applied and the rate at which it is applied [19]. The CRF on the other hand, is a resistance component of the cornea for which corneal deformation is proportional only to the applied force [19]. The operating principles of the NCT compared to that of the GAT would suggest that the rate at which the applanation force is applied is significant only for the NCT and thus corneal viscoelasticity would be expected to exert a greater influence on IOP measurements made with a NCT, as was shown in this study. Copt et al. [33] demonstrated the importance that corneal thickness-adjusted IOP measurements could have on the diagnosis and management of glaucoma. They showed that correcting IOP for corneal thickness would cause over 50% of ocular hypertensives to be reclassified as normal, and 30% of normal tension glaucoma patients to be reclassified as having primary open angle glaucoma. It is likely that glaucoma misclassification errors would increase if CH and CRF were also taken into account, giving the importance that several studies have demonstrated for both parameters influencing the measured IOP [24]. In conclusion, this study demonstrated good repeatabilities and test-retest reproducibilities for the ORA biomechanical properties. The Goldmann tonometer provides estimates of true IOP that are less influenced by CCT, CH and CRF than the estimates provided by the CT80 noncontact tonometer. The biomechanical properties of the cornea influenced the differences between the Goldmann tonometer and the CT80 noncontact tonometer but these effects were small and inconsistent. The CRF has the greatest influence on the variation of measured IOP and, when the ORA biomechanical properties are accounted for, CCT has a small to negligible effect on the IOP measured with the Goldmann tonometer. On the same group of oculovisually normal subjects in good health, despite CCT, CH, CRF, GATIOP , and CT80IOP all showing good between-sessions reproducibility, the effect of CCT on the GATmeasured IOP was not consistent between sessions, in contrast those of CRF and CH. This suggests that the ORA biomechanical properties are better, more consistent predictors (than CCT) of the variation in trans-corneally measured IOP. A limitation of this study is the non-inclusion of a wider variety of subjects, including those with anterior segment disease, thereby restricting the findings of this study to young oculovisual normals. As such, prospective, longitudinal studies in a wider variety of subjects are required to ascertain the association between the ORA biomechanical properties and IOP measured by tonometry. Conflict of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
7
Acknowledgements The authors extend their appreciation to the Research Centre, College of Applied Medical Sciences and to the Deanship of Scientific Research at King Saud University for funding this research.
References [1] Collaborative Normal-Tension Glaucoma Study Group. Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures. Am J Ophthalmol 1998;126(4):487–97. [2] Collaborative Normal-Tension Glaucoma Study Group. The effectiveness of intraocular pressure reduction in the treatment of normal-tension glaucoma. Am J Ophthalmol 1998;126(4):498–505. [3] The 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(4):429–40. [4] Kass MA, Heuer DK, Higginbotham EJ, Johnson CA, Keltner JL, Miller JP, 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(6):829–30, 701-713; discussion. [5] Gordon MO, Beiser JA, Brandt JD, Heuer DK, Higginbotham EJ, Johnson CA, et al. The ocular hypertension treatment study: baseline factors that predict the onset of primary open-angle glaucoma. Arch Ophthalmol 2002;120(6):829–30, 714-720; discussion. [6] 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):205–9. [7] 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):1965–72. [8] Ehlers N, Bramsen T, Sperling S. Applanation tonometry and central corneal thickness. Acta Ophthalmol (Copenh) 1975;53(1):34–43. [9] Whitacre MM, Stein R. Sources of error with use of Goldmann-type tonometers. Surv Ophthalmol 1993;38(1):1–30. [10] Dielemans I, Vingerling JR, Hofman A, Grobbee DE, de Jong PT. Reliability of intraocular pressure measurement with the Goldmann applanation tonometer in epidemiological studies. Graefes Arch Clin Exp Ophthalmol 1994;232(3):141–4. [11] Doughty MJ, Zaman ML. Human corneal thickness and its impact on intraocular pressure measures: a review and meta-analysis approach. Surv Ophthalmol 2000;44(5):367–408. [12] Liu J, Roberts CJ. Influence of corneal biomechanical properties on intraocular pressure measurement: quantitative analysis. J Cataract Refract Surg 2005;31(1):146–55. [13] Feltgen N, Leifert D, Funk J. Correlation between central corneal thickness, applanation tonometry, and direct intracameral IOP readings. Br J Ophthalmol 2001;85(1):85–7. [14] Yu AY, Duan SF, Zhao YE, Li XY, Lu F, Wang J, et al. Correlation between corneal biomechanical properties, applanation tonometry and direct intracameral tonometry. Br J Ophthalmol 2012;96(5):640–4. [15] Whitacre MM, Stein RA, Hassanein K. The effect of corneal thickness on applanation tonometry. Am J Ophthalmol 1993;115(5):592–6. [16] Foster PJ, Wong JS, Wong E, Chen FG, Machin D, Chew PT. Accuracy of clinical estimates of intraocular pressure in Chinese eyes. Ophthalmology 2000;107(10):1816–21. [17] Luce DA. Determining in vivo biomechanical properties of the cornea with an ocular response analyzer. J Cataract Refract Surg 2005;31(1):156–62. [18] Shah S, Laiquzzaman M, Bhojwani R, Mantry S, Cunliffe I. Assessment of the biomechanical properties of the cornea with the ocular response analyzer in normal and keratoconic eyes. Invest Ophthalmol Vis Sci 2007;48(7):3026–31. ˜ D, Shabayek MH, Arnalich-Montiel F, Alió JL. Corneal biome[19] Ortiz D, Pinero chanical properties in normal, post-laser in situ keratomileusis, and keratoconic eyes. J Cataract Refract Surg 2007;33(8):1371–5. [20] Pepose JS, Feigenbaum SK, Qazi MA, Sanderson JP, Roberts CJ. Changes in corneal biomechanics and intraocular pressure following LASIK using static, dynamic, and noncontact tonometry. Am J Ophthalmol 2007;143:39–47. [21] Hager A, Loge K, Füllhas MO, Schroeder B, Grossherr M, Wiegand W. Changes in corneal hysteresis after clear corneal cataract surgery. Am J Ophthalmol 2007;144(3):341–6. [22] González-Méijome JM, Queirós A, Jorge J, Díaz-Rey A, Parafita MA. Intraoffice variability of corneal biomechanical parameters and intraocular pressure (IOP). Optom Vis Sci 2008;85(6):457–62. [23] Congdon NG, Broman AT, Bandeen-Roche K, Grover D, Quigley HA. Central corneal thickness and corneal hysteresis associated with glaucoma damage. Am J Ophthalmol 2006;141(5):868–75. [24] Medeiros FA, Weinreb RN. Evaluation of the influence of corneal biomechanical properties on intraocular pressure measurements using the ocular response analyzer. J Glaucoma 2006;15(5):364–70. ˜ J, Maldonado MJ, García N, Mendiluce L, García-Gómez PJ, [25] Moreno-Montanés Seguí-Gómez M. Reproducibility and clinical relevance of the ocular response
