Corneal biomechanical evaluation could improve the safety of refractive surgery because many of today’s cases of ectasia following LASIK are ascribed to undetected corneal biomechanical instability or subclinical keratoconus1 (ie, corneas without the typical topography findings that develop manifest ectasia after years of follow-up). The detection of this increased susceptibility to postoperative ectasia has been greatly aided by imaging tools that analyze early corneal signs of keratoconus development such as anterior surface higher-order aberrations,2 abnormal epithelial thickness,3 pachymetric progression,4 or posterior surface elevation.5 There is nonetheless the possibility that a surgical candidate with ectasia propensity may consult before these morphological changes arise, especially considering the variable age of manifest keratoconus onset and its asymmetric development,6 and thus a reliable method to assay corneal biomechanics might be able to uncover the stromal instability before the actual corneal morphology is modified by the ectatic process.
The Ocular Response Analyzer (ORA) (Reichert Ophthalmic Instruments, Depew, NY) was the first corneal biomechanical testing device approved for clinical use, but the lack of normative databases and ample clinical experience precluded its utility in the office setting. Corneal evaluation with the ORA yields two parameters, corneal hysteresis (CH) and corneal resistance factor (CRF), which convey useful information but are partially dependent on central corneal thickness (CCT).7,8 We have previously developed a method for correcting for this influence, which markedly improved the detection rate of subclinical keratoconus cases (high sensitivity) and is practical for clinical use.9 However, our approach suffered from low specificity and it was our experience that this was especially true when dealing with non-keratoconic myopic eyes, which seem to exhibit a different biomechanical profile10 but are not necessarily linked to a higher risk for post-LASIK ectasia. This is in agreement with previous reports that show that the corneas and sclera of highly myopic eyes are characterized by ultrastructural changes11–13 and, in line with these studies, several groups have reported reduced CH and/or CRF values in non-keratoconic myopic eyes.10,14–18
Our aim was to explore the biomechanical profile of non-keratoconic myopic eyes as measured by the ORA and to evaluate whether the collected information could be included in a multiparametric model to increase the aforementioned performance for subclinical keratoconus detection.
Patients and Methods
The study was an observational case series. The research protocol followed the tenets of the Declaration of Helsinki and was approved by an ethics committee. All participants were told of the purpose of the study and gave written informed consent before inclusion. Patients were recruited between March 2010 and November 2011 at ECOS Laboratory and had been referred for spectacle or contact lens prescription or keratoconus diagnosis. Each participant underwent slit-lamp examination, anterior segment optical coherence tomography (OCT) evaluation (software version 126.96.36.199, Visante OCT; Carl Zeiss Meditec, Dublin, CA), Placido disk topography and aberrometry (software version 4.0 iTrace; Tracey Technologies, Houston, TX), and ORA measurements (software version 2.04; Reichert Ophthalmic Instruments, Depew, NY).
For topography and keratoconus grading, the Keratoconus Severity Score (KSS)19 was used, which is based on average corneal power (in diopters [D]) and corneal higher-order aberrations (HOA, expressed in μm as root-mean-square values). The KSS scale ranges from 0 to 5: 0 (unaffected, normal topography), 1 (unaffected, atypical topography), 2 (suspect), 3 (mild keratoconus), 4 (moderate keratoconus), and 5 (severe keratoconus). Two separate datasets were compiled for different purposes. For dataset 1, only non-keratoconic participants were recruited from ammetropic patients without any clinical signs of ectasia referred for spectacle prescription or contact lens fitting if the KSS of both eyes was 0 (unremarkable topography defined as typical axial pattern and HOA < 0.65 μm). Such strict topographical criteria for non-keratoconic eyes were adapted from Buhren et al.,2 because some subclinical keratoconus cases initially show increased HOA. From March to December 2010, one eye per patient was randomly chosen and included in dataset 1 depending on the objective refraction’s spherical equivalent: greater than −5 D was considered non-myopic (group 1A) and −5 D or less was considered myopic (group 1B).
Dataset 2, which included non-keratoconic and keratoconic eyes, was compiled from December 2010 to November 2011. Non-keratoconic patients, defined as for dataset 1, were recruited in group 2A irrespective of objective refraction, whereas verified subclinical keratoconus cases were included in group 2B. Following the criteria of Buhren et al.,2 the latter were operationally defined as eyes with KSS 0, 1, or 2 from patients with manifest keratoconus (KSS 3 or higher) in the fellow eye: axial topography pattern consistent with keratoconus, may have positive slit-lamp findings, no corneal scarring, average corneal power greater than 49.00 D, and HOA greater than 1.50 μm. Participant exclusion criteria were: age younger than 18 years, previous eye surgery, any eye disease for dataset 1 and group 2A or eye disease other than keratoconus for group 2B, and chronic use of topical medications or corneal opacities. Contact lenses were removed at least 72 hours before examination.
