CLINICAL SCIENCE
Air-pulse Corneal Applanation Signal Curve Parameters for Characterization of Astigmatic Corneas Omer Trivizki, MD,* Jonathan Shahar, MD,* Samuel Levinger, MD,† and Eliya Levinger, MD*†
Purpose: The aim of this study was to test the 42 parameters of the ocular response analyzer for distinguishing between the biomechanical properties of emmetropic eyes with normal topography and eyes with moderate-to-high with-the-rule astigmatism (WTA) and against-the-rule astigmatism (ATA) that have symmetric bowtie topography.
Methods: This retrospective case series study included 37 patients (37 studied eyes) with WTA astigmatism and 35 patients (35 studied eyes) with ATA astigmatism. The control group consisted of 70 patients with emmetropia (70 studied eyes) with normal topography. We first tested correlations of the parameters that describe the applanation curve during ocular response analyzer measurements with the maximum keratometry values and the corneal thickness in all 3 groups. We then evaluated the significant parameters among them in search of any group differences in the biomechanical properties of the cornea. Results: Fifteen parameters correlated with Kmax reading values or corneal thickness values. The correlation coefficients (r) were low. The best correlated parameters were p1area, p2area, h1, dive1, p2area1, h11, h2, and h21. The ATA group had the highest number of parameters (n = 6) with significant differences compared with the control group. Only p2area was predictive for ATA. In contrast, the WTA group had only 1 parameter (p2area1) that was found to be significantly different compared with the control group. Conclusions: Some of the new waveform parameters can distinguish between patients with ATA and WTA and normal topography patterns and may delineate the differences in biomechanical properties between these groups that may predict the risk of corneal ectasia after laser in situ keratomileusis. Key Words: ORA parameters, ectasia risk factors, cornea biomechanical properties, astigmatism (Cornea 2014;33:721–725)
Received for publication January 26, 2014; revision received April 7, 2014; accepted April 9, 2014. Published online ahead of print June, 2014. From the *Department of Ophthalmology, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel; and †Enaim Refractive Surgery Center, Jerusalem, Israel. The authors have no funding or conflicts of interest to disclose. Reprints: Omer Trivizki, MD, Department of Ophthalmology, Tel Aviv Sourasky Medical Center, 6 Weizman St, Tel Aviv 6423906, Israel (e-mail: [email protected]). Copyright © 2014 by Lippincott Williams & Wilkins
Cornea Volume 33, Number 7, July 2014
C
orneal ectasia represents one of the rare and the most dreaded complications of laser in situ keratomileusis (LASIK).1–3 There is a general consensus not to operate on corneas with keratoconus, pellucid marginal degeneration, or topographic features suggestive of keratoconus (the so-called “forme fruste keratoconus”), although there is still no firm understanding of the cause of this condition.2,4–8 In 2008, Randleman et al7 developed the Ectasia Risk Score System, which took into account 5 parameters of risk assessment, one of which was preoperative corneal topography (CT). More than 40% of their ectasia cases had abnormal preoperative topography patterns, although 32% of those cases had exhibited normal or symmetrical bowtie topographic patterns preoperatively.7 Topography patterns that were considered by others as a risk factor for ectasia had involved against-the-rule (ATA) astigmatism.6 Another means of assessing the risk for postoperative LASIK ectasia is by measuring the biomechanical properties of the cornea. One medical device that estimates the biomechanical properties of the cornea in vivo is the ocular response analyzer (ORA). The ORA is an air-pulse tonometer that uses a bidirectional applanation process to determine the intraocular pressure (IOP). It also assesses various parameters related to biomechanical properties of the eye,8 The most well-established parameters are corneal hysteresis (CH) and corneal resistance factor (CRF).9 It has been widely reported that CH and CRF values are lower in eyes with glaucoma, Fuchs dystrophy, and keratoconus.8–13 Thirty-seven new parameters have since been added to the original 5 to quantitatively describe several aspects of the applanation waveform curve during ORA measurements. Of them, 12 demonstrated an excellent ability to distinguish between early stages of keratoconus and normal eyes.11 We tested the utility of all these additional ORA parameters for distinguishing between the biomechanical properties of emmetropic eyes that have normal topography patterns and those of eyes with moderate-to-high ATA and eyes with-therule astigmatism (WTA) that have symmetric bowtie topography patterns, given the possibility that astigmatism might be a risk factor for post-LASIK ectasia.