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
8
[26]
[27] [28]
[29]
[30]
[31]
[32] [33]
[34]
[35]
[36]
[37]
[38]
[39]
analyzer in nonoperated eyes: corneal biomechanical and tonometric implications. Invest Ophthalmol Vis Sci 2008;49(3):968–74. Sullivan-Mee M, Billingsley SC, Patel AD, Halverson KD, Alldredge BR, Qualls C. Ocular response analyzer in subjects with and without glaucoma. Optom Vis Sci 2008;85(6):463–70. Morita T, Shoji N, Kamiya K, Fujimura F, Shimizu K. Corneal biomechanical properties in normal-tension glaucoma. Acta Ophthalmol 2012;90(1):e48–53. Kara N, Altinkaynak H, Baz O, Goker Y. Biomechanical evaluation of cornea in topographically normal relatives of patients with keratoconus. Cornea 2012, http://dx.doi.org/10.1097/ICO.0b013e3182490924 (cited 7.9.12). De Moraes CV, Hill V, Tello C, Liebmann JM, Ritch R. Lower corneal hysteresis is associated with more rapid glaucomatous visual field progression. J Glaucoma 2012;21(4):209–13. Mansouri K, Leite MT, Weinreb RN, Tafreshi A, Zangwill LM, Medeiros FA. Association between corneal biomechanical properties and glaucoma severity. Am J Ophthalmol 2012;153(3):419–27. Franco S, Lira M. Biomechanical properties of the cornea measured by the ocular response analyzer and their association with intraocular pressure and the central corneal curvature. Clin Exp Optom 2009;92(6):469–75. Morad Y, Sharon E, Hefetz L, Nemet P. Corneal thickness and curvature in normal-tension glaucoma. Am J Ophthalmol 1998;125(2):164–8. Copt RP, Thomas R, Mermoud A. Corneal thickness in ocular hypertension, primary open-angle glaucoma, and normal tension glaucoma. Arch Ophthalmol 1999;117(1):14–6. Choi HJ, Kim DM, Hwang SS. Relationship between central corneal thickness and localized retinal nerve fiber layer defect in normal-tension glaucoma. J Glaucoma 2006;15(2):120–3. 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):3262–70. Kirwan C, O’Keefe M, Lanigan B. Corneal hysteresis and intraocular pressure measurement in children using the Reichert ocular response analyzer. Am J Ophthalmol 2006;142(6):990–2. Sullivan-Mee M, Gerhardt G, Halverson KD, Qualls C. Repeatability and reproducibility for intraocular pressure measurement by dynamic contour, ocular response analyzer, and Goldmann applanation tonometry. J Glaucoma 2009;18(9):666–73. Wasielica-Poslednik J, Berisha F, Aliyeva S, Pfeiffer N, Hoffmann EM. Reproducibility of ocular response analyzer measurements and their correlation with central corneal thickness. Graefes Arch Clin Exp Ophthalmol 2010;248(11):1617–20. Detry-Morel M, Jamart J, Pourjavan S. Evaluation of corneal biomechanical properties with the Reichert ocular response analyzer. Eur J Ophthalmol 2011;21(2):138–48.
[40] Mangouritsas G, Mourtzoukos S, Mantzounis A, Alexopoulos L. Comparison of Goldmann and Pascal tonometry in relation to corneal hysteresis and central corneal thickness in nonglaucomatous eye. Clin Ophthalmol 2011;5: 1071–80. [41] Tranchina L, Lombardo M, Oddone F, Serrao S, Lomoriello DS, Ducoli P. Influence of corneal biomechanical properties on intraocular pressure differences between an air-puff tonometer and the Goldmann applanation tonometer. J Glaucoma 2012, http://dx.doi.org/10.1097/IJG.0b013e31824cafc9 (cited 7.9.12). [42] Liu JH, Kripke DF, Twa MD, Hoffman RE, Mansberger SL, Rex KM, et al. Twentyfour-hour pattern of intraocular pressure in the aging population. Invest Ophthalmol Vis Sci 1999;40(12):2912–20. [43] AlMubrad TM, Ogbuehi KC. On repeated corneal applanation with the Goldmann and two non-contact tonometers. Clin Exp Optom 2010;93(2):77–82. [44] AlMubrad TM, Ogbuehi KC. The effect of repeated applanation on subsequent IOP measurements. Clin Exp Optom 2008;91(6):524–9. [45] Stocker FW. On changes in intraocular pressure after application of the tonometer; in the same eye and in the other eye. Am J Ophthalmol 1958;45(2):192–6. [46] Moses RA, Liu CH. Repeated applanation tonometry. Am J Ophthalmol 1968;66(1):89–91. [47] Sorensen PN. The non-contact tonometer. Clinical evaluation on normal and diseased eyes. Acta Ophthalmol (Copenh) 1975;53(4):513–21. [48] Chauhan BC, Henson DB. Clinical evaluation of the non-contact tonometer Mark II. Am J Optom Physiol Opt 1988;65:751–6. [49] Bynke HG, Schele B. On the origin of the ocular pressure pulse. Ophthalmologica 1967;153:29–36. [50] Leydhecker W. The intraocular pressure clinical aspects. Ann Ophthalmol 1976;8:389–401. [51] Perkins ES. The ocular pulse. Curr Eye Res 1981;1:19–23. [52] Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioural, and biomedical sciences. Behav Res Methods 2007;39:175–91. [53] Bland JM, Altman DG. Measurement error. BMJ 1996;313(7059):744. [54] O’Brien RM. A caution regarding rules of thumb for variance inflation factors. Qual Quant 2007;41:673–90. [55] Kaushik S, Pandav SS, Banger A, Aggarwal K, Gupta A. Relationship between corneal biomechanical properties, central corneal thickness, and intraocular pressure across the spectrum of glaucoma. Am J Ophthalmol 2012;153(5):840–9. [56] Tonnu PA, Ho T, Newson T, El Sheikh A, Sharma K, White E, et al. The influence of central corneal thickness and age on intraocular pressure measured by pneumotonometry, non-contact tonometry, the Tono-Pen XL, and Goldmann applanation tonometry. Br J Ophthalmol 2005;89(7):851–4.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
Contact Lens & Anterior Eye xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Contact Lens & Anterior Eye journal homepage: www.elsevier.com/locate/clae
Corneal biomechanical properties: Precision and influence on tonometry Kelechi C. Ogbuehi ∗ , Uchechukwu L. Osuagwu 1 Cornea Research Chair, Department of Optometry and Vision Sciences College of Applied Medical Sciences, King Saud University, P.O. Box 10219, Riyadh 11433, Saudi Arabia
a r t i c l e
i n f o
Article history: Received 31 October 2012 Received in revised form 19 August 2013 Accepted 16 September 2013 Keywords: Goldmann tonometer Ocular Response Analyzer ORA Noncontact tonometer Repeatability Tonometry.