Patients underwent a complete clinical evaluation and testing with an ORA, corneal topography, and tomography, all of which were performed by two trained ophthalmologists (MD and JGG) between 1 PM and 6 PM. Four consecutive ORA measurements without topical anesthesia were obtained from both eyes and averaged (only good-quality readings, as defined by the manufacturer, were stored). The details of the ORA function and the applanation pressures from which both CH and CRF are derived have been addressed elsewhere.20 The methodology for calculating DifCH and DifCRF to account for the influence of corneal thickness has been recently reported,9 and that for CH-CRF was published by Touboul et al.7 Topographic examinations with artifacts or irregularities were discarded. Topographic indices such as average corneal power and HOA of the corneal 5-mm central surface were provided by iTrace software. CCT was obtained from the mean value of the 2-mm central area of the OCT Visante pachymetry map obtained after centering in the pupil.
Combinations of biomechanical descriptors (DifCH, DifCRF, and CH-CRF) and ACP were constructed by logistic regression performed on dataset 2 (non-keratoconic and keratoconic eyes), either by entering all variables at once or by stepwise inclusion. Receiver operating characteristic (ROC) curves were used to calculate sensitivity, specificity, and area under the curve (AROC) of each biomechanical descriptor taken separately and of each logistic function. ROC curve analysis and all statistical tests were performed with Prism 5 software (GraphPad Software, La Jolla, CA) and SPSS 17 software (SPSS, Inc., Chicago, IL). Statistical significance was set at a P value less than .05 and data are shown as mean ± standard deviation unless otherwise stated. Data collection and sorting were done with the aid of Microsoft Excel 2010 software (Microsoft Corporation, Redmond, WA).
Corneal Biomechanics in Non-Keratoconic Non-Myopic, and Non-Keratoconic Myopic Eyes
Dataset 1 comprised 52 non-keratoconic non-myopic eyes (group 1A) and 97 non-keratoconic myopic eyes (group 1B), as summarized in Table A, available in the online version of this article. Mean age (38.0 ± 13.9 vs 33.8 ± 9.9 years, P = .03), but not gender distribution, was different between groups. CCT, ACP, and corneal HOA were similar, whereas ACD was higher in group 1B (3.07 ± 0.41 vs 3.23 ± 0.30 mm, P = .03). CH and CRF were lower in non-keratoconic myopic eyes, but not CH-CRF. CH and CRF were positively correlated with CCT in both groups (group 1A: r = 0.336, P = .015 and r = 0.517, P < .001, respectively; group 1B: r = 0.391, P < .001 and r = 0.450, P < .001, respectively), whereas ACP, but not SE, was positively correlated with CH (r = 0.344, P = .001), CRF (r = 0.231, P = .02), DifCH (r = 0.350, P < .001), and DifCRF (r = 0.228, P = .03) in non-keratoconic myopic eyes only (Figure 1).
Figure 1. Correlation between corneal power and biomechanical indices in non-keratoconic eyes. Correlation graphs for corneal hysteresis and resistance factor and average corneal power are shown for dataset 1 eyes. Pearson’s r correlation and P values are shown where statistically significant. CH = corneal hysteresis; CRF = corneal resistance factor; ACP = average corneal power
We reasoned that the aforementioned correlation between corneal power and biomechanical properties could be explained by heterogeneity of the non-keratoconic myopic group, comprising eyes with axial myopia and lower ACP and eyes with curvature myopia with comparatively higher corneal power. Supporting this hypothesis, the CH and CRF distribution in non-keratoconic myopic eyes, but not in non-keratoconic non-myopic eyes, seemed to contain two subgroups. To explore this possibility, non-keratoconic myopic (group 1B) observations were stratified into low-ACP (< 44.0 D) and high-ACP myopic eyes (⩾ 44.0 D) by taking the mean ACP as the cutoff point. In agreement with this line of reasoning, there was no statistically significant correlation between the biomechanical indices and ACP if both non-keratoconic myopic subgroups were analyzed separately. Compared to non-keratoconic non-myopic eyes, only low-ACP non-keratoconic myopic eyes showed lower mean CH (10.02 ± 1.34 vs 9.01 ± 1.58, P < .01) and CRF (8.86 ± 1.71, P < .01), whereas high-ACP non-keratoconic myopic eyes exhibited higher mean CH-CRF (0.53 ± 1.03, P = .03). After correcting for the influence of CCT on ORA measurements, DifCH was not significantly different in non-keratoconic myopic eyes as a whole (Figure 2), but was lower in the low-ACP subgroup (0.11 ± 1.27 vs −0.79 ± 1.50, P < .01). Non-keratoconic myopic eyes also showed reduced mean DifCRF (0.24 ± 0.16 vs −0.47 ± 0.17, P < .01), which was significantly lower only in low-ACP eyes (0.24 ± 0.16 vs 0.70 ± 1.59, P < .01). From these results, we chose low ACP as a surrogate indicator of a non-keratoconic myopic biomechanical pattern that is potentially different from the normal biomechanical signature and included this variable in the logistic regression analysis for subclinical keratoconus detection.