MATERIALS AND METHODS This is a retrospective case series study of patients with astigmatism who were recruited from our outpatient clinic. Institutional review board approval was obtained from the ethics committee of the Tel-Aviv Medical Center (protocol number 0289-13, December 2013). The control group consisted www.corneajrnl.com |
721
Cornea Volume 33, Number 7, July 2014
Trivizki et al
of patients with normal topography. Patients with a maximal simulated keratometry (Kmax) reading .47.2 diopters (D), an inferosuperior asymmetry (I-S) value .1.4 as measured with the EyeSys Vista topographer (EyeSys Vision, Inc, Houston, TX), previous corneal surgery, concurrent ocular disease, or diagnosed as having keratoconus were excluded. The ORA (Reichert Ophthalmic Instruments, Depew, NY) is an air-pulse IOP tonometer that assesses 4 basic parameters: corneal corrected IOP, Goldmann-simulated IOP, CH, and CRF (software version 1.1). CH is a measure of viscous damping in the cornea, which is caused by the amount and viscosity of glycosaminoglycans and proteoglycans as well as collagen matrix interactions. The CRF characterizes the overall resistance of the cornea to deformation, and it consists of viscous damping and elastic resistance. An infrared electro-optical system monitors the induced inward and outward movement of the cornea. Two specific pressure values are used for further calculations: pressure P1 corresponds to the first inward applanation and pressure P2 corresponds to the second outward applanation. With the introduction of new software in 2009 (version 2.0), the ORA computes 37 new parameters that describe the waveform of the ORA response curve. The first set of these additional parameters (n = 23) describes the upper 75% of the peak height in the response curve with respect to the area under the curve (p1area, p2area), the upward slope (uslope1, uslope2), the downward slope (dslope1, dslope2), width (w1, w2), height (h1, h2), and aspect ratio of the peaks (aspect1, aspect2), the path length around the peaks (path1, path2), the roughness of the peaks (aindex, bindex) and the roughness of the region between the peaks (aplhf), as well as 6 more specific parameters (dive1, dive2, mslew1, mslew2, slew1, and slew2). The manufacturer does not provide specific properties of each parameter, but rather general ones, such as the following: the area under the curve (p1area and p2area) is proportional to the time needed to change from the convex to concave form of the cornea and vice versa. Small values of p1area and p2area represent rapid change and indicate that the cornea is characterized by less damping. The ORA calculates a waveform score, which is used to select the best value for each parameter out of the 4 measured ones. We used these best values for our analysis, as recommended by the manufacturer. Each eye was analyzed separately. The statistical analysis was performed in 3 steps. To find any significant differences between the 42 ORA parameters, all studied eyes were analyzed by the nonparametric Spearman test for correlations to both Kmax values and CT values. Parameters that were found to be significantly correlated were compared between the groups using the Kruskal–Wallis 1-way analysis of variance or the Wilcoxon test when variables deviated from a normal distribution. A post hoc comparison was made with the Mann–Whitney test, and Bonferroni corrections were made when appropriate. Box plots with medians (error bars) were used for the presentation of the statistically significant parameters. Confidence coefficients were set to 0.95. Finally, a logistic regression was performed to determine the effects of the highly significant parameters on group pertinence. Because previous studies demonstrated that the prevalence of astigmatism
722
| www.corneajrnl.com
increases and that the axis turns to ATA with age,14 we also added the factor of age to our analysis. The reported P values are 2-sided. A P value ,0.05 was considered to be significant. All analyses were performed using the SPSS statistical software version 22 (IBM, Armonk, NY).
RESULTS The original study cohort had included a total of 72 astigmatic patients (72 studied eyes, 1 eye from each patient) divided into 2 groups according to their astigmatism axis of refraction: 37 eyes with WTA and 35 eyes with ATA. These 72 eyes had symmetrical bowtie topographic patterns manifesting a sphere refraction of 60.5 D and a cylinder value .3 D. The control group consisted of 70 nonastigmatic patients (70 eyes) with round or oval topography patterns manifesting refraction of a sphere range of 60.5 D and a cylinder ,1 D. The pertinent clinical and demographic characteristics of each group are presented in Table 1. There were no group differences in age, gender, eye laterality, or CT. Only 15 of the 42 ORA parameters correlated with either Kmax or CT readings (Table 2). The correlation coefficients (r) were moderately low. Overall, the properties of 8 parameters showed significant differences between the control, WTA, and ATA groups (Table 3). Three of them (CH, CRF, and p2area) were relatively highly correlated with the CT readings (values of 0.42, 0.52, and 0.33, respectively). All tested parameters correlated negatively with the Kmax and positively with the CT values. After Bonferroni correction, significant differences emerged between the ATA patients and the control group in 6 of the 8 parameters (h1, h11, p2area, p2area1, dive1, and p1area), whereas only 1 parameter differed significantly between the WTA group and the control group (p2area1). Box plots with median error bars are presented in Figure 1. We further analyzed those parameters for predictive capability. Specifically, we performed a logistic regression test to determine the likelihood of an eye to be classified under the ATA or normal group. Age was also added to this analysis. Because only 1 parameter revealed a significant difference between the control group and the WTA group, no further analysis was performed. The logistic regression model was statistically significant, x2 = 21.1, P = 0.002. The model explained 25% (Nagelkerke R2) of the variance between the groups and correctly classified 70.5% of the cases. Only 1 of the 6 predictor variables (p2area) reached a level of
TABLE 1. Demographics of the Study Groups Variable
ATA (n = 35)
WTA (n = 37)
Control (n = 70)
Age (6SD), yr 32 6 7.9 31 6 6.5 38 6 13 Males, % 73 66 80 Right eyes, % 49 54 50 Corneal thickness, mm 532 6 48 528 6 38 547 6 37 SER (mean 6 SD) 23.28 6 0.54 22.58 6 0.4 20.52 6 0.2
P 0.41 0.94 0.88 0.73
SER, spherical equivalent refraction.
Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 7, July 2014
ORA Parameters of Astigmatic Corneas
TABLE 2. Pearson Correlation of ORA Parameters to Kmax and CT Values r Kmax P1 P2 IOPg aindex bindex p1area p2area aspect1 aspect2 uslope1 uslope2 dslope1 dslope2 w1 w2 h1 h2 dive1 dive2 path1 path2 mslew1 mslew2 slew1 slew2 aplhf p1area1 p2area1 aspect11 aspect21 uslope11 uslope21 dslope11 dslope21 w11 w21 h11 h21 path11 path21 CH CRF
20.22 20.34 20.29 0.01 20.12 20.23 20.25 20.14 20.08 20.01 20.14 20.17 20.07 20.14 20.04 20.23 20.23 20.14 20.14 0.12 0.07 20.01 20.25 20.05 20.11 20.21 20.17 20.23 20.12 20.09 20.06 20.07 20.11 20.02 20.08 20.05 20.23 20.23 0.07 0.11 0.22 0.08
P 0.07 0.00 0.01 0.92 0.30 0.05 0.04 0.26 0.50 0.91 0.25 0.17 0.59 0.24 0.73 0.05 0.05 0.23 0.25 0.33 0.55 0.94 0.03 0.69 0.37 0.08 0.15 0.05 0.31 0.47 0.61 0.56 0.37 0.84 0.52 0.66 0.05 0.05 0.54 0.38 0.06 0.51
r CT 0.49 0.28 0.40 0.05 0.04 0.04 0.33 0.07 20.03 20.03 20.10 0.12 20.01 20.04 0.16 0.05 0.17 20.06 0.15 20.09 20.19 20.05 20.08 20.02 20.12 20.27 0.01 0.34 0.03 20.09 20.01 20.04 0.11 20.03 0.04 0.20 0.05 0.17 0.01 20.18 0.42 0.56
P 0.00 0.02 0.00 0.70 0.75 0.77 0.00 0.54 0.79 0.78 0.40 0.33 0.95 0.72 0.18 0.65 0.15 0.62 0.20 0.48 0.10 0.69 0.49 0.88 0.31 0.02 0.92 0.00 0.79 0.43 0.92 0.77 0.37 0.78 0.77 0.09 0.65 0.15 0.94 0.13 0.00 0.00
CT, corneal thickness.
significance. Decreasing the p2area increased the likelihood of being classified as ATA (Fig. 2). Age was not found to be a true predictor in our analysis.
DISCUSSION Post-LASIK ectasia is an unanticipated and progressive corneal steepening and thinning condition associated with increasing myopic astigmatism after LASIK.3,15 The most Ó 2014 Lippincott Williams & Wilkins
TABLE 3. Differences Between Refractive Groups (1-way ANOVA) ORA Parameter h1 h11 p2area p2area1 dive1 p1area h2 h21 mslew2 CRF P2 CH aplhf
Kruskal– Wallis Test, P
ATA Versus WTA, P
ATA Versus Control Group, P
WTA Versus Control Group, P
0.00 0.00 0.00 0.01 0.01 0.02 0.02 0.02 0.41 0.42 0.47 0.64 0.68
0.43 0.43 1.00 0.83 0.66 1.00 0.98 0.98
0.00 0.00 0.01 0.02 0.02 0.03 0.06 0.06
0.06 0.06 0.06 0.04 0.10 0.23 0.06 0.06
common finding in cases of ectasia was abnormal preoperative topography suggestive of a pre-ectatic disorder, such as keratoconus or forme fruste keratoconus.1,7,16 There have been many attempts to define and demonstrate common clinical features for the potential ectatic patient. Moreover, investigators, such as Randleman et al,7 designed scoring systems to define the relative risk for post-LASIK ectasia based on preoperative topography, residual stromal bed thickness, age, preoperative CT, and degree of myopia. The analysis of topographical data continues to be the preferred method of screening irregular astigmatism that is considered to be suggestive of forme fruste keratoconus.15 The removal of tissue from the cornea in a refractive procedure may alter its flexibility and stiffness,17 thereby reducing its strength and ability to protect the eye. It is therefore essential to uncover and understand the biomechanics of the cornea. Among the many devices that describe corneal morphology, the ORA is the first to enable exploration of the biomechanical properties of the cornea in vivo.9 All the parameters that are calculated by the ORA software represent elastic behavior of the cornea. The P1 group represents this behavior as the cornea flattens by the ORA air puff inward, whereas the P2 group represents this behavior as the cornea moves to its original position outward. In this study, we sought to determine whether there are any significant differences in the ORA parameters between 2 common types of astigmatism. These parameters can detect differences in the biomechanical properties of astigmatic eyes that call for caution in deciding whether or not to perform LASIK surgery on patients with high astigmatism. Our results showed that there are significant differences in the biomechanical properties between WTA, ATA, and normal eyes. The ATA group had the highest number of parameters (n = 6) that were significantly different from normal values, but only 1 parameter was also found to be a significant predictor in the regression analysis. Those parameters were compared with Kmax or CT based on the suggestion that those 2 measurements represented vulnerable biomechanics of the cornea.7,18 www.corneajrnl.com |
723
Trivizki et al
Cornea Volume 33, Number 7, July 2014
FIGURE 1. Box plots with median (error bar) of the 6 parameters that were significantly different between the ATA and the normal corneas.
The median values of the parameters that were significantly different between the ATA group and the control group were both negatively correlated with Kmax and relatively lower in the ATA group (Fig. 1). This finding might indicate a trend
for the ATA group to have higher Kmax values. The 2 parameters that were found to be highly correlated and significantly different between the ATA and the control groups were h1 and h11, both of which represent the peak of the signal during
FIGURE 2. Scatter plot of the p2area values for the 2 study groups and the control group.