a b s t r a c t Purpose: To assess the precision and reproducibility of the corneal biomechanical parameters, and their relationships with the intraocular pressure (IOP) measured with the Goldmann tonometer and a noncontact tonometer. Methods: Readings for biomechanical properties and for IOP measured with the Goldmann and noncontact tonometers, were taken on one randomly selected eye of 106 normal subjects, on each one of two measurement sessions. Measurements with the ocular response analyzer (ORA) and the noncontact tonometer were randomized, followed by the measurement of central corneal thickness and with the Goldmann tonometer. Results: Repeatability coefficients for CCT, corneal hysteresis (CH) and corneal resistance factor (CRF) in Session 1 were ±0.01 m, ±3.05 mmHg and ±2.62 mmHg, respectively. The mean CCT, CH, CRF, Goldmann and noncontact tonometry did not vary significantly between sessions. Reproducibility coefficients for CCT, CH and CRF were ±0.02 m, ±2.19 mmHg and ±1.97 mmHg, respectively. Univariate regression analysis showed that CCT, CH and CRF significantly (P < 0.0001) correlated with the IOP measured with the Goldmann and noncontact tonometers (and with the differences between tonometers) in Session 1. There were no significant correlations with the differences between tonometers in Session 2. Multivariate analysis revealed a minimal effect of CCT on Goldmann measurements but a significant effect on those of the noncontact tonometer. Conclusions: Measurement of the biomechanical properties of the cornea, using the ORA, are repeatable and reproducible, affect Goldmann tonometry less than noncontact tonometry, and have a minimal influence on the difference in measured intraocular pressure between tonometers. © 2013 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
1. Introduction Accurate measurement of intraocular pressure (IOP) is vital for the diagnosis and management of glaucoma because, as has been shown by all the major glaucoma trials, it is the only variable which can be altered to prevent or delay the onset and/or progression of glaucoma [1–7]. The Goldmann applanation tonometer (GAT) is the clinical gold standard for IOP assessment, but as with most tonometers, its measurements are influenced by the biomechanical properties of the cornea [8–12]. In contrast the true IOP, as measured intracamerally, is not subject to the biomechanical properties of the cornea [13,14].
∗ Corresponding author. Tel.: +966 1 4693530; fax: +966 1 4693536. E-mail addresses: [email protected], [email protected] (K.C. Ogbuehi), [email protected], [email protected] (U.L. Osuagwu). 1 Fax: +966 1 4693536.
However, the exact relationship between the corneal biomechanical properties and the IOP measured through the cornea is still unclear. The central corneal thickness (CCT) is supposed to influence the IOP measured through the cornea with an overestimation of IOP in thicker corneas and an underestimation in thinner corneas. A number of formulae [8,11,15,16] have been computed to correct trans-corneally measured IOP for the effect of CCT but their usage has failed to gain wide acceptance to date. Some of the studies have even come to contradictory conclusions. Foster et al. [16] found no relationship between corneal thickness and the difference in the measured IOP (between applanation tonometry and intracameral cannulation), leading them to provide a correction formula for IOP measured by applanation tonometry which did not take corneal thickness into consideration. This result was in contrast to studies by Ehlers et al. [8] and Whitacre et al. [15]. Both latter studies reported significant relationships between the corneal thickness and the measurement error between the IOP measured by applanation tonometry and the true IOP (as measured by intracameral cannulation). Data from the Ehlers et al. [8] and Whitacre et al. [15] studies suggest that a 10% increase in corneal thickness would
1367-0484/$ – see front matter © 2013 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clae.2013.09.006
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8 2
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
result in a 3.5 mmHg [8] and 1.1 mmHg [15] increase in the IOP measured by applanation tonometry. The recent introduction of the Ocular Response Analyzer® (ORA) – Reichert Ophthalmic Instruments, Depew, NY, USA – means that corneal biomechanical properties other than CCT may be measured, indirectly, in vivo. The ORA fires a metered air pulse at the cornea, causing it indent and thus pass through two (inward and outward) applanation events. From the two IOP readings, the ORA computes two corneal biomechanical properties. Corneal hysteresis (CH) is supposed to represent the viscoelastic properties of the cornea. The corneal resistance factor (CRF) is thought to predominantly reflect the elastic resistance of the cornea, and could reflect the overall resistance of the eye [17,18]. Previous studies have demonstrated statistically significant reductions in CH and CRF in post-lasik eyes [19,20], pseudophakic eyes [21], keratoconus [18,19,22], in eyes with glaucoma [23–27], and in Fuchs corneal dystrophy [17]. It seems that CH and CRF are significantly lower in normal relatives of keratoconus patients [28] and in patients who showed a more rapid progression of glaucomatous visual field loss [23,29]. However, one recent study [30] reported only a weak relationship between CH and CRF on the one hand, and structural (retinal nerve fibre layer thickness) and functional (visual field) oculovisual damage in glaucoma, on the other. Since both parameters decrease along with CCT in post-LASIK corneas, and have also been shown to be higher in thicker corneas [31], it would appear that CH and CRF are functional surrogates for CCT, which itself has been shown to be a risk factor for the onset and/or progression of primary open-angle glaucoma (POAG) and normal tension glaucoma (NTG) [5,7,32–34]. Corneal hysteresis and CRF tend to be positively correlated with CCT in a number of anterior segment diseases notably keratoconus and glaucoma. In some of these cases, CH and CRF appear to be risk factors for the development and/or progression of the disease process independent of CCT [23,35,36]. The increasing relevance of the corneal biomechanical properties makes it important to thoroughly assess repeatability, as well as long- and short-term reproducibility, for each of these biomechanical properties. Also, since most forms of glaucoma are chronic disease processes, multiple, longitudinal comparisons of IOP measurements are a necessary and crucial element of patient care [37]. A number of studies have reported on the repeatability and reproducibility of CH and CRF [22,25,38] and on the influence of these parameters on the IOP measured by indentation and applanation methods [14,24,36,39,40] but none has compared the consistency of the effect (on the measured IOP) of CH and CRF (measured on separate days), on the one hand, with that of CCT on the other. Also there are no reports in the literature of the comparative influence of these properties on the IOP measured with the GAT and that measured with a non-ORA noncontact tonometer (NCT). Only one study to date has reported on the influence of CH and CRF on the difference in measured IOP between an NCT and the GAT [41]. The goals of this study were to assess the repeatability and reproducibility of the biomechanical properties (CH and CRF) measured by the ORA, determine their effects on the IOP measured with the GAT and a NCT in comparison to the effect of CCT on those same parameters, and to study the influence of CCT, CH and CRF on the difference between the IOPs measured by the GAT and a NCT. With specific regard to the effect on the measured IOP in two separate sessions, we sought to test the null hypotheses of:
(1) No significant difference between sessions for IOP readings with the GAT and noncontact tonometer, and for CH, CRF and CCT.
(2) No difference in the effect of CCT or CH or CRF, between sessions, on the GAT-measured IOP and on the noncontact IOP. (3) No difference in session 1 and in session 2 between the effects on measured IOP, of: CCT versus CH; CCT versus CRF; and CH versus CRF.