Figure 2. Biomechanical indices of emmetropic and myopic eyes. Mean ± standard error of measurement of DifCH, DifCRF, and CH-CRF indices are shown for non-keratoconic non-myopic (group 1A, empty bars) and non-keratoconic myopic (group 1B, black bars) eyes, and for the latter, observations with low (light blue bars) and with high average corneal power (dark blue bars) are plotted separately for comparison. * indicates a statistical significant difference in means by analysis of variance with Dunnett’s post hoc test. DifCH and DifCRF = central corneal thickness-corrected indices; CH = corneal hysteresis; CRF = corneal resistance factor
Corneal Biomechanics in Non-Keratoconic and Keratoconic Eyes
For dataset 2 (Table B, available in the online version of this article), group 2A consisted of 87 non-keratoconic eyes (36 non-myopic and 51 myopic eyes) and group 2B comprised 73 eyes with subclinical keratoconus (KSS distribution: 45 [61%] KSS 0, 15 [21%] KSS 1 and 13 [18%] KSS 2 eyes). Gender distribution (male, 35% vs 63%), but not age, was significantly different between groups. Mean ACP was similar in both groups, whereas mean HOA were higher (0.229 ± 0.083 vs 0.586 ± 0.382 μm, P < .01) and mean CCT was lower (513.4 ± 30.0 vs 497.8 ± 31.3 μm, P < .01) in group 2B eyes. In these keratoconic eyes (Figure 3), DifCH (−0.17 ± 1.30 vs −1.04 ± 1.19, P < .01) and DifCRF (−0.20 ± 1.23 vs −1.87 ± 1.13, P < .01) were reduced, whereas CH-CRF (0.27 ± 0.96 vs 1.13 ± 0.77, P < .01) was increased compared to non-keratoconic eyes. In the latter, DifCH (r = 0.299, P = .005) and DifCRF (r = 0.244, P = .023) but not CH-CRF were positively correlated to ACP.
Figure 3. Biomechanical indices of emmetropic, myopic and keratoconic eyes. Whisker-box plots of DifCH, DifCRF, and CH-CRF indices are shown for non-keratoconic non-myopic (group 2A with spherical equivalent > −5 D, empty boxes), non-keratoconic myopic (group 2A observations with spherical equivalent ⩽ −5 D, light blue boxes), and keratoconic (group 2B, dark blue boxes) eyes. * indicates a statistical significant difference in means by analysis of variance with Bonferroni’s post hoc test. DifCH and DifCRF = central corneal thickness-corrected indices; CH = corneal hysteresis; CRF = corneal resistance factor
The non-keratoconic group included both non-myopic and myopic eyes to evaluate their confounding effect. Combinations of DifCH, DifCRF, CH-CRF, and ACP were generated by logistic regression and evaluated for their capacity to discriminate between non-keratoconic eyes and those with keratoconus (Table 1). The resulting logistic functions were as follows: LF1 = 1.397*DifCH + 0.309*DifCRF + 2.274*CH-CRF – 2.365; LF2 = −1.112*DifCRF + 0.900*CH-CRF – 2.034; LF3 = −0.673*DifCH – 0.494*DifCRF + 1.410*CH-CRF + 0.186*ACP – 10.319; and LF4 = 1.182*DifCRF + 0.743*CH-CRF + 0.194*ACP – 10.509. The previously derived DifCRF cutoff point (< −0.695)9 correctly diagnosed 71.3% of non-keratoconic eyes (80.6% of non-keratoconic non-myopic and 64.7% of non-keratoconic myopic eyes) and 86.3% of keratoconic eyes, whereas the logistic function LF2 combining DifCRF and CH-CRF correctly identified 81.6% of non-keratoconic eyes (91.7% of non-keratoconic non-myopic and 74.5% of non-keratoconic myopic eyes) and 78.1% of keratoconic eyes. All indices and logistic functions but CH-CRF were correlated to spherical equivalent in the control group.