724
| www.corneajrnl.com
Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 7, July 2014
the inward movement. This can be interpreted as indicating that ATA causes the light reflected from the applanated cornea to be more dispersed during the loading phase compared with normal healthy corneas. Mikielewicz et al11 recently demonstrated that CRF, p2area1, p2area, CH, dive1, and path2 were the 6 best parameters for showing differences between normal patients and patients with early stage keratoconus. We identified 8 parameters that were significantly different, and all of them represent the intensity of the signal in one way or another. Specifically, the h-x parameters represent the signal peak height, and the p-x-area parameters represent the sum of the signals during movement of the cornea. Only the findings of 3 of the parameters (p2area, p2area1, and dive1) were similar to those of the study by Mikielewicz et al.11 Analysis of these parameters by a regression analysis demonstrated that a lower p2area could eventually predict ATA (Fig. 2). Taken together with the findings reported by Miklielewicz et al,11 these results show that the sum of the signals during the outward movement is low in both ATA and early stage keratoconus, but this similarity is more interesting as it is relevant to the understanding of the biomechanics of the cornea. Based on the outcome of a recent analysis by Binder and Trattler5 that involved a 9700-eye database, the presence of ATA .2.0 D did not increase the risk for developing ectasia over mean follow-up periods exceeding 2 years. Because our study included only cases of very severe ATA (ie, .3.0 D), it is possible that they may have biomechanical characteristics that pose a greater risk for developing ectasia than less severe ones. Guirao19 described a thin spherical model for measuring the risk factors for keratectasia and noted that other parameters, such as Young modulus of elasticity and Poisson ratio, are related to other qualities in the cornea (eg, viscoelasticity). The resistance to bending that probably represents the interlamellar cohesive strength of the cornea and plasticity, which is the failure of the material to return to its previous stress relationship, all are relatively unexplained aspects that require further investigation.19,20 In conclusion, this study demonstrated a number of similar ORA waveform parameters between patients with early keratoconus and those with ATA. Although topography alone is not enough to predict the corneal behavior, the important question is whether these findings indicate similar biomechanical properties between the 2 groups. Although our findings are based on retrospective data of a limited study sample, as clinicians, we believe that they should be borne in mind when encountering a patient with ATA astigmatism who has been approved for refractive surgery intervention. Patients with extreme values of astigmatism may best be
Ó 2014 Lippincott Williams & Wilkins
ORA Parameters of Astigmatic Corneas
referred to a surface ablation surgery as a safety measure. Further research is needed to determine the biomechanics of corneal pathologies and provide evidence-based risk predictors of LASIK procedures. REFERENCES 1. Randleman JB, Russell B, Ward MA, et al. Risk factors and prognosis for corneal ectasia after LASIK. Ophthalmology. 2003;110:267–275. 2. Randleman JB. Post-laser in-situ keratomileusis ectasia: current understanding and future directions. Curr Opin Ophthalmol. 2006;17:406–412. 3. Amoils SP, Deist MB, Gous P, et al. Iatrogenic keratectasia after laser in situ keratomileusis for less than -4.0 to -7.0 diopters of myopia. J Cataract Refract Surg. 2000;26:967–977. 4. Binder PS. Ectasia after laser in situ keratomileusis. J Cataract Refract Surg. 2003;29:2419–2429. 5. Binder PS, Trattler WB. Evaluation of a risk factor scoring system for corneal ectasia after LASIK in eyes with normal topography. J Refract Surg. 2010;26:241–250. 6. Jabbur NS, Stark WJ, Green WR. Corneal ectasia after laser-assisted in situ keratomileusis. Arch Ophthalmol. 2001;119:1714–1716. 7. Randleman JB, Woodward M, Lynn MJ, et al. Risk assessment for ectasia after corneal refractive surgery. Ophthalmology. 2008;115:37–50. 8. Shah S, Laiquzzaman M, Bhojwani R, et al. Assessment of the biomechanical properties of the cornea with the ocular response analyzer in normal and keratoconic eyes. Invest Ophthalmol Vis Sci. 2007;48: 3026–3031. 9. Luce DA. Determining in vivo biomechanical properties of the cornea with an ocular response analyzer. J Cataract Refract Surg. 2005;31:156–162. 10. del Buey MA, Cristóbal JA, Ascaso FJ, et al. Biomechanical properties of the cornea in Fuchs’ corneal dystrophy. Invest Ophthalmol Vis Sci. 2009;50:3199–3202. 11. Mikielewicz M, Kotliar K, Barraquer RI, et al. Air-pulse corneal applanation signal curve parameters for the characterisation of keratoconus. Br J Ophthalmol. 2011;95:793–798. 12. Johnson RD, Nguyen MT, Lee N, et al. Corneal biomechanical properties in normal, forme fruste keratoconus, and manifest keratoconus after statistical correction for potentially confounding factors. Cornea. 2011;30: 516–523. 13. Saad A, Lteif Y, Azan E, et al. Biomechanical properties of keratoconus suspect eyes. Invest Ophthalmol Vis Sci. 2010;51:2912–2916. 14. Asano K, Nomura H, Iwano M, et al. Relationship between astigmatism and aging in middle-aged and elderly Japanese. Jpn J Ophthalmol. 2005; 49:127–133. 15. Seiler T, Koufala K, Richter G. Iatrogenic keratectasia after laser in situ keratomileusis. J Refract Surg. 1998;14:312–317. 16. Binder PS, Lindstrom RL, Stulting RD, et al. Keratoconus and corneal ectasia after LASIK. J Cataract Refract Surg. 2005;31:2035–2038. 17. Dawson DG, Edelhauser HF, Grossniklaus HE. Long-term histopathologic findings in human corneal wounds after refractive surgical procedures. Am J Ophthalmol. 2005;139:168–178. 18. Greenstein SA, Hersh PS. Characteristics influencing outcomes of corneal collagen crosslinking for keratoconus and ectasia: implications for patient selection. J Cataract Refract Surg. 2013;39:1133–1140. 19. Guirao A. Theoretical elastic response of the cornea to refractive surgery: risk factors for keratectasia. J Refract Surg. 2005;21:176–185. 20. Dupps WJ Jr. Biomechanical modeling of corneal ectasia. J Refract Surg. 2005;21:186–190.
www.corneajrnl.com |
725
Air-pulse Corneal Applanation Signal Curve Parameters for Characterization of Astigmatic Corneas Omer Trivizki, MD,* Jonathan Shahar, MD,* Samuel Levinger, MD,† and Eliya Levinger, MD*†
Purpose: The aim of this study was to test the 42 parameters of the ocular response analyzer for distinguishing between the biomechanical properties of emmetropic eyes with normal topography and eyes with moderate-to-high with-the-rule astigmatism (WTA) and against-the-rule astigmatism (ATA) that have symmetric bowtie topography.