2. Methods The study cohort included one randomly selected eye of 106 routine Optometry patients, who met the criteria for participation. Randomization was carried out by a designated researcher, using a sequence of random numbers generated on a Microsoft Excel spreadsheet. The purpose of the investigation, and the rights of the participants (before, during and after the study) was explained to each subject before his/her participation, and each subject gave informed consent to participate in the study, in accordance with the 1975 Helsinki Declaration, as modified in Edinburgh 2000. The study protocol received local ethical committee approval. Prior to inclusion in the study, each subject underwent a comprehensive ophthalmic examination which included slit-lamp biomicroscopy of the external eye and anterior segment, monocular direct ophthalmoscopy, central visual field assessment with automated static perimetry, objective & subjective refraction and pupil evaluation. Subjects with a positive history for one of the following were excluded: participation in one of the previous studies by our research group; a positive history for (or objective evidence of) anterior segment disease or surgery; a history of contact lens wear; a history of ocular hypertension or glaucoma. Therefore all subjects were oculovisual normals in good general health. The advantage of selecting a young sample of oculovisual normals is that the parameters measured in this study would be expected to vary within narrow limits compared with those in patients with anterior segment disease, for example. Therefore, any significant variation from one week to the next would suggest a poor reproducibility of the technique in question, within this time period. Of the one hundred and ten subjects recruited for this study, four (three men) were lost to follow up. Each subject was required to visit the clinic for two separate measurement sessions separated by approximately one week. In both sessions, IOP measurements were made between 14:00 h and 16:00 h to ensure that the IOP was assessed at the period of the day when it is known to be most stable [42]. The measurements made in the second session were to confirm the results of the first session, to establish reproducibility indices for the biomechanical properties measured with the ORA, and to judge the consistency of the effects of CH and CRF (on the IOP measured with the GAT and a NCT) compared with that of CCT. In each session, the order of IOP measurements with the Ocular Response Analyzer (Reichert Ophthalmic Instruments, Depew, New York, USA) and Topcon CT80 noncontact tonometer (Topcon Medical Systems Inc., Oakland, New Jersey) was randomized by a designated researcher, using a sequence of random numbers generated on a Microsoft Excel spreadsheet. Following the measurements with the noncontact tonometers, the cornea was anaesthetized with one drop of 0.4% oxybuprocaine for assessment of the corneal thickness with the ultrasound pachymeter attached to the ORA followed by Goldmann applanation tonometry. As there are no studies in the published literature showing that ultrasound pachymetry causes an ocular massage effect, we deemed that the potential effect of performing pachymetry before Goldmann tonometry would be less than that if the order of measurements were reversed (i.e. if Goldmann tonometry was performed before pachymetry, we believed that the effect on the corneal thickness measured would have been greater). Our reasoning is supported by the conclusion, by AlMubrad and Ogbuehi
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
[43], that even the (approximately 1 mmHg) ocular massage effect attributed to the Goldmann tonometer in their study was almost identical to the effect caused by topical anaesthesia alone in an earlier study [44]. This suggested to us that the applanation of the Goldmann tonometer by itself is not enough to cause a significant ocular massage effect without topical anaesthesia. Therefore we reasoned that the much lighter touch of the much smaller ultrasound probe would cause an even smaller effect that would be both statistically and clinically insignificant. To preclude the possible effect of an ocular massage, Goldmann tonometry was performed after IOP assessment with the noncontact tonometers [45–48]. Contact of the applanation probe with the cornea was kept to between 4 and 5 s to simultaneously minimize the effects of aqueous massage on repeated applanation readings [45,46] and the possible errors caused by oscillatory fluctuations in IOP due mainly to the ocular pulse and respiratory cycles [49–51]. To further minimize the aforementioned errors, triplicate readings were made (with the Goldmann tonometer) and averaged to get the IOP reading for an eye. Quadruplicate measurements of IOP were taken with the Topcon CT80 non-contact tonometer but only the last three of measurements were used to compute the average IOP for statistical analyses. This procedure was adopted, as recommended by the manufacturer, to suit the principle of IOP measurement used by the Topcon CT80 non-contact tonometer in which, after the first pulse is fired at the cornea, subsequent pulses are automatically adjusted to the IOP of the subject to minimize the risk of excessive air pressure. Four measurements were also taken with the ORA in each session. The four measurements made with the ORA (software version 1) had to be of good quality as described by the inventor – ‘a good bell-shaped pressure signal with tall, relatively equal applanation peaks and relatively smooth raw and filtered applanation signals’ [17]. For some subjects, up to seven measurements had to be taken to obtain four good quality readings. Repeatability was computed using all ORA biomechanical measurements, which were also used to calculate the average corneal hysteresis and corneal resistance factor per patient per session. The averages so determined were used for further statistical analyses. The GAT readings were obtained in a masked fashion so that the clinician who performed GAT did not know the IOP of a subject before performing GAT on him/her and the IOP data records were maintained by another clinician. Each time a Goldmann IOP measurement was made, it was read off the scale by the randomizing researcher who, subsequently, reset the scale to an arbitrary value between 7 and 10 mmHg. This procedure was adopted to avoid tactile clues which might aid the tonometrist to determine the endpoint of a reading [44]. In each session, prior to IOP measurement with the GAT, CCT was assessed with the ultrasound pachymeter attached to the ORA. 2.1. Statistical analyses Statistical analyses were conducted with the Graphpad Instat for Windows programme, version 3.00 (Graphpad Software Inc., San Diego, California, USA, www.graphpad.com), and with Microsoft Excel 2007. For sample size determination, the statistical freeware G*Power (version 3.0.10) was downloaded from the web and employed as described elsewhere [52]. A sample size power calculation was based on a pilot sample of 20 subjects with a mean (±SD) CH of 11.9 mmHg (±1.9) and CRF of 11.5 mmHg (±2.0). The sample size necessary to achieve a statistical power of 95% at an ˛ level of 0.05 was 96 subjects. 2.2. Assessment of repeatability and reproducibility To assess repeatability, the differences (between triplicate readings, or quadruplicate readings in the case of the ORA) for each
3
