How Abnormal Is the Noncorneal Biometry of Keratoconic Eyes?

BASIC INVESTIGATION

Jos J. Rozema, PhD,*† Nadia Zakaria, MD, PhD,*† Irene Ruiz Hidalgo, MSc,*† Sien Jongenelen, MD,* Marie-José Tassignon, MD, PhD,*† and Carina Koppen, MD, PhD*†

here seems to be a general consensus in the literature that Purpose: To determine whether the noncorneal biometry in Tkeratoconus is associated with an increased ocular axial keratoconic eyes deviates from that in healthy eyes. length. This idea was first proposed in the 1970s to partially Methods: The right eyes of 200 healthy subjects and 76 patients explain the excessive degrees of myopia often seen after penetrating keratoplasty (PK), alongside surgery-related is- with keratoconus were measured with an autorefractometer, 1,2 a Scheimpflug tomographer, and an optical biometer. The analysis sues such as graft size and method of suturing. The decades consisted of a general linear model (GLM), correcting for age and that followed saw a great many authors continue this work, gender effects, comparing keratoconic eyes with healthy eyes, and adding the difference in trephination size between donor and fl emmetropic eyes. Next, 20 patients with mild keratoconus in one eye recipient as a major in uence on postoperative myopia (see and moderate keratoconus in the fellow eye were analyzed to assess Table, Supplemental Digital Content 1, http://links.lww.com/ the effect of disease stage. ICO/A397), and suggesting trephination size as a potential method to control it.3 Often these reports reiterated the idea of Results: As expected the keratoconic group differed significantly axial elongation, supported by other authors who found from both the healthy and the emmetropic groups for most refractive, keratoconic eyes were significantly longer compared with corneal, and anterior chamber parameters (GLM, P , 0.001). For the a healthy control group.4–8 Meanwhile, 2 other studies vitreous depth V and the axial length L, the keratoconic group reported finding no such difference.9,10 showed significantly larger values than the emmetropic group It is interesting to note, however, that all studies (V: +0.68 6 0.19 mm, P = 0.001; L: +0.63 6 0.14 mm, P , reporting a significant axial elongation in keratoconus used 0.001), but not larger than the healthy group (V: +0.18 6 0.20 mm, patients with emmetropic eyes as the control group,4–8 while P = 0.364; L: +0.09 6 0.14 mm, P = 0.519). Besides the corneal the only study that did not impose refractive criteria on their parameters, the stage of the disease led to a significantly deeper control group reported no significant difference.10 Although aqueous depth (+0.07 6 0.02 mm; Wilcoxon signed rank, P = the choice for an emmetropic control group is understandable, 0.002), but no significant difference in axial length (P = 0.940). it may lead to a bias in the distribution of the axial length as these group tends to have shorter eyes than the general Conclusions: Current analysis does not confirm the previously suggested association between keratoconus and higher amounts of population, which sees an ever-increasing prevalence of axial myopia, which may have been the result of selection bias by myopia. Moreover, Li et al observed that 77.2% of patients using emmetropic eyes as a healthy control group. with keratoconus were already myopic before the age of 15 years,11 the typical age for early-onset keratoconus. Emme- Key Words: keratoconus, biometry, axial length tropic eyes are therefore unlikely to be a suitable reference for comparison with those with keratoconus, which suggests that (Cornea 2016;35:860–865) previously observed differences were potentially based on an artifact. It is therefore the aim of this article to repeat the prior statistical analyses of the corneal and noncorneal biometry in Received for publication December 29, 2015; revision received January 8, keratoconic eyes and compare these values to those of healthy 2016; accepted January 12, 2016. Published online ahead of print March eyes with a wide range of refractions, as well as to a subgroup 18, 2016. From the *Department of Ophthalmology, Antwerp University Hospital, of patients with emmetropic eyes. Furthermore, the biometry Edegem, Belgium; and †Department of Medicine and Health Science, of eyes with mild keratoconus are compared to those of their University of Antwerp, Wilrijk, Belgium. companion eyes with moderate keratoconus in order to Supplemental digital content is available for this article. Direct URL citations establish whether keratoconus stage affects noncorneal biom- appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.corneajrnl.com). etry. Finally, since the idea of noncorneal biometry alterations Supported by research grants from the Flemish government agency for in keratoconus originated from observing excessive amounts Innovation by Science and Technology (Grant IWT/110684). of myopia after a corneal transplant, we also simulated what The authors have no conflicts of interest to disclose. kind of refractive outcome one could theoretically expect Reprints: Jos J. Rozema, PhD, Department of Ophthalmology, Antwerp University Hospital, Wilrijkstraat 10, 2650 Edegem, Belgium (e-mail: jos. in such a situation. These results may serve to readjust [email protected]). the clinical expectations on refractive outcomes after Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. a corneal transplant.

860 | www.corneajrnl.com Cornea Volume 35, Number 6, June 2016

Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Cornea Volume 35, Number 6, June 2016 The Noncorneal Biometry of Keratoconic Eyes

METHODS R2011b (The MathWorks, MA). These calculations were performed for the case in which no postoperative increase in Subjects corneal curvature had taken place, as well as for a modest This work presents the biometry of 200 healthy postoperative increase of 1D. volunteers (100 men, 100 women, aged 36.1 6 9.0 years, 96.9% Caucasians), recruited from the employees of our hospital and inhabitants of the surrounding area, and 76 Statistics patients with keratoconus (39 men, 37 women, aged 32.7 6 Statistical calculations were performed using Excel 8.2 years, 96.1% Caucasians), recruited from the keratoconus 2010 (Microsoft Corp, WA) and SPSS 23 (SPSS Inc., consultation at our department. The diagnosis of keratoconus Chicago). The comparison between groups was performed was based on a combination of typical clinical signs on using a general linear model, correcting for the influence of a Scheimpflug topography (eg, asymmetric bow tie pattern, subject age and gender. Significance levels of P , 0.05 were conical protrusion near the corneal apex, skewed axes), as used, with the application of a Bonferroni correction in case well as findings seen with a slit lamp examination (eg, of multiple comparisons to avoid a inflation. Fleischer ring, Vogt striae). The study was approved by the Ethical Committee of the Antwerp University Hospital and all subjects gave written informed consent before testing, in RESULTS accordance with the Declaration of Helsinki. Comparison of Biometry In the analysis these data are divided over 3 groups: As one would expect, the differences between the group patients with keratoconus (KTC), healthy subjects (Healthy), with keratoconus and the 2 other groups lie mostly in the and a subgroup of the latter with those with emmetropic eyes refractive and corneal parameters, with significant differences having a spherical equivalent (SE) refraction between 60.75D seen for SE, J , K , J , and CCT (Table 1). Keratoconic (Emm). In all subjects both eyes were measured, but only right 45 m 45,c eyes also had significantly deeper anterior chambers than the eye data were used for the intergroup comparisons. 2 other groups, but no significant differences were seen for either lens thickness T or lens power PL. Although for the refractive and anterior segment Biometry and Calculations parameters the comparisons between the keratoconic group In all subjects, the uncyclopleged refraction was and either healthy group yielded very similar results, a very determined using an autorefractometer (ARK-700; Nidek, different outcome was seen when the axial length L and Gamagori, Japan), which was used to calculate the SE and the 12 vitreous depth V were considered. For both parameters, the Jackson cylinders J0 and J45. Next, the keratometry was fi fl keratoconic group was found to have signi cantly higher measured with a Scheimp ug tomographer (Pentacam HR; values compared with emmetropic eyes, but not compared Oculus GmbH, Wetzlar, Germany), from which the mean with the entire group of healthy eyes (Table 1). The presence anterior keratometry Km and the anterior corneal Jackson of keratoconus therefore does not seem to lead to cylinders J0,c and J45,c were calculated. This device also a deeper vitreous. provided measures of the central corneal thickness (CCT) and Furthermore, a correlation analysis shows that 59.6% of the aqueous depth (AD) (ie, distance between the posterior the SE refraction in healthy eyes is determined by the axial corneal vertex and the anterior lenticular vertex). Axial length 2 length L (r , P , 0.001), and only 4.1% by keratometry Km L was determined using an optical biometer (IOL Master 500; (P = 0.004; Fig. 1). In keratoconic eyes, however, these Carl Zeiss, Jena, Germany/Lenstar; Haag-Streit, Koeniz, influences are reversed, with 69.2% (P , 0.001) of the Switzerland). Although axial length was determined using 2 refraction determined by corneal power and 4.5% by axial different biometric devices, we do not anticipate this to length (P = 0.070). influence the results based on a recent meta-analysis on the compatibility of biometers for measuring axial length.13 For the subjects in whom Lenstar was used, measurements of lens Influence of Keratoconus Stage thickness T were also available, which permitted calculating Another comparison was done between both eyes vitreous depth V as V=L2 T 2 AD – CCT in these subjects, of subjects that have one eye with mild keratoconus (defined as well as the crystalline lens power PL using the Bennett as Km , 45D and corneal astigmatism Ast , –1.5D), and one fi 14,15 formula with adjusted coef cients. eye with moderate keratoconus (Km $ 45D or Ast $ –1.5D). Here a significant difference in AD was also seen, but not in vitreous depth or axial length (Table 2). Hence axial length Simulated Corneal Transplantation and vitreous depth do not appear to be associated with The biometry data of the healthy cohort may also be keratoconus stage. used to determine the refraction one may theoretically expect after corneal transplantation. Since donor corneas are usually not matched by their curvature to the axial length of the Simulation of Refraction After acceptor, this clinical situation can be simulated by randomly Corneal Transplantation permuting the corneas of the healthy cohort and calculating The algorithm that randomly permutes the mean the resulting refraction by means of matrix optics16 in Matlab corneal curvature was applied to the data of the 128 healthy

Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Rozema et al Cornea Volume 35, Number 6, June 2016

TABLE 1. Overview of the Average and Standard Deviation Values for the 3 Groups and the Differences Between Them, Estimated Using a General Linear Model KTC 2 Healthy* KTC 2 Emm* Parameter KTC, Mean 6 SD Healthy, Mean 6 SD Emm, Mean 6 SD Mean 6 SEM P† Mean 6 SEM P† Age (y) 32.7 6 8.2 36.1 6 9.0 38.2 6 9.0 ———— SE (D)‡ 25.2 6 4.8 21.4 6 2.3 20.1 6 0.4 23.7 6 0.4 ,0.001 25.2 6 0.5 ,0.001

J0 (D)‡ 20.1 6 1.4 0.02 6 0.4 0.04 6 0.3 20.1 6 0.1 0.420 20.1 6 0.2 0.448

J45 (D)‡ 1.0 6 1.2 0.01 6 0.2 0.03 6 0.1 1.0 6 0.1 ,0.001 0.9 6 0.1 ,0.001

Km (D)‡ 46.6 6 4.2 43.2 6 1.2 43.0 6 1.1 3.4 6 0.3 ,0.001 3.6 6 0.5 ,0.001

J0,c (D)‡ 20.5 6 1.2 20.3 6 0.3 20.3 6 0.3 20.2 6 0.1 0.024 20.2 6 0.1 0.190

J45,c (D)‡ 20.4 6 1.1 20.1 6 0.2 20.1 6 0.1 20.4 6 0.1 ,0.001 20.3 6 0.1 0.009 CCT (mm)‡ 501.4 6 38.7 547.1 6 31.8 543.5 6 31.4 244.8 6 4.6 ,0.001 239.0 6 5.8 ,0.001 AD (mm)‡ 3.19 6 0.36 2.94 6 0.39 2.84 6 0.31 0.18 6 0.04 ,0.001 0.27 6 0.05 ,0.001 T (mm)§ 3.9 6 0.3 4.0 6 0.4 4.1 6 0.3 0.01 6 0.1 0.869 20.1 6 0.1 0.070 V (mm)§ 16.6 6 1.2 16.4 6 1.1 15.9 6 0.6 0.2 6 0.2 0.364 0.7 6 0.2 0.001 L (mm)‡ 24.1 6 1.0 23.9 6 1.1 23.5 6 0.7 0.1 6 0.1 0.519 0.6 6 0.1 ,0.001

PL (D)§ 23.37 6 2.1 23.6 6 2.1 23.8 6 2.1 20.4 6 0.4 0.269 20.7 6 0.4 0.093

*General linear model, correcting for age and gender. †t test, P , 0.05/12 = 0.004 (Bonferroni correction) indicates a significant difference (boldface). ‡N = 74 (KTC), 200 (Healthy) and 86 (Emm). §N = 44 (KTC), 128 (Healthy) and 51 (Emm). Emm, emmetropic eyes; J0, J45, Jackson cylinders of the refraction; J0,c, J45,c, Jackson cylinders of the corneal astigmatism; Km, mean keratometry; KTC, keratoconus; L, axial length; PL, lens power; SEM, standard error of the mean; T, lens thickness; V, vitreous depth. eyes for which all required biometric parameters were avail- the localized corneal protrusion results in progressive increase able, thus simulating one random combination of donor and of corneal curvature and astigmatism, as well as thinning of recipient biometry for each eye. Based on these data, the the cornea.17 This is consistent with a model in which the corresponding refraction was calculated, which was then used weakened biomechanical structure permits the cornea to to estimate the mean theoretical distribution of the refraction gradually expand under the pressure of the aqueous humor.18 after corneal transplantation. This procedure was subse- This expansion also has repercussions for the anterior quently repeated 100 times to estimate the variations induced chamber, which tends to be significantly deeper in patients by these random combinations. with keratoconus,6–8,11,19–21 as was also seen in Table 1. The As a way of verifying the refractive calculations, first influence on anterior chamber depth was also found to the refractive distribution calculated from the original (non- increase progressively with the degree of keratoconus (Table permuted) biometry data was compared with that of the 2), which again confirms the outward expansion of the cornea clinically measured refraction and was found to correspond with disease progression.11,20 almost perfectly (Fig. 2). Next the eyes with the permuted The influence of keratoconus on the deeper structures of corneas were analyzed, assuming no postoperative increase in the eye is far less clear, however. No significant effect of mean keratometry had taken place. Here the mean calculated keratoconus on crystalline lens power PL or lens thickness T SE refraction did not differ significantly from the original data could be seen, which is in agreement with the literature,6,7 but (21.34 6 2.76D and 21.30 6 2.50D, respectively; t test P = no associations with axial length L could be found in these 0.651), but the distribution of the permuted data was flatter data either, which contradicts almost all earlier reports.4–8 and broader than that of the original data. Finally, assuming Similarly, no significant difference could be seen for the a postoperative increase in mean keratometry of 1D led to vitreous depth V, again in opposition to the literature.5–7 a corresponding myopic shift to 22.38 6 2.83D, which was Vitreous depth was also not influenced by the degree of significantly different from the original data (t test, P keratoconus (Table 2), as was also reported by Tuft et al.5 = 0.011). The sole reason for this disagreement lies in the choice This demonstrates that postoperative myopia is to be of the reference group in the articles reporting axial length expected after corneal transplantation from a theoretical point differences, all of which used emmetropic eyes as the healthy of view and is not necessarily associated with an axial control. Meanwhile, the only prior study that also included elongation due to keratoconus. nonemmetropic eyes in their healthy control found no significant difference in axial length.10 This illustrates the importance of using the correct reference group, which was DISCUSSION also demonstrated by the observation that removing all ametropic eyes from our control group (ie, retaining only Statistical Comparison the emmetropic eyes) causes the differences seen in vitreous The processes underlying keratoconus have a very depth and axial length to become significant (Table 1). Given pronounced effect on the shape of the anterior segment, as the relatively high prevalence of myopia in the West

Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Cornea Volume 35, Number 6, June 2016 The Noncorneal Biometry of Keratoconic Eyes

FIGURE 1. Scatterplot of the SE refraction in healthy and keratoconic eyes as a function of the axial length (A) and keratometry (B).

European population, removing all ametropic eyes from our Longitudinal Analyses cohort consequently reduces the mean axial length by about Without a doubt, the best way to conﬁrm the existence 0.6 mm, thus unduly increasing the difference with the of a greater axial elongation in patients with untreated keratoconic axial lengths. Hence, the reported inﬂuence of keratoconus would be through longitudinal observation, keratoconus on vitreous depth could be the result of a bias in alongside a healthy control group. To the best of our the choice of control group. knowledge such a study has yet to be reported, but there

TABLE 2. Comparison of the Least and the Most Symptomatic Eyes in 20 Patients With Keratoconus Difference Parameter N Mild KTC, Mean 6 SD Moderate KTC, Mean 6 SD Mean 6 SEM P* SE (D) 20 22.0 6 1.8 23.9 6 4.0 21.9 6 0.9 0.048

Km (D) 20 42.8 6 1.2 44.5 6 2.4 1.7 6 0.6 0.014 Ast (D) 20 20.8 6 0.6 23.1 6 1.1 22.3 6 0.3 ,0.001 CCT (mm) 20 525.9 6 39.2 508.7 6 32.2 217.2 6 5.6 0.012 AD (mm) 20 3.1 6 0.4 3.2 6 0.4 0.1 6 0.02 0.002 T (mm) 13 3.9 6 0.3 3.9 6 0.3 0.02 6 0.02 0.834 V (mm) 13 16.8 6 0.9 16.8 6 1.0 0.0 6 0.1 0.529 L (mm) 20 24.4 6 0.9 24.4 6 1.0 0.03 6 0.1 0.940

PL (mm) 13 23.2 6 1.7 23.3 6 2.9 0.1 6 0.6 0.861

*Wilcoxon signed-rank test; P , 0.05/9 = 0.0056 (Bonferroni correction) indicates a signiﬁcant difference (boldface). Ast, corneal astigmatism; Km, mean keratometry; KTC, keratoconus; L, axial length; PL, lens power; SEM, standard error of the mean; T, lens thickness; V, vitreous depth.

Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Rozema et al Cornea Volume 35, Number 6, June 2016

FIGURE 2. Simulated distribution of the SE refraction in case the keratometry was randomly permuted between the 128 healthy eyes from Table 1 for which all data was available, shown for a postoperative keratometry increase of 0D (A) and 1D (B). The solid black line corresponds to the average of 100 runs of the algorithm and the dashed lines with the 95% confidence interval. are 2 studies looking at the axial length as part of a post- Digital Content 1, http://links.lww.com/ICO/A397). For both operative follow-up. In a recent follow-up of 14 keratoconic graft types, the weighted mean of these parameters led to eyes after epi-off cross-linking, Goldich et al23 reported no refractive values that were about 1.25D more myopic, and significant change in axial length 1 year postoperatively, keratometry about 1.30D steeper than the corresponding followed by a sudden increase of 0.3 mm in the next year. values of our healthy cohort, which corresponds very well Meanwhile, the mean keratometry and SE refraction remained with the results of the simulation. Another verification can be the same between 1 and 2 years of follow-up, despite the axial done by using the observation by Jones et al29 that 24% of length increase. This is only possible if at the same time the patients with a same-sized graft have myopia larger than crystalline lens would have reduced power by about 1D, 23D, which increased to 38% for patients with an oversized which is far more than the 0.12D per 5 years normally seen in graft. Assuming a postoperative keratometry increase of 1D, subjects of that age.24 As the mean lens power in keratoconic we found this to be the case for 35.7 6 2.2% of eyes in our and healthy eyes did not differ significantly from each other simulation, which is also a very good agreement. Finally, Tuft (Table 1), these results seem rather unlikely. reported that after PK in keratoconus there was a strong Similarly, Italon et al25 reported a 0.2-mm increase in correlation of the SE with axial length (r2 = 0.545), which is vitreous depth 2 years after PK in 30 eyes. Unfortunately, also similar in our simulation (r2 = 0.609). These results preoperative refraction and keratometry was missing from this confirm that our model provides a realistic description of article, rendering a deeper analysis of the observed refractive myopia after PK. changes impossible. Moreover, biometry was measured using Incidentally, the weighted means in Table S1 (available ultrasound, which has a relatively large reproducibility error as supplementary material) also have a large overlap in the compared with optical methods.13 standard errors of both oversized and same-sized grafts, making it unlikely that these differences are statistically significant. This supports the report by Jaycock et al30 that Myopia After PK the use of same-sized donor and recipient trephines did not The issue of the high prevalence of myopia after PK reduce myopia compared with use of oversized donor buttons. reported in the literature can be mostly explained by the random combination of a donor cornea with a recipient axial length that was not attuned to its specific curvature during the Biological Arguments Against emmetropization process. The result is a broad distribution of Axial Elongation the postoperative refraction centered around low myopia (see Axial elongation involves the sclera, which has great Fig. 2), which shifts further to the myopic side if there is an structural similarities to the cornea. Both structures form the increase in mean postoperative keratometry. This postopera- outer limit of the eye, and both require strength to maintain tive myopia was shown to be associated with graft button the pressure within the eye and to resist forces applied by the size,3 suture technique,1,2 time after suture removal, and axial extraocular muscles during eye movement. Moreover, both length.26–28 In order to assess the realism of the simulated are made up of connective tissue consisting of collagen fibrils post-PK results in Figure. 2, we collected all articles reporting embedded in a proteoglycan-rich extrafibrillar matrix.31 But mean and SD of the refraction or keratometry after full PK despite these similarities, there is one major structural with a donor button either larger or the same size as the difference between them because the cornea consists of opening in the acceptor cornea (see Table, Supplemental regularly stacked, narrow collagen fibrils with a uniform

Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Cornea Volume 35, Number 6, June 2016 The Noncorneal Biometry of Keratoconic Eyes diameter, making it transparent,22,32 whereas in the sclera the 14. Bennett A. A method of determining the equivalent powers of the eye and its crystalline lens without resort to phakometry. Ophthal Physiol collagen arrangement shows wider, irregularly interwoven – fibrils, rendering it opaque. This is associated with keratan Opt. 1988;8:53 59. 33 15. Rozema JJ, Atchison DA, Tassignon M. Comparing methods to estimate sulfate proteoglycan (KS-PG), which is present in signifi- the human lens power. Invest Ophthalmol Vis Sci. 2011;52:7937–7942. cantly higher concentrations in the cornea compared with the 16. Gerrard A, Burch J. Introduction to Matrix Methods in Optics. New sclera or the limbus.34 Since in keratoconus the structural York, NY: Dover; 1994. – alteration of KS-PG reduces the number of keratin sulfate 17. Rabinowitz YS. Keratoconus. Surv Ophthalmol. 1998;42:297 319. 35 18. Roberts CJ, Dupps WJ. Biomechanics of corneal ectasia and bio- chains, and malfunctioning keratocytes lead to a disturbance mechanical treatments. J Cataract Refr Surg. 2014;40:991–998. of KS-PG metabolism, this results in dysregulation of 19. Edmonds CR, Wung S-F, Pemberton B, et al. Comparison of anterior proteoglycan turnover in the extracellular matrix36 and chamber depth of normal and keratoconus eyes using Scheimpflug a consequent biomechanical weakening of the cornea. This photography. Eye Contact Lens. 2009;35:120–122. may also be associated with recently observed reductions in 20. Abolbashari F, Mohidin N, Hosseini SMA, et al. Anterior segment 37,38 characteristics of keratoconus eyes in a sample of Asian population. corneal transparency in keratoconic corneas. As the Contact Lens Ant Eye. 2013;36:191–195. concentration of KS-PG is considerably lower in the sclera, 21. Fontes BM, Ambrósio R, Jardim D, et al. Corneal biomechanical metrics this could explain why keratoconus does not seem to affect and anterior segment parameters in mild keratoconus. Ophthalmology. – the sclera. This was confirmed by Schlatter et al,10 who found 2010;117:673 679. fi 22. Feuk T. On the transparency of the stroma in the mammalian cornea. no signi cant differences in anterior scleral thickness between IEEE Trans Biomed Eng. 1970;17:186–190. healthy and keratoconic eyes. 23. Goldich Y, Barkana Y, Lior OW, et al. Corneal collagen cross-linking for All these considerations confirm that keratoconus has the treatment of progressive keratoconus: 3-year prospective outcome. no major influence on axial myopia and that the pathology is Can J Ophthalmol. 2014;49:54–59. 24. Jongenelen S, Rozema JJ, Tassignon MJ. Distribution of the lens power purely corneal in nature. The increase in anterior depth, on the in vivo as a function of age. Invest Ophthalmol Vis Sci. 2015;56: other hand, could be explained by a combination of the 7029–7035. thinning and outward expansion of the cornea. Clinically, 25. Italon C, Pieh S, Hanselmayer G, et al. Changes of axial length and these findings may be important to better understand the keratometry after keratoplasty for keratoconus using the guided trephine – refractive development in patients with keratoconus, as well system. Am J Ophthalmol. 2002;134:696 700. 26. Javadi MA, Feizi S, Rastegarpour A. Effect of vitreous length and as for lens power calculations in patients with keratoconus, trephine size disparity on post-DALK refractive status. Cornea. 2011;30: where axial length is the most important factor for lens power. 419–423. 27. Feizi S, Javadi MA. Factors predicting refractive outcomes after deep anterior lamellar keratoplasty in keratoconus. Am J Ophthalmol. 2015; 160:648–653.e2. REFERENCES 28. Shimmura S, Ando M, Ishioka M, et al. Same-size donor corneas for 1. Troutman R, Meltzer M. Astigmatism and myopia in keratoconus. Trans myopic keratoconus. Cornea. 2004;23:345–349. Am Ophthalmol Soc. 1972;70:265. 29. Jones MN, Armitage WJ, Ayliffe W, et al. Penetrating and deep 2. Jensen AD, Maumenee A. Refractive errors following keratoplasty. anterior lamellar keratoplasty for keratoconus: a comparison of graft Trans Am Ophthalmol Soc. 1974;72:123. outcomes in the United Kingdom. Invest Ophthalmol Vis Sci. 2009;50: 3. Wilson SE, Bourne WM. Effect of recipient-donor trephine size disparity 5625–5629. on refractive error in keratoconus. Ophthalmology. 1989;96:299–305. 30. Jaycock PD, Jones MN, Males J, et al. Outcomes of same-sizing versus 4. Brooks A, Robertson IF, Mahoney A. Ocular rigidity and intraocular oversizing donor trephines in keratoconic patients undergoing first pressure in keratoconus. Aust J Opthalmol. 1984;12:317–324. penetrating keratoplasty. Ophthalmology. 2008;115:268–275. 5. Tuft SJ, Fitzke FW, Buckley RJ. Myopia following penetrating 31. Maurice DM. The cornea and sclera. In: Davson H, ed. The Eye. New keratoplasty for keratoconus. Br J Ophthalmol. 1992;76:642–645. York, NY: Academic Press; 1984:1–158. 6. Touzeau O, Scheer S, Allouch C, et al. The relationship between 32. Maurice DM. The structure and transparency of the cornea. J Physiol. keratoconus and axial myopia [in French]. J Fr Ophthalmol. 2004;27: 1957;136:263–286. 765–771. 33. Müller LJ, Pels E, Schurmans LR, et al. A new three-dimensional model 7. Ernst BJ, Hsu HY. Keratoconus association with axial myopia: a pro- of the organization of proteoglycans and collagen fibrils in the human spective biometric study. Eye Contact Lens. 2011;37:2–5. corneal stroma. Exp Eye Res. 2004;78:493–501. 8. Günes¸ A. Keratoconus association with axial myopia [in Turkish]. Turk J 34. Akhtar S, Kerr BC, Hayes AJ, et al. Immunochemical localization of Ophthalmol. 2012;42:246–248. keratan sulfate proteoglycans in cornea, sclera, and limbus using 9. Lanier JD, Bullington RH Jr, Prager TC. Axial length in keratoconus. a keratanase-generated neoepitope monoclonal antibody. Invest Oph- Cornea. 1992;11:250–254. thalmol Vis Sci. 2008;49:2424. 10. Schlatter B, Beck M, Frueh BE, et al. Evaluation of scleral and corneal 35. Funderburgh J, Panjwani N, Conrad G, et al. Altered keratan sulfate thickness in keratoconus patients. J Cataract Refr Surg. 2015;41:1073–1080. epitopes in keratoconus. Invest Ophthalmol Vis Sci. 1989;30:2278–2281. 11. Li S, Li Z, Shi W, et al. Clinical features of 233 cases of keratoconus [in 36. Akhtar S, Bron A, Hayes AJ, et al. Role of keratan sulphate (sulphated Chinese]. Zhonghua Yan Ke Za Zhi. 2005;41:610–613. poly-N-acetyllactosamine repeats) in keratoconic cornea, histochemical, 12. Salmon TO, Thibos LN. Videokeratoscope–line-of-sight misalignment and ultrastructural analysis. Graefes Arch Clin Exp Ophthalmol. 2011; and its effect on measurements of corneal and internal ocular aberrations. 249:413–420. J Opt Soc Am A. 2002;19:657–669. 37. Lopes B, Ramos I, Ambrósio R Jr. Corneal densitometry in keratoconus. 13. Rozema JJ, Wouters K, Mathysen DG, et al. Overview of the Cornea. 2014;33:1282–1286. repeatability, reproducibility, and agreement of the biometry values 38. Anayol M, Sekeroglu M, Ceran B, et al. Quantitative assessment of provided by various ophthalmic devices. Am J Ophthalmol. 2014;158: corneal clarity in keratoconus: a case control study of corneal densitom- 1111–1120.e1. etry. Eur J Ophthalmol. 2015;26:18–23.