Methods: This retrospective case series study included 37 patients (37 studied eyes) with WTA astigmatism and 35 patients (35 studied eyes) with ATA astigmatism. The control group consisted of 70 patients with emmetropia (70 studied eyes) with normal topography. We first tested correlations of the parameters that describe the applanation curve during ocular response analyzer measurements with the maximum keratometry values and the corneal thickness in all 3 groups. We then evaluated the significant parameters among them in search of any group differences in the biomechanical properties of the cornea. Results: Fifteen parameters correlated with Kmax reading values or corneal thickness values. The correlation coefficients (r) were low. The best correlated parameters were p1area, p2area, h1, dive1, p2area1, h11, h2, and h21. The ATA group had the highest number of parameters (n = 6) with significant differences compared with the control group. Only p2area was predictive for ATA. In contrast, the WTA group had only 1 parameter (p2area1) that was found to be significantly different compared with the control group. Conclusions: Some of the new waveform parameters can distinguish between patients with ATA and WTA and normal topography patterns and may delineate the differences in biomechanical properties between these groups that may predict the risk of corneal ectasia after laser in situ keratomileusis. Key Words: ORA parameters, ectasia risk factors, cornea biomechanical properties, astigmatism (Cornea 2014;33:721–725)
Received for publication January 26, 2014; revision received April 7, 2014; accepted April 9, 2014. Published online ahead of print June, 2014. From the *Department of Ophthalmology, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel; and †Enaim Refractive Surgery Center, Jerusalem, Israel. The authors have no funding or conflicts of interest to disclose. Reprints: Omer Trivizki, MD, Department of Ophthalmology, Tel Aviv Sourasky Medical Center, 6 Weizman St, Tel Aviv 6423906, Israel (e-mail: [email protected]). Copyright © 2014 by Lippincott Williams & Wilkins
Cornea Volume 33, Number 7, July 2014
C
orneal ectasia represents one of the rare and the most dreaded complications of laser in situ keratomileusis (LASIK).1–3 There is a general consensus not to operate on corneas with keratoconus, pellucid marginal degeneration, or topographic features suggestive of keratoconus (the so-called “forme fruste keratoconus”), although there is still no firm understanding of the cause of this condition.2,4–8 In 2008, Randleman et al7 developed the Ectasia Risk Score System, which took into account 5 parameters of risk assessment, one of which was preoperative corneal topography (CT). More than 40% of their ectasia cases had abnormal preoperative topography patterns, although 32% of those cases had exhibited normal or symmetrical bowtie topographic patterns preoperatively.7 Topography patterns that were considered by others as a risk factor for ectasia had involved against-the-rule (ATA) astigmatism.6 Another means of assessing the risk for postoperative LASIK ectasia is by measuring the biomechanical properties of the cornea. One medical device that estimates the biomechanical properties of the cornea in vivo is the ocular response analyzer (ORA). The ORA is an air-pulse tonometer that uses a bidirectional applanation process to determine the intraocular pressure (IOP). It also assesses various parameters related to biomechanical properties of the eye,8 The most well-established parameters are corneal hysteresis (CH) and corneal resistance factor (CRF).9 It has been widely reported that CH and CRF values are lower in eyes with glaucoma, Fuchs dystrophy, and keratoconus.8–13 Thirty-seven new parameters have since been added to the original 5 to quantitatively describe several aspects of the applanation waveform curve during ORA measurements. Of them, 12 demonstrated an excellent ability to distinguish between early stages of keratoconus and normal eyes.11 We tested the utility of all these additional ORA parameters for distinguishing between the biomechanical properties of emmetropic eyes that have normal topography patterns and those of eyes with moderate-to-high ATA and eyes with-therule astigmatism (WTA) that have symmetric bowtie topography patterns, given the possibility that astigmatism might be a risk factor for post-LASIK ectasia.