subject were averaged. These averaged differences and the associated standard deviations were used to plot the 95% limits of repeatability for CCT, CH and CRF in each session. To assess the consistency of the within-session differences for each corneal biomechanical property, paired t-test comparisons were performed for the column of within-session differences derived from session one with that derived from session two, for CCT, CH and for CRF. Repeatability coefficients were computed by taking the square root of the average within-subject variance, to get the average within-subject standard deviation (Sw). The repeatability coefficient was calculated as 2.77Sw [53] for CH and CRF in each session, and one reproducibility coefficient each was plotted for CH and for CRF. To assess reproducibility for CCT, CH and CRF (for the same observer on a different day), a reproducibility coefficient was calculated using the mean measurement in session 1 and that in session 2 to compute a within-subject standard deviation, between sessions. 2.3. Influence of corneal parameters on clinical IOP measurements Regression analysis was used to examine the one-to-one relationship between the independent variables (CCT, CH and CRF) measured with the ORA, and the dependent variables (IOP measured with the Goldmann applanation tonometer – GATIOP –, the IOP measured with the CT80 noncontact tonometer – NCTIOP – and the difference between them – NCTIOP − GATIOP ). Regression Analysis was used to evaluate the associations between combined independent variables GATIOP , NCTIOP , and GATIOP − NCTIOP . The aim was to determine if a combination of biomechanical properties was better than any one biomechanical property in explaining the variation in the IOP measured with the Goldmann tonometer, with the noncontact tonometer, as well as the variation in the difference between contact and noncontact tonometers. In combining the corneal biomechanical properties, there were problems with multicollinearity caused by the high correlation (R = 0.9) between CH and CRF, but the variance inflation factors (VIFs) for CH and CRF (5.29 and 6.62, respectively) were not large enough to necessitate an attempt to reduce multicollinearity, by, for example, eliminating either CH or CRF. The presence of multicollinearity means that the results relating to the individual coefficients of determination (for CH and CRF) should be interpreted with caution. This is especially true if the null hypothesis is not rejected. The reason is that moderately large VIFs (as are present with multicollinearity) would make it more likely that the null hypothesis will not be rejected. In addition, O’Brien [54] cautioned against the inappropriate application of rules of thumb (the rule of 4; the rule of 10, etc.) concerning VIFs, stating that several factors affect the variance of a regression coefficient, and that sometimes, these factors can reduce the variance of regression coefficients far more than they are inflated by the VIF. The correlations of CH and CRF with CCT (as the independent variable) were also assessed using Pearson’s product–moment correlation coefficient. The level of statistical significance for this study was set at 0.05. 3. Results Table 1 presents the demographic data for the subject cohort, and descriptive statistics for measurements of intraocular pressure and the corneal biomechanical properties. Mean values for IOP with both tonometers as well as the mean values for CCT, CH and CRF did not vary significantly (P > 0.05) between sessions.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
4 Table 1 Biometric and demographic data of the subjects. Numbers and parameters (106 subjects) Female gender (no.) Right eyes (no.) Age (yrs)
Mean
Standard deviation
Range
49 61 22.5
2.1
Session 1 GATIOP (mmHg) CT80IOP (mmHg) CRF (mmHg) CH (mmHg) CCT (m)
14.2 14.5 10.6 11.0 560
3.0 3.3 2.2 1.8 37
7.7–19.7 7.3–22.7 6.7–17.1 7.1–16.2 480–665
Session 2 GATIOP (mm/Hg) CT80IOP (mm/Hg) CRF (mm/Hg) CH (mm/Hg) CCT (m)
14.1 14.4 10.5 11.1 559
2.8 3.0 2.1 1.9 36
8.3–22.7 7.3–24.7 6.7–16.1 7.6–16.6 483–678
19–30
GATIOP , intraocular pressure measured with the Goldmann applanation tonometer; CT80IOP , intraocular pressure measured with the Topcon CT80 noncontact tonometer; CRF, corneal resistance factor; CH, corneal hysteresis; CCT, central corneal thickness.
Repeatability indices did not differ significantly (P > 0.05) between sessions for the corneal biomechanical properties and the repeatability coefficients for CCT, CH and CRF (Fig. 1) in Session 1 were ±0.01 m, ±3.05 mmHg and ±2.62 mmHg, respectively. The corresponding repeatability coefficients for Session 2 were: ±0.02 m; ±2.74 mmHg; and ±2.52 mmHg, respectively. In comparison to the repeatability coefficients for CH and CRF, those for the GAT (measured in the same units) were 2.28 mmHg and 2.34 mmHg for Sessions 1 and 2 respectively. Neither the averages nor the columns of differences (computed by averaging the differences between the triplicate or quadruplicate readings for each subject) for CCT, CH or CRF varied significantly (P > 0.05) between sessions. The reproducibility coefficients for CCT, CH and CRF were ±0.02 m, ±2.19 mmHg and ±1.97 mmHg, respectively. In comparison to CH and CRF, the reproducibility coefficient for the GAT was ±4.17 mmHg. Univariate analysis between the independent variables (CCT, CH, and CRF) and GATIOP in Session 1, returned coefficients of determination of: R2 = 0.178 (F1,104 = 22.46, P < 0.0001), R2 = 0.104 (F1,104 = 12.09, P < 0.0001), and R2 = 0.335 (F1,104 = 52.40, P < 0.0001), for CCT, CH and CRF (Fig. 2) respectively. The corresponding coefficients of determination for Session 2, were: R2 = 0.309 (F1,104 = 46.50, P < 0.0001), R2 = 0.149 (F1,104 = 18.22, P < 0.0001), and R2 = 0.384 (F1,104 = 64.85, P < 0.0001). The regression of CT80IOp on CCT, CH and CRF (Fig. 3), respectively, returned values of: R2 = 0.464 (F1,104 = 89.96, P < 0.0001), R2 = 0.319 (F1,104 = 48.70, P < 0.0001), and R2 = 0.595 (F1,104 = 153.04, P < 0.0001) in Session 1. For Session 2, the corresponding coefficients of determination were: R2 = 0.405 (F1,104 = 70.86, P < 0.0001); R2 = 0.222 (F1,104 = 29.64, P < 0.0001); and R2 = 0.473 (F1,104 = 93.26, P < 0.0001).
Therefore CRF best explained the variation of both GATIOP and CT80IOP . Of the three biomechanical properties, it also had the most consistent effect on GATIOP between sessions. The coefficient of determination for the regression of GATIOP on CCT showed the most (two-fold) variation between sessions suggesting that the effect of CCT on GATIOP (but not CT80IOP ) is very variable. All three biomechanical properties had a greater effect on CT80IOP than on GATIOP in both sessions. Regression of ‘CT80IOp – GATIOP ’ on CCT, CH and CRF (Fig. 4), respectively, returned values of: R2 = 0.207 (F1,104 = 27.15, P < 0.0001), R2 = 0.171 (F1,104 = 21.50, P < 0.0001), and R2 = 0.142 (F1,104 = 17.21, P < 0.0001) in Session 1. For Session 2, the corresponding coefficients of determination were: R2 = 0.021 (F1,104 = 2.19, P = 0.14); R2 = 0.019 (F1,104 = 1.98, P = 0.16); and R2 = 0.017 (F1,104 = 1.83, P = 0.18). The effect of the individual biomechanical properties on the difference between the Goldmann IOP and the CT80IOP can therefore be surmised as ‘small and inconsistent’. The adjusted multiple linear regression coefficients of determination, for Sessions 1 and 2, are detailed in Table 2. In the multiple linear regression models, the corneal biomechanical properties also showed a consistently greater effect on the noncontact tonometer than on the Goldmann tonometer. When the biomechanical properties were paired, CCT and CH together had the least predictive ability of the variation in GATIOP readings and CT80IOP readings. The combination of CH and CRF best explained the variations in both the Goldman-measured IOP and CT80-measured IOP in both sessions, and, in the first session, the combination of CH and CRF better explained the variation of GATIOP than the combination of CH, CCT and CRF. Generally, once CH and CRF were taken
Fig. 1. Repeatability of corneal biomechanical properties – depicted as the mean of the differences between triplicate readings, and the 95% limits of repeatability (±1.96 SD) – in Session 1 and Session 2: (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF).
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
5
Fig. 2. Regression of the Goldmann-measured IOP on the corneal biomechanical properties in Session 1 and Session 2 (R2 , multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF).
Fig. 3. Regression of the CT80-measured IOP on the corneal biomechanical properties in Session 1 and Session 2 (R2 , multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF).