Table 1: Diagnostic Performance of Individual Biomechanical Indices and Their Combinations
The integration of corneal biomechanical testing with the ORA to clinical practice holds promise because there is the prospect of early detection of subtle changes in corneal structure that predate keratoconus onset. However, much needs to be learned about the biomechanical pattern of healthy and diseased corneas and the different factors that influence ORA readings. This work takes on an already validated approach to interpreting biomechanical descriptors in non-keratoconic non-myopic, non-keratoconic myopic, and keratoconic eyes and further explores its diagnostic capabilities. It should be emphasized that this subclinical keratoconus group only included eyes with insufficient topographical findings to be diagnosed as keratoconic but with verified ectasia predisposition defined by the manifest disease in the fellow eye. Placido topography was markedly ineffective with these eyes, only flagging 18% of the subclinical keratoconus cases as suspect. With this benchmark, the performance of the corneal thickness-corrected CRF index (86% sensitivity, 72% specificity) represents a significant improvement and is comparable to our preceding report.9 Nonetheless, the false-positive rate was unacceptably low with non-keratoconic myopic eyes, leading us to focus on this subgroup.
The aforementioned finding supported the analysis of an independent dataset of non-keratoconic non-myopic, and myopic eyes for additional descriptors that might improve the diagnostic capacity of corneal biomechanical testing for keratoconus. The non-keratoconic myopic eyes exhibited a different biomechanical profile, with an even reduction in CH and CRF values. In comparison, the previously described keratoconus profile is characterized by an asymmetric and more profound drop in these biomechanical indicators, which is more marked for CRF than for CH when considered independently of the influence of corneal thickness.9 These findings are in agreement with previous reports that analyzed only keratoconic eyes21 and indirectly described this tendency as an increase in the CH-CRF index.
Although they were reduced in non-keratoconic myopic eyes, there was considerable overlap in CH and CRF between non-keratoconic non-myopic and non-keratoconic myopic eyes, which led us to believe that these indices would not be clinically useful for differentiating a non-keratoconic myopic pattern when considered separately. Indeed, mean DifCH was not significantly lower if all non-keratoconic myopic eyes were analyzed altogether, suggesting that there was an additional confounding effect of corneal thickness to be considered. Remarkably, there was a positive correlation between corneal curvature (or ACP) and these biomechanical indices only in non-keratoconic myopic eyes. This correlation could not be attributed to an effect of corneal curvature on ORA readings, as discussed by Franco et al.,22 because there was no difference in mean ACP between non-keratoconic myopic and non-myopic eyes. We reasoned that this could be due to a heterogeneous non-keratoconic myopic sample comprising axially myopic eyes with flatter corneas and myopic eyes with steeper corneal curvature and normal axial length.
Post hoc analysis confirmed that low-ACP non-keratoconic myopic eyes had markedly reduced DifCH and lower DifCRF values, whereas high-ACP non-keratoconic myopic eyes showed DifCH and DifCRF values comparable to non-keratoconic non-myopic eyes. We cannot conclude from these results that axial length is actually influencing ORA measurements because this study was never intended to evaluate this eye property. Of note, Lim et al.15 found a similar correlation between CH and CRF and corneal curvature in children, but no correlation of these biomechanical indices with axial length. In addition, others have shown that flatter corneas in and of themselves are less rigid,23 and thus it is plausible that the reduced CH and CRF values could be exposing an underlying corneal structure that is more frequent in non-keratoconic myopic eyes. In any case, from these results we hypothesized that the inclusion of corneal curvature into a multi-parametric model might contribute to its diagnostic performance by reducing the number of false-positive non-keratoconic myopic eyes.