MATERIALS AND METHODS This is a retrospective case series study of patients with astigmatism who were recruited from our outpatient clinic. Institutional review board approval was obtained from the ethics committee of the Tel-Aviv Medical Center (protocol number 0289-13, December 2013). The control group consisted www.corneajrnl.com |
721
Cornea Volume 33, Number 7, July 2014
Trivizki et al
of patients with normal topography. Patients with a maximal simulated keratometry (Kmax) reading .47.2 diopters (D), an inferosuperior asymmetry (I-S) value .1.4 as measured with the EyeSys Vista topographer (EyeSys Vision, Inc, Houston, TX), previous corneal surgery, concurrent ocular disease, or diagnosed as having keratoconus were excluded. The ORA (Reichert Ophthalmic Instruments, Depew, NY) is an air-pulse IOP tonometer that assesses 4 basic parameters: corneal corrected IOP, Goldmann-simulated IOP, CH, and CRF (software version 1.1). CH is a measure of viscous damping in the cornea, which is caused by the amount and viscosity of glycosaminoglycans and proteoglycans as well as collagen matrix interactions. The CRF characterizes the overall resistance of the cornea to deformation, and it consists of viscous damping and elastic resistance. An infrared electro-optical system monitors the induced inward and outward movement of the cornea. Two specific pressure values are used for further calculations: pressure P1 corresponds to the first inward applanation and pressure P2 corresponds to the second outward applanation. With the introduction of new software in 2009 (version 2.0), the ORA computes 37 new parameters that describe the waveform of the ORA response curve. The first set of these additional parameters (n = 23) describes the upper 75% of the peak height in the response curve with respect to the area under the curve (p1area, p2area), the upward slope (uslope1, uslope2), the downward slope (dslope1, dslope2), width (w1, w2), height (h1, h2), and aspect ratio of the peaks (aspect1, aspect2), the path length around the peaks (path1, path2), the roughness of the peaks (aindex, bindex) and the roughness of the region between the peaks (aplhf), as well as 6 more specific parameters (dive1, dive2, mslew1, mslew2, slew1, and slew2). The manufacturer does not provide specific properties of each parameter, but rather general ones, such as the following: the area under the curve (p1area and p2area) is proportional to the time needed to change from the convex to concave form of the cornea and vice versa. Small values of p1area and p2area represent rapid change and indicate that the cornea is characterized by less damping. The ORA calculates a waveform score, which is used to select the best value for each parameter out of the 4 measured ones. We used these best values for our analysis, as recommended by the manufacturer. Each eye was analyzed separately. The statistical analysis was performed in 3 steps. To find any significant differences between the 42 ORA parameters, all studied eyes were analyzed by the nonparametric Spearman test for correlations to both Kmax values and CT values. Parameters that were found to be significantly correlated were compared between the groups using the Kruskal–Wallis 1-way analysis of variance or the Wilcoxon test when variables deviated from a normal distribution. A post hoc comparison was made with the Mann–Whitney test, and Bonferroni corrections were made when appropriate. Box plots with medians (error bars) were used for the presentation of the statistically significant parameters. Confidence coefficients were set to 0.95. Finally, a logistic regression was performed to determine the effects of the highly significant parameters on group pertinence. Because previous studies demonstrated that the prevalence of astigmatism
722
| www.corneajrnl.com
increases and that the axis turns to ATA with age,14 we also added the factor of age to our analysis. The reported P values are 2-sided. A P value ,0.05 was considered to be significant. All analyses were performed using the SPSS statistical software version 22 (IBM, Armonk, NY).
RESULTS The original study cohort had included a total of 72 astigmatic patients (72 studied eyes, 1 eye from each patient) divided into 2 groups according to their astigmatism axis of refraction: 37 eyes with WTA and 35 eyes with ATA. These 72 eyes had symmetrical bowtie topographic patterns manifesting a sphere refraction of 60.5 D and a cylinder value .3 D. The control group consisted of 70 nonastigmatic patients (70 eyes) with round or oval topography patterns manifesting refraction of a sphere range of 60.5 D and a cylinder ,1 D. The pertinent clinical and demographic characteristics of each group are presented in Table 1. There were no group differences in age, gender, eye laterality, or CT. Only 15 of the 42 ORA parameters correlated with either Kmax or CT readings (Table 2). The correlation coefficients (r) were moderately low. Overall, the properties of 8 parameters showed significant differences between the control, WTA, and ATA groups (Table 3). Three of them (CH, CRF, and p2area) were relatively highly correlated with the CT readings (values of 0.42, 0.52, and 0.33, respectively). All tested parameters correlated negatively with the Kmax and positively with the CT values. After Bonferroni correction, significant differences emerged between the ATA patients and the control group in 6 of the 8 parameters (h1, h11, p2area, p2area1, dive1, and p1area), whereas only 1 parameter differed significantly between the WTA group and the control group (p2area1). Box plots with median error bars are presented in Figure 1. We further analyzed those parameters for predictive capability. Specifically, we performed a logistic regression test to determine the likelihood of an eye to be classified under the ATA or normal group. Age was also added to this analysis. Because only 1 parameter revealed a significant difference between the control group and the WTA group, no further analysis was performed. The logistic regression model was statistically significant, x2 = 21.1, P = 0.002. The model explained 25% (Nagelkerke R2) of the variance between the groups and correctly classified 70.5% of the cases. Only 1 of the 6 predictor variables (p2area) reached a level of
TABLE 1. Demographics of the Study Groups Variable
ATA (n = 35)
WTA (n = 37)
Control (n = 70)
Age (6SD), yr 32 6 7.9 31 6 6.5 38 6 13 Males, % 73 66 80 Right eyes, % 49 54 50 Corneal thickness, mm 532 6 48 528 6 38 547 6 37 SER (mean 6 SD) 23.28 6 0.54 22.58 6 0.4 20.52 6 0.2
P 0.41 0.94 0.88 0.73
SER, spherical equivalent refraction.
Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 7, July 2014
ORA Parameters of Astigmatic Corneas
TABLE 2. Pearson Correlation of ORA Parameters to Kmax and CT Values r Kmax P1 P2 IOPg aindex bindex p1area p2area aspect1 aspect2 uslope1 uslope2 dslope1 dslope2 w1 w2 h1 h2 dive1 dive2 path1 path2 mslew1 mslew2 slew1 slew2 aplhf p1area1 p2area1 aspect11 aspect21 uslope11 uslope21 dslope11 dslope21 w11 w21 h11 h21 path11 path21 CH CRF
20.22 20.34 20.29 0.01 20.12 20.23 20.25 20.14 20.08 20.01 20.14 20.17 20.07 20.14 20.04 20.23 20.23 20.14 20.14 0.12 0.07 20.01 20.25 20.05 20.11 20.21 20.17 20.23 20.12 20.09 20.06 20.07 20.11 20.02 20.08 20.05 20.23 20.23 0.07 0.11 0.22 0.08
P 0.07 0.00 0.01 0.92 0.30 0.05 0.04 0.26 0.50 0.91 0.25 0.17 0.59 0.24 0.73 0.05 0.05 0.23 0.25 0.33 0.55 0.94 0.03 0.69 0.37 0.08 0.15 0.05 0.31 0.47 0.61 0.56 0.37 0.84 0.52 0.66 0.05 0.05 0.54 0.38 0.06 0.51
r CT 0.49 0.28 0.40 0.05 0.04 0.04 0.33 0.07 20.03 20.03 20.10 0.12 20.01 20.04 0.16 0.05 0.17 20.06 0.15 20.09 20.19 20.05 20.08 20.02 20.12 20.27 0.01 0.34 0.03 20.09 20.01 20.04 0.11 20.03 0.04 0.20 0.05 0.17 0.01 20.18 0.42 0.56
P 0.00 0.02 0.00 0.70 0.75 0.77 0.00 0.54 0.79 0.78 0.40 0.33 0.95 0.72 0.18 0.65 0.15 0.62 0.20 0.48 0.10 0.69 0.49 0.88 0.31 0.02 0.92 0.00 0.79 0.43 0.92 0.77 0.37 0.78 0.77 0.09 0.65 0.15 0.94 0.13 0.00 0.00
CT, corneal thickness.
significance. Decreasing the p2area increased the likelihood of being classified as ATA (Fig. 2). Age was not found to be a true predictor in our analysis.
DISCUSSION Post-LASIK ectasia is an unanticipated and progressive corneal steepening and thinning condition associated with increasing myopic astigmatism after LASIK.3,15 The most Ó 2014 Lippincott Williams & Wilkins
TABLE 3. Differences Between Refractive Groups (1-way ANOVA) ORA Parameter h1 h11 p2area p2area1 dive1 p1area h2 h21 mslew2 CRF P2 CH aplhf
Kruskal– Wallis Test, P
ATA Versus WTA, P
ATA Versus Control Group, P
WTA Versus Control Group, P
0.00 0.00 0.00 0.01 0.01 0.02 0.02 0.02 0.41 0.42 0.47 0.64 0.68
0.43 0.43 1.00 0.83 0.66 1.00 0.98 0.98
0.00 0.00 0.01 0.02 0.02 0.03 0.06 0.06
0.06 0.06 0.06 0.04 0.10 0.23 0.06 0.06
common finding in cases of ectasia was abnormal preoperative topography suggestive of a pre-ectatic disorder, such as keratoconus or forme fruste keratoconus.1,7,16 There have been many attempts to define and demonstrate common clinical features for the potential ectatic patient. Moreover, investigators, such as Randleman et al,7 designed scoring systems to define the relative risk for post-LASIK ectasia based on preoperative topography, residual stromal bed thickness, age, preoperative CT, and degree of myopia. The analysis of topographical data continues to be the preferred method of screening irregular astigmatism that is considered to be suggestive of forme fruste keratoconus.15 The removal of tissue from the cornea in a refractive procedure may alter its flexibility and stiffness,17 thereby reducing its strength and ability to protect the eye. It is therefore essential to uncover and understand the biomechanics of the cornea. Among the many devices that describe corneal morphology, the ORA is the first to enable exploration of the biomechanical properties of the cornea in vivo.9 All the parameters that are calculated by the ORA software represent elastic behavior of the cornea. The P1 group represents this behavior as the cornea flattens by the ORA air puff inward, whereas the P2 group represents this behavior as the cornea moves to its original position outward. In this study, we sought to determine whether there are any significant differences in the ORA parameters between 2 common types of astigmatism. These parameters can detect differences in the biomechanical properties of astigmatic eyes that call for caution in deciding whether or not to perform LASIK surgery on patients with high astigmatism. Our results showed that there are significant differences in the biomechanical properties between WTA, ATA, and normal eyes. The ATA group had the highest number of parameters (n = 6) that were significantly different from normal values, but only 1 parameter was also found to be a significant predictor in the regression analysis. Those parameters were compared with Kmax or CT based on the suggestion that those 2 measurements represented vulnerable biomechanics of the cornea.7,18 www.corneajrnl.com |
723
Trivizki et al
Cornea Volume 33, Number 7, July 2014
FIGURE 1. Box plots with median (error bar) of the 6 parameters that were significantly different between the ATA and the normal corneas.
The median values of the parameters that were significantly different between the ATA group and the control group were both negatively correlated with Kmax and relatively lower in the ATA group (Fig. 1). This finding might indicate a trend
for the ATA group to have higher Kmax values. The 2 parameters that were found to be highly correlated and significantly different between the ATA and the control groups were h1 and h11, both of which represent the peak of the signal during
FIGURE 2. Scatter plot of the p2area values for the 2 study groups and the control group.