Fig. 4. Regression of the difference between the CT80-measured IOP and the Goldmann-measured IOP (CT80IOP − GATIOP ) on the corneal biomechanical properties in Session 1 and Session 2. (R2 , Multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) central corneal thickness (CCT), (B) corneal hysteresis (CH) and (C) corneal resistance factor (CRF). Table 2 Adjusted multiple regression Coefficients of determination. Dependent variables
Independent variables Session 1 CCT + CH CCT + CRF CH + CRF CCT + CH + CRF Session 2 CCT + CH CCT + CRF CH + CRF CCT + CH + CRF
GATIOP
CT80IOP
CT80IOP − GATIOP
Adjusted R2 = 0.165, F2 ,103 = 11.38, P < 0.0001 Adjusted R2 = 0.322, F2,103 = 25.97, P < 0.0001 Adjusted R2 = 0.540, F2,103 = 62.61, P < 0.0001 Adjusted R2 = 0.536, F3,102 = 41.46, P < 0.0001
Adjusted R2 = 0.477, F2,103 = 48.86, P < 0.0001 Adjusted R2 = 0.614, F2,103 = 84.44, P < 0.0001 Adjusted R2 = 0.682, F2,103 = 113.59, P < 0.0001 Adjusted R2 = 0.703, F3,102 = 83.66, P < 0.0001
Adjusted R2 = 0.215, F2,103 = 15.35, P < 0.0001 Adjusted R2 = 0.195, F2,103 = 13.72, P < 0.0001 Adjusted R2 = 0.155, F2,103 = 10.65, P < 0.0001 Adjusted R2 = 0.220, F3,102 = 10.88, P < 0.0001
Adjusted R2 = 0.301, F2,103 = 23.65, P < 0.0001 Adjusted R2 = 0.404, F2,103 = 36.61, P < 0.0001 Adjusted R2 = 0.522, F2,103 = 58.33, P < 0.0001 Adjusted R2 = 0.537, F3,102 = 41.52, P < 0.0001
Adjusted R2 = 0.409, F2,103 = 37.32, P < 0.0001 Adjusted R2 = 0.514, F2,103 = 56.47, P < 0.0001 Adjusted R2 = 0.573, F2,103 = 71.47, P < 0.0001 Adjusted R2 = 0.606, F3,102 = 54.81, P < 0.0001
Adjusted R2 = 0.006, F2,103 = 1.32, P = 0.27 Adjusted R2 = 0.004, F2,103 = 1.19, P = 0.31 Adjusted R2 = 0.000, F2,103 = 1.00, P = 0.37 Adjusted R2 = −0.003, F3,102 = 0.88, P = 0.45
GATIOP , intraocular pressure measured with the Goldmann applanation tonometer; CT80IOP , intraocular pressure measured with the Topcon CT80 Noncontact Tonometer; CT80IOP − GATIOP , difference between the intraocular pressure measured with the Topcon CT80 noncontact tonometer and that measured with the Goldmann applanation tonometer; CRF, corneal resistance factor; CH, corneal hysteresis; CCT, central corneal thickness; R2 , multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
6
Fig. 5. Regression of the ORA-measured biomechanical properties of the cornea on central corneal thickness in Session 1 and Session 2 (R2 , Multiple coefficient of determination; Fd,r , variance ratiodegree of freedom , residual ): (A) corneal hysteresis (CH) and (B) corneal resistance factor (CRF).
into account, the addition of CCT to the multiple regression model brought slight but statistically significant increases in the predictive ability of the model, except with the GATIOP for Session 1 in which the contribution of CCT to the model was not statistically significant. In Session 2, also with the GATIOP , the contribution of CCT just reached the level of significance (t = 2.06, P = 0.04) set for this study. The multiple linear regressions of ‘CT80IOp − GATIOP ’ on CCT, CH and CRF followed the same trend as the simple linear regressions, showing statistically significant coefficients of determination in the first session but not in the second session. The multiple linear regressions also showed that CCT alone best explained the variation in the differences between tonometers and that once CCT had been accounted for, adding CH and CRF to the model only minimally influenced the predictive ability of the model (the contribution of CRF was not statistically significant – t = −1.31, P = 0.19 – and the contribution of CH just reached statistical significance – t = 2.08, P = 0.04), with the combination of CCT and CRF even having a reduced predictive ability than with CCT alone. All coefficients of determination in the second session were not statistically significant and thus these effects can also be surmised as: ‘small and inconsistent’. Corneal hysteresis and corneal resistance factor both demonstrated consistent, high and statistically significant correlations with CCT, with CRF demonstrating a higher correlation with CCT than CH (Fig. 5). 4. Discussion The ORA is a relatively new noncontact tonometer designed to measure functional biomechanical properties of the cornea in vivo and incorporate these properties to adjust, and possibly improve the accuracy of, the IOP measured by tonometry. The ORA offers the clinician a means to assess biomechanical properties of patients other than CCT. It is therefore imperative to ensure adequate repeatability and reproducibility of these new ORA biomechanical properties as they become more widely incorporated into routine clinical practice. Though earlier studies have assessed repeatability of CH and CRF, none has assessed reproducibility in the manner in which it was assessed in this study. Further, no study has compared the effect on the measured IOP of CCT on the one hand, and CH and CRF on the other, on the same group of subjects and on two separate occasions. The results from this study showed adequate repeatability indices for CCT, CH and CRF. The repeatability indices for CH and
CRF were comparable to the index (measured in the same units) for the Goldmann tonometer. The average CCT did not vary significantly between sessions and the column of differences (generated from the triplicate readings for each subject) did not vary between sessions. Therefore the reproducibility for CCT between sessions was good. The same was true for CH and CRF between sessions. The reproducibility indices for CH and CRF were twice as good for the index with the Goldmann tonometer in this study. The averages, repeatability and reproducibility indices for the corneal biomechanical properties did not vary significantly between men and women. Univariate analysis showed that CRF explained the variation in GATIOP and CT80IOP better than either CCT or CH, but CCT explained the differences between tonometers better than CRF or CH. The effect of the biomechanical properties of the differences between tonometers was not repeatable between sessions. In the univariate analysis (with respect to GATIOP , CT80IOP , and CT80IOP − GATIOP ) CCT, CH and CRF (the independent variables) were all able to explain a significant amount of the variation in the dependent variables, except in session 2 in which none of the independent variables was able to explain a statistically significant portion of the variation of CT80IOP − GATIOP . These results are in agreement with results from studies that have demonstrated CRF to be the most important factor influencing GAT measurements [24,55], with studies showing a significant correlation between GATIOP and CCT [14,24,56], and with studies which reported that CCT is the primary biomechanical parameter influencing the measurement differences between two tonometers [24]. Other studies have reported results in conflict with the findings from this study. Notably, Yu et al. [14] reported that the difference in measured IOP between the tonopen and intracamerally measured IOP was only associated with CRF. Tranchina et al. [41] showed that while CCT, CH and CRF all significantly influenced the difference between a NCT and the GAT, CRF was the best predictor of this difference. In the first measurement session of this study, all three biomechanical parameters influenced the differences between the CT80IOp and the GATIOP , but CCT (not CRF) was the best predictor of these differences. Multivariate analyses of the data from the first session this study showed that CCT, CH and CRF together accounted for 22% (R2 = 0.22, F3,102 = 10.88, P < 0.0001) of the variance between tonometers, similar to the 25% in the Tranchina et al. [41] study. However, these results were not repeated in the second session suggesting that the small effects of the corneal biomechanical properties on the differences between the CT80IOP and the GATIOP are rather variable.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