Regarding the ability of biomechanical testing to detect keratoconus, the simple DifCRF approach led to the highest sensitivity and acceptable specificity (> 80%) if only non-keratoconic non-myopic eyes were considered for the control group. Although the actual performance in the clinical setting would probably be not as good, this type of analysis reflects that ORA testing is indeed measuring an intrinsic corneal property, (the CRF) that readily differentiates keratoconic eyes if other confounding factors are considered. In addition to the previously described influence of corneal thickness, biomechanical traits that seem to be associated with non-keratoconic myopic eyes with flat corneas should be taken into account to improve its usability. Confirming this hypothesis, different multiparametric models that included DifCH, CH-CRF, and ACP along with DifCRF were able to increase specificity to more than 80%, but at the cost of some sacrifice in sensitivity. The differences in performance between these models were not clinically meaningful, suggesting that the consideration of one additional variable such as CH-CRF would be sufficient to achieve these changes. It remains to be established how these multiparametric models fare with an independent dataset to truly assess their diagnostic capacity.
As delineated by Terai et al.,24 CH and CRF are influenced by corneal thickness and intraocular pressure, highlighting the importance of correcting for these factors when deriving standardized indices that can be successfully employed in the office setting. For subclinical keratoconus detection, the limited intraocular pressure range of refractive surgery candidates should not bear a significant effect on ORA metrics and we strived for the most straightforward approach by considering only participants with normal intraocular pressure. In this way, we have further improved our diagnostic model by considering more than one biomechanical index for detecting non-keratoconic corneas. Nonetheless, we believe that much more useful information can be obtained by analyzing the entire corneal waveform signals obtained with the ORA than rather focusing on two or three metrics. The newer ORA software versions provide 39 additional waveform descriptors that, according to the analysis of Mikielewicz et al.,25 could contribute relevant information to the detection of underlying ectasia. According to the manufacturer, the newest ORA software version 3.01 is currently bundled with a multiparametric algorithm that improves its clinical usability. Unfortunately, to the best of our knowledge, there are no independent published validations of these claims and the fact that the underlying model has not been disclosed calls for more work on this subject.
This study highlights the importance of considering multiple factors when attempting to characterize an eye from a biomechanical viewpoint for keratoconus diagnosis. Moreover, it lends credit to the observation that some non-keratoconic myopic eyes behave differently from non-keratoconic non-myopic eyes under ORA testing, and perhaps this might also apply to other biomechanical testing devices that are under development. This finding could be due to the reported ultrastructural changes in axial myopia. However, this study has some limitations regarding this latter assertion because it was not conceived for this type of analysis and thus this conclusion needs to be tested. Nonetheless, the inclusion of corneal curvature or hysteresis into the diagnostic model empirically improved its specificity for myopic eyes, which was a weak point of the DifCRF approach. Overall, we believe that the findings in this work expand the clinical experience with the ORA and could contribute to the screening of refractive surgery candidates.
- Klein SR, Epstein RJ, Randleman JB, Stulting RD. Corneal ectasia after laser in situ keratomileusis in patients without apparent preoperative risk factors. Cornea. 2006;25:388–403. doi:10.1097/01.ico.0000222479.68242.77 [CrossRef]
- Buhren J, Kuhne C, Kohnen T. Defining subclinical keratoconus using corneal first-surface higher-order aberrations. Am J Ophthalmol. 2007;143:381–389. doi:10.1016/j.ajo.2006.11.062 [CrossRef]
- Reinstein DZ, Archer TJ, Gobbe M. Stability of LASIK in topographically suspect keratoconus confirmed non-keratoconic by Artemis VHF digital ultrasound epithelial thickness mapping: 1-year follow-up. J Refract Surg. 2009;25:569–577.
- Ambrósio R, Alonso RS, Luz A, Coca Velarde LG. Corneal-thickness spatial profile and corneal-volume distribution: tomographic indices to detect keratoconus. J Cataract Refract Surg. 2006;32:1851–1859. doi:10.1016/j.jcrs.2006.06.025 [CrossRef]
- de Sanctis U, Loiacono C, Richiardi L, Turco D, Mutani B, Grignolo FM. Sensitivity and specificity of posterior corneal elevation measured by Pentacam in discriminating keratoconus/subclinical keratoconus. Ophthalmology. 2008;115:1534–1539. doi:10.1016/j.ophtha.2008.02.020 [CrossRef]
- Li X, Rabinowitz YS, Rasheed K, Yang H. Longitudinal study of the normal eyes in unilateral keratoconus patients. Ophthalmology. 2004;111:440–446. doi:10.1016/j.ophtha.2003.06.020 [CrossRef]
- Touboul D, Roberts C, Kerautret J, et al. Correlations between corneal hysteresis, intraocular pressure, and corneal central pachymetry. J Cataract Refract Surg. 2008;34:616–622. doi:10.1016/j.jcrs.2007.11.051 [CrossRef]
- Fontes BM, Ambrósio R, Velarde GC, Nosé W. Ocular response analyzer measurements in keratoconus with normal central corneal thickness compared with matched normal control eyes. J Refract Surg. 2011;27:209–215.