724
| www.corneajrnl.com
Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 7, July 2014
the inward movement. This can be interpreted as indicating that ATA causes the light reflected from the applanated cornea to be more dispersed during the loading phase compared with normal healthy corneas. Mikielewicz et al11 recently demonstrated that CRF, p2area1, p2area, CH, dive1, and path2 were the 6 best parameters for showing differences between normal patients and patients with early stage keratoconus. We identified 8 parameters that were significantly different, and all of them represent the intensity of the signal in one way or another. Specifically, the h-x parameters represent the signal peak height, and the p-x-area parameters represent the sum of the signals during movement of the cornea. Only the findings of 3 of the parameters (p2area, p2area1, and dive1) were similar to those of the study by Mikielewicz et al.11 Analysis of these parameters by a regression analysis demonstrated that a lower p2area could eventually predict ATA (Fig. 2). Taken together with the findings reported by Miklielewicz et al,11 these results show that the sum of the signals during the outward movement is low in both ATA and early stage keratoconus, but this similarity is more interesting as it is relevant to the understanding of the biomechanics of the cornea. Based on the outcome of a recent analysis by Binder and Trattler5 that involved a 9700-eye database, the presence of ATA .2.0 D did not increase the risk for developing ectasia over mean follow-up periods exceeding 2 years. Because our study included only cases of very severe ATA (ie, .3.0 D), it is possible that they may have biomechanical characteristics that pose a greater risk for developing ectasia than less severe ones. Guirao19 described a thin spherical model for measuring the risk factors for keratectasia and noted that other parameters, such as Young modulus of elasticity and Poisson ratio, are related to other qualities in the cornea (eg, viscoelasticity). The resistance to bending that probably represents the interlamellar cohesive strength of the cornea and plasticity, which is the failure of the material to return to its previous stress relationship, all are relatively unexplained aspects that require further investigation.19,20 In conclusion, this study demonstrated a number of similar ORA waveform parameters between patients with early keratoconus and those with ATA. Although topography alone is not enough to predict the corneal behavior, the important question is whether these findings indicate similar biomechanical properties between the 2 groups. Although our findings are based on retrospective data of a limited study sample, as clinicians, we believe that they should be borne in mind when encountering a patient with ATA astigmatism who has been approved for refractive surgery intervention. Patients with extreme values of astigmatism may best be
Ó 2014 Lippincott Williams & Wilkins
ORA Parameters of Astigmatic Corneas
referred to a surface ablation surgery as a safety measure. Further research is needed to determine the biomechanics of corneal pathologies and provide evidence-based risk predictors of LASIK procedures. REFERENCES 1. Randleman JB, Russell B, Ward MA, et al. Risk factors and prognosis for corneal ectasia after LASIK. Ophthalmology. 2003;110:267–275. 2. Randleman JB. Post-laser in-situ keratomileusis ectasia: current understanding and future directions. Curr Opin Ophthalmol. 2006;17:406–412. 3. Amoils SP, Deist MB, Gous P, et al. Iatrogenic keratectasia after laser in situ keratomileusis for less than -4.0 to -7.0 diopters of myopia. J Cataract Refract Surg. 2000;26:967–977. 4. Binder PS. Ectasia after laser in situ keratomileusis. J Cataract Refract Surg. 2003;29:2419–2429. 5. Binder PS, Trattler WB. Evaluation of a risk factor scoring system for corneal ectasia after LASIK in eyes with normal topography. J Refract Surg. 2010;26:241–250. 6. Jabbur NS, Stark WJ, Green WR. Corneal ectasia after laser-assisted in situ keratomileusis. Arch Ophthalmol. 2001;119:1714–1716. 7. Randleman JB, Woodward M, Lynn MJ, et al. Risk assessment for ectasia after corneal refractive surgery. Ophthalmology. 2008;115:37–50. 8. Shah S, Laiquzzaman M, Bhojwani R, et al. Assessment of the biomechanical properties of the cornea with the ocular response analyzer in normal and keratoconic eyes. Invest Ophthalmol Vis Sci. 2007;48: 3026–3031. 9. Luce DA. Determining in vivo biomechanical properties of the cornea with an ocular response analyzer. J Cataract Refract Surg. 2005;31:156–162. 10. del Buey MA, Cristóbal JA, Ascaso FJ, et al. Biomechanical properties of the cornea in Fuchs’ corneal dystrophy. Invest Ophthalmol Vis Sci. 2009;50:3199–3202. 11. Mikielewicz M, Kotliar K, Barraquer RI, et al. Air-pulse corneal applanation signal curve parameters for the characterisation of keratoconus. Br J Ophthalmol. 2011;95:793–798. 12. Johnson RD, Nguyen MT, Lee N, et al. Corneal biomechanical properties in normal, forme fruste keratoconus, and manifest keratoconus after statistical correction for potentially confounding factors. Cornea. 2011;30: 516–523. 13. Saad A, Lteif Y, Azan E, et al. Biomechanical properties of keratoconus suspect eyes. Invest Ophthalmol Vis Sci. 2010;51:2912–2916. 14. Asano K, Nomura H, Iwano M, et al. Relationship between astigmatism and aging in middle-aged and elderly Japanese. Jpn J Ophthalmol. 2005; 49:127–133. 15. Seiler T, Koufala K, Richter G. Iatrogenic keratectasia after laser in situ keratomileusis. J Refract Surg. 1998;14:312–317. 16. Binder PS, Lindstrom RL, Stulting RD, et al. Keratoconus and corneal ectasia after LASIK. J Cataract Refract Surg. 2005;31:2035–2038. 17. Dawson DG, Edelhauser HF, Grossniklaus HE. Long-term histopathologic findings in human corneal wounds after refractive surgical procedures. Am J Ophthalmol. 2005;139:168–178. 18. Greenstein SA, Hersh PS. Characteristics influencing outcomes of corneal collagen crosslinking for keratoconus and ectasia: implications for patient selection. J Cataract Refract Surg. 2013;39:1133–1140. 19. Guirao A. Theoretical elastic response of the cornea to refractive surgery: risk factors for keratectasia. J Refract Surg. 2005;21:176–185. 20. Dupps WJ Jr. Biomechanical modeling of corneal ectasia. J Refract Surg. 2005;21:186–190.
www.corneajrnl.com |
725