The results from this study also showed that, in univariate analyses, the GATIOP significantly correlated with CCT in both sessions. In multivariate analyses, once CH and CRF had been taken into account, the addition of CCT to the multivariate model either, did not significantly increase the correlation with GATIOP (in Session 1), or the increased correlation with the GATIOP just reached statistical significance (t = 2.06, P = 0.04, in Session 2). These findings concur with those of other studies [14,53] and suggest that CH and CRF are not completely independent of CCT and may even be functional surrogates of CCT and the tissue material properties of the cornea. This conclusion is supported by the high correlation (in both sessions) of CH and CRF with CCT. The biomechanical properties of the cornea all had a markedly greater influence on the CT80IOP than on the GATIOP . This effect could be explained in part by the greater influence that CH would be expected to have on noncontact tonometers. The corneal viscoelasticity (which CH is supposed to represent) is a dynamic resistance component for which corneal deformation is proportional to the force applied and the rate at which it is applied [19]. The CRF on the other hand, is a resistance component of the cornea for which corneal deformation is proportional only to the applied force [19]. The operating principles of the NCT compared to that of the GAT would suggest that the rate at which the applanation force is applied is significant only for the NCT and thus corneal viscoelasticity would be expected to exert a greater influence on IOP measurements made with a NCT, as was shown in this study. Copt et al. [33] demonstrated the importance that corneal thickness-adjusted IOP measurements could have on the diagnosis and management of glaucoma. They showed that correcting IOP for corneal thickness would cause over 50% of ocular hypertensives to be reclassified as normal, and 30% of normal tension glaucoma patients to be reclassified as having primary open angle glaucoma. It is likely that glaucoma misclassification errors would increase if CH and CRF were also taken into account, giving the importance that several studies have demonstrated for both parameters influencing the measured IOP [24]. In conclusion, this study demonstrated good repeatabilities and test-retest reproducibilities for the ORA biomechanical properties. The Goldmann tonometer provides estimates of true IOP that are less influenced by CCT, CH and CRF than the estimates provided by the CT80 noncontact tonometer. The biomechanical properties of the cornea influenced the differences between the Goldmann tonometer and the CT80 noncontact tonometer but these effects were small and inconsistent. The CRF has the greatest influence on the variation of measured IOP and, when the ORA biomechanical properties are accounted for, CCT has a small to negligible effect on the IOP measured with the Goldmann tonometer. On the same group of oculovisually normal subjects in good health, despite CCT, CH, CRF, GATIOP , and CT80IOP all showing good between-sessions reproducibility, the effect of CCT on the GATmeasured IOP was not consistent between sessions, in contrast those of CRF and CH. This suggests that the ORA biomechanical properties are better, more consistent predictors (than CCT) of the variation in trans-corneally measured IOP. A limitation of this study is the non-inclusion of a wider variety of subjects, including those with anterior segment disease, thereby restricting the findings of this study to young oculovisual normals. As such, prospective, longitudinal studies in a wider variety of subjects are required to ascertain the association between the ORA biomechanical properties and IOP measured by tonometry. Conflict of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
7
Acknowledgements The authors extend their appreciation to the Research Centre, College of Applied Medical Sciences and to the Deanship of Scientific Research at King Saud University for funding this research.
References [1] Collaborative Normal-Tension Glaucoma Study Group. Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures. Am J Ophthalmol 1998;126(4):487–97. [2] Collaborative Normal-Tension Glaucoma Study Group. The effectiveness of intraocular pressure reduction in the treatment of normal-tension glaucoma. Am J Ophthalmol 1998;126(4):498–505. [3] The 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(4):429–40. [4] Kass MA, Heuer DK, Higginbotham EJ, Johnson CA, Keltner JL, Miller JP, 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(6):829–30, 701-713; discussion. [5] Gordon MO, Beiser JA, Brandt JD, Heuer DK, Higginbotham EJ, Johnson CA, et al. The ocular hypertension treatment study: baseline factors that predict the onset of primary open-angle glaucoma. Arch Ophthalmol 2002;120(6):829–30, 714-720; discussion. [6] 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):205–9. [7] 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):1965–72. [8] Ehlers N, Bramsen T, Sperling S. Applanation tonometry and central corneal thickness. Acta Ophthalmol (Copenh) 1975;53(1):34–43. [9] Whitacre MM, Stein R. Sources of error with use of Goldmann-type tonometers. Surv Ophthalmol 1993;38(1):1–30. [10] Dielemans I, Vingerling JR, Hofman A, Grobbee DE, de Jong PT. Reliability of intraocular pressure measurement with the Goldmann applanation tonometer in epidemiological studies. Graefes Arch Clin Exp Ophthalmol 1994;232(3):141–4. [11] Doughty MJ, Zaman ML. Human corneal thickness and its impact on intraocular pressure measures: a review and meta-analysis approach. Surv Ophthalmol 2000;44(5):367–408. [12] Liu J, Roberts CJ. Influence of corneal biomechanical properties on intraocular pressure measurement: quantitative analysis. J Cataract Refract Surg 2005;31(1):146–55. [13] Feltgen N, Leifert D, Funk J. Correlation between central corneal thickness, applanation tonometry, and direct intracameral IOP readings. Br J Ophthalmol 2001;85(1):85–7. [14] Yu AY, Duan SF, Zhao YE, Li XY, Lu F, Wang J, et al. Correlation between corneal biomechanical properties, applanation tonometry and direct intracameral tonometry. Br J Ophthalmol 2012;96(5):640–4. [15] Whitacre MM, Stein RA, Hassanein K. The effect of corneal thickness on applanation tonometry. Am J Ophthalmol 1993;115(5):592–6. [16] Foster PJ, Wong JS, Wong E, Chen FG, Machin D, Chew PT. Accuracy of clinical estimates of intraocular pressure in Chinese eyes. Ophthalmology 2000;107(10):1816–21. [17] Luce DA. Determining in vivo biomechanical properties of the cornea with an ocular