- Galletti JG, Pförtner T, Bonthoux FF. Improved keratoconus detection by ocular response analyzer testing after consideration of corneal thickness as a confounding factor. J Refract Surg. 2012;28:202–208. doi:10.3928/1081597X-20120103-03 [CrossRef]
- Shen M, Fan F, Xue A, Wang J, Zhou X, Lu F. Biomechanical properties of the cornea in high myopia. Vision Res. 2008;48:2167–2171. doi:10.1016/j.visres.2008.06.020 [CrossRef]
- Rada JAS, Shelton S, Norton TT. The sclera and myopia. Exp Eye Res. 2006;82:185–200. doi:10.1016/j.exer.2005.08.009 [CrossRef]
- McBrien NA, Jobling AI, Gentle A. Biomechanics of the sclera in myopia: extracellular and cellular factors. Optom Vis Sci. 2009;86:E23–E30. doi:10.1097/OPX.0b013e3181940669 [CrossRef]
- Grossniklaus HE, Green WR. Pathologic findings in pathologic myopia. Retina. 1992;12:127–133. doi:10.1097/00006982-199212020-00009 [CrossRef]
- Chang PY, Chang SW, Wang JY. Assessment of corneal biomechanical properties and intraocular pressure with the Ocular Response Analyzer in childhood myopia. Br J Ophthalmol. 2010;94:877–881. doi:10.1136/bjo.2009.158568 [CrossRef]
- Lim L, Gazzard G, Chan YH, et al. Cornea biomechanical characteristics and their correlates with refractive error in Singaporean children. Invest Ophthalmol Vis Sci. 2008;49:3852–3857. doi:10.1167/iovs.07-1670 [CrossRef]
- Song Y, Congdon N, Li L, et al. Corneal hysteresis and axial length among Chinese secondary school children: the Xichang Pediatric Refractive Error Study (X-PRES) report no. 4. Am J Ophthalmol. 2008;145:819–826. doi:10.1016/j.ajo.2007.12.034 [CrossRef]
- Xu S, Xu A, Tao A, Wang J, Fan F, Lu F. Corneal biomechanical properties and intraocular pressure in high myopic anisometropia. Eye Contact Lens. 2010;36:204–209. doi:10.1097/ICL.0b013e3181e4a60a [CrossRef]
- Jiang Z, Shen M, Mao G, et al. Association between corneal biomechanical properties and myopia in Chinese subjects. Eye (Lond). 2011;25:1083–1089. doi:10.1038/eye.2011.104 [CrossRef]
- McMahon TT, Szczotka-Flynn L, Barr JT, et al. A new method for grading the severity of keratoconus: the Keratoconus Severity Score (KSS). Cornea. 2006;25:794–800. doi:10.1097/01.ico.0000226359.26678.d1 [CrossRef]
- Luce D. Methodology for cornea compensated IOP and corneal resistance factor for the Reichert Ocular Response Analyzer. Invest Ophthalmol Vis Sci. 2006;47:2266.
- Saad A, Lteif Y, Azan E, Gatinel D. Biomechanical properties of keratoconus suspect eyes. Invest Ophthalmol Vis Sci. 2010;51:2912–2916. doi:10.1167/iovs.09-4304 [CrossRef]
- 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;92:469–475. doi:10.1111/j.1444-0938.2009.00414.x [CrossRef]
- Francis BA, Hsieh A, Lai M-Y, et al. Effects of corneal thickness, corneal curvature, and intraocular pressure level on Goldmann applanation tonometry and dynamic contour tonometry. Ophthalmology. 2007;114:20–26. doi:10.1016/j.ophtha.2006.06.047 [CrossRef]
- Terai N, Raiskup F, Haustein M, Pillunat LE, Spoerl E. Identification of biomechanical properties of the cornea: the Ocular Response Analyzer. Curr Eye Res. 2012;37:553–562. doi:10.3109/02713683.2012.669007 [CrossRef]
- Mikielewicz M, Kotliar K, Barraquer RI, Michael R. Air-pulse corneal applanation signal curve parameters for the characterisation of keratoconus. Br J Ophthalmol. 2011;95:793–798. doi:10.1136/bjo.2010.188300 [CrossRef]
Diagnostic Performance of Individual Biomechanical Indices and Their Combinations