- Email: [email protected]
A combination of topographic and pachymetric parameters in keratoconus diagnosis
G Model
ARTICLE IN PRESS
CLAE-799; No. of Pages 6
Contact Lens & Anterior Eye xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Contact Lens & Anterior Eye journal homepage: www.elsevier.com/locate/clae
A combination of topographic and pachymetric parameters in keratoconus diagnosis Ibrahim Toprak a,∗ , Volkan Yaylalı b , Cem Yildirim b a b
Department of Ophthalmology, Servergazi State Hospital, Denizli, Turkey Department of Ophthalmology, Faculty of Medicine, Pamukkale University, Denizli, Turkey
a r t i c l e
i n f o
Article history: Received 27 December 2014 Received in revised form 30 March 2015 Accepted 3 April 2015 Keywords: Diagnosis Index Keratoconus Maximum keratometry Pachymetry
a b s t r a c t Purpose: To evaluate the utility of topographic and pachymetric parameters of Scheimpflug system in keratoconus diagnosis. Methods: This study included 183 eyes of 183 patients with keratoconus (keratoconus group) and 131 eyes of 131 age and sex-matched healthy subjects (control group). Mean keratometry (K, front), topographic astigmatism, pupil-center pachymetry, apical pachymetry, thinnest pachymetry (TP), corneal volume and maximum K (Kmax) were obtained from the Scheimpflug imaging system. A receiver operating characteristic (ROC) analysis was performed and area under the curve (AUC) was calculated to determine the diagnostic ability of each parameter in eyes with ≤ stage 3, ≤ stage 2 and stage 1 keratoconus based on the Amsler–Krumeich grading system. Results: The Kmax and TP showed the highest individual performance (with sensitivity–specificity of 92.9–92.4% and 89.6–93.3%, respectively) in diagnosis of keratoconus. The AUCs and sensitivity–specificity values for the Kmax/TP and Kmax2 /TP were calculated to improve the diagnostic performance. As expected, sensitivity–specificity values significantly increased by using Kmax/TP (97.3–94.7% at the level ≥0.08) and Kmax2 /TP (99.5–95.7% at the level ≥4.1) in discrimination of keratoconic eyes from normals. Moreover, Kmax2 /TP had very high sensitivity (>99%) and specificity (>94%) in diagnosis of stage 1 and stage 2 keratoconus. Conclusions: Although Kmax and TP appear to have high diagnostic ability in keratoconus, the use of either single parameter in isolation might be unsatisfactory in differential diagnosis. Therefore, the Kmax2 /TP ratio has been introduced, which reflects major characteristics of keratoconus and might be used as an important criterion in keratoconus diagnosis. © 2015 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
1. Introduction Keratoconus is a non-inflammatory corneal disease with progressive thinning and apical protrusion, which result in visual deterioration secondary to high myopia and irregular astigmatism [1]. The incidence of keratoconus in the general population is 1 per 2000 and no gender predominance was reported. The disease begins at puberty and shows progression until the third-fourth decade of life [1]. Clinical symptoms vary on the stage of the disease. A detailed patient history, refraction, keratometry and typical findings such as stromal thinning, iron deposits within the corneal epithelium (Fleischer’s ring), vertical stress lines in the deep stroma
∗ Corresponding author at: Department of Ophthalmology, Servergazi State Hospital, Bereketli Beldesi, Denizli 20070, Turkey. Tel.: +90 505 495 37 91; fax: +90 258 361 31 01. E-mail address: [email protected] (I. Toprak).
(Vogt’s striae), anterior stromal scars, Munson’s sign, Rizzuti’s phenomenon, scissoring reflex and oil droplet sign on retinoscopy sign on retinoscopy guide the clinician in diagnosis of moderate to advanced keratoconus [1–3]. However, in early cases and forme fruste (or subclinical) keratoconus, corrected distance visual acuity (CDVA) is generally 20/20 (Snellen equivalent) and typical signs are absent [1–3]. Corneal topography devices are widely used, and accepted as the reference method for keratoconus diagnosis [2,3]. A relatively new technology termed as Scheimpflug imaging, captures highresolution slit images of the anterior segment structures from the anterior corneal surface to the posterior of the lens using a 360◦ rotating camera. The Scheimpflug system allows anterior chamber and lens assessment, pachymetric mapping of the entire cornea and corneal wavefront, as well as corneal topography [2,4]. Although topography devices provide valuable corneal data and typical patterns for keratoconus, it can be challenging to discriminate very early or subclinical keratoconus from normal. Thus,