response analyzer. J Cataract Refract Surg 2005;31(1):156–62. [18] Shah S, Laiquzzaman M, Bhojwani R, Mantry S, Cunliffe I. Assessment of the biomechanical properties of the cornea with the ocular response analyzer in normal and keratoconic eyes. Invest Ophthalmol Vis Sci 2007;48(7):3026–31. ˜ D, Shabayek MH, Arnalich-Montiel F, Alió JL. Corneal biome[19] Ortiz D, Pinero chanical properties in normal, post-laser in situ keratomileusis, and keratoconic eyes. J Cataract Refract Surg 2007;33(8):1371–5. [20] Pepose JS, Feigenbaum SK, Qazi MA, Sanderson JP, Roberts CJ. Changes in corneal biomechanics and intraocular pressure following LASIK using static, dynamic, and noncontact tonometry. Am J Ophthalmol 2007;143:39–47. [21] Hager A, Loge K, Füllhas MO, Schroeder B, Grossherr M, Wiegand W. Changes in corneal hysteresis after clear corneal cataract surgery. Am J Ophthalmol 2007;144(3):341–6. [22] González-Méijome JM, Queirós A, Jorge J, Díaz-Rey A, Parafita MA. Intraoffice variability of corneal biomechanical parameters and intraocular pressure (IOP). Optom Vis Sci 2008;85(6):457–62. [23] Congdon NG, Broman AT, Bandeen-Roche K, Grover D, Quigley HA. Central corneal thickness and corneal hysteresis associated with glaucoma damage. Am J Ophthalmol 2006;141(5):868–75. [24] Medeiros FA, Weinreb RN. Evaluation of the influence of corneal biomechanical properties on intraocular pressure measurements using the ocular response analyzer. J Glaucoma 2006;15(5):364–70. ˜ J, Maldonado MJ, García N, Mendiluce L, García-Gómez PJ, [25] Moreno-Montanés Seguí-Gómez M. Reproducibility and clinical relevance of the ocular response
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006
G Model CLAE-658; No. of Pages 8
ARTICLE IN PRESS K.C. Ogbuehi, U.L. Osuagwu / Contact Lens & Anterior Eye xxx (2013) xxx–xxx
8
[26]
[27] [28]
[29]
[30]
[31]
[32] [33]
[34]
[35]
[36]
[37]
[38]
[39]
analyzer in nonoperated eyes: corneal biomechanical and tonometric implications. Invest Ophthalmol Vis Sci 2008;49(3):968–74. Sullivan-Mee M, Billingsley SC, Patel AD, Halverson KD, Alldredge BR, Qualls C. Ocular response analyzer in subjects with and without glaucoma. Optom Vis Sci 2008;85(6):463–70. Morita T, Shoji N, Kamiya K, Fujimura F, Shimizu K. Corneal biomechanical properties in normal-tension glaucoma. Acta Ophthalmol 2012;90(1):e48–53. Kara N, Altinkaynak H, Baz O, Goker Y. Biomechanical evaluation of cornea in topographically normal relatives of patients with keratoconus. Cornea 2012, http://dx.doi.org/10.1097/ICO.0b013e3182490924 (cited 7.9.12). De Moraes CV, Hill V, Tello C, Liebmann JM, Ritch R. Lower corneal hysteresis is associated with more rapid glaucomatous visual field progression. J Glaucoma 2012;21(4):209–13. Mansouri K, Leite MT, Weinreb RN, Tafreshi A, Zangwill LM, Medeiros FA. Association between corneal biomechanical properties and glaucoma severity. Am J Ophthalmol 2012;153(3):419–27. Franco S, Lira M. Biomechanical properties of the cornea measured by the ocular response analyzer and their association with intraocular pressure and the central corneal curvature. Clin Exp Optom 2009;92(6):469–75. Morad Y, Sharon E, Hefetz L, Nemet P. Corneal thickness and curvature in normal-tension glaucoma. Am J Ophthalmol 1998;125(2):164–8. Copt RP, Thomas R, Mermoud A. Corneal thickness in ocular hypertension, primary open-angle glaucoma, and normal tension glaucoma. Arch Ophthalmol 1999;117(1):14–6. Choi HJ, Kim DM, Hwang SS. Relationship between central corneal thickness and localized retinal nerve fiber layer defect in normal-tension glaucoma. J Glaucoma 2006;15(2):120–3. 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):3262–70. Kirwan C, O’Keefe M, Lanigan B. Corneal hysteresis and intraocular pressure measurement in children using the Reichert ocular response analyzer. Am J Ophthalmol 2006;142(6):990–2. Sullivan-Mee M, Gerhardt G, Halverson KD, Qualls C. Repeatability and reproducibility for intraocular pressure measurement by dynamic contour, ocular response analyzer, and Goldmann applanation tonometry. J Glaucoma 2009;18(9):666–73. Wasielica-Poslednik J, Berisha F, Aliyeva S, Pfeiffer N, Hoffmann EM. Reproducibility of ocular response analyzer measurements and their correlation with central corneal thickness. Graefes Arch Clin Exp Ophthalmol 2010;248(11):1617–20. Detry-Morel M, Jamart J, Pourjavan S. Evaluation of corneal biomechanical properties with the Reichert ocular response analyzer. Eur J Ophthalmol 2011;21(2):138–48.
[40] Mangouritsas G, Mourtzoukos S, Mantzounis A, Alexopoulos L. Comparison of Goldmann and Pascal tonometry in relation to corneal hysteresis and central corneal thickness in nonglaucomatous eye. Clin Ophthalmol 2011;5: 1071–80. [41] Tranchina L, Lombardo M, Oddone F, Serrao S, Lomoriello DS, Ducoli P. Influence of corneal biomechanical properties on intraocular pressure differences between an air-puff tonometer and the Goldmann applanation tonometer. J Glaucoma 2012, http://dx.doi.org/10.1097/IJG.0b013e31824cafc9 (cited 7.9.12). [42] Liu JH, Kripke DF, Twa MD, Hoffman RE, Mansberger SL, Rex KM, et al. Twentyfour-hour pattern of intraocular pressure in the aging population. Invest Ophthalmol Vis Sci 1999;40(12):2912–20. [43] AlMubrad TM, Ogbuehi KC. On repeated corneal applanation with the Goldmann and two non-contact tonometers. Clin Exp Optom 2010;93(2):77–82. [44] AlMubrad TM, Ogbuehi KC. The effect of repeated applanation on subsequent IOP measurements. Clin Exp Optom 2008;91(6):524–9. [45] Stocker FW. On changes in intraocular pressure after application of the tonometer; in the same eye and in the other eye. Am J Ophthalmol 1958;45(2):192–6. [46] Moses RA, Liu CH. Repeated applanation tonometry. Am J Ophthalmol 1968;66(1):89–91. [47] Sorensen PN. The non-contact tonometer. Clinical evaluation on normal and diseased eyes. Acta Ophthalmol (Copenh) 1975;53(4):513–21. [48] Chauhan BC, Henson DB. Clinical evaluation of the non-contact tonometer Mark II. Am J Optom Physiol Opt 1988;65:751–6. [49] Bynke HG, Schele B. On the origin of the ocular pressure pulse. Ophthalmologica 1967;153:29–36. [50] Leydhecker W. The intraocular pressure clinical aspects. Ann Ophthalmol 1976;8:389–401. [51] Perkins ES. The ocular pulse. Curr Eye Res 1981;1:19–23. [52] Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioural, and biomedical sciences. Behav Res Methods 2007;39:175–91. [53] Bland JM, Altman DG. Measurement error. BMJ 1996;313(7059):744. [54] O’Brien RM. A caution regarding rules of thumb for variance inflation factors. Qual Quant 2007;41:673–90. [55] Kaushik S, Pandav SS, Banger A, Aggarwal K, Gupta A. Relationship between corneal biomechanical properties, central corneal thickness, and intraocular pressure across the spectrum of glaucoma. Am J Ophthalmol 2012;153(5):840–9. [56] Tonnu PA, Ho T, Newson T, El Sheikh A, Sharma K, White E, et al. The influence of central corneal thickness and age on intraocular pressure measured by pneumotonometry, non-contact tonometry, the Tono-Pen XL, and Goldmann applanation tonometry. Br J Ophthalmol 2005;89(7):851–4.
Please cite this article in press as: Ogbuehi KC, Osuagwu UL. Corneal biomechanical properties: Precision and influence on tonometry. Contact Lens Anterior Eye (2013), http://dx.doi.org/10.1016/j.clae.2013.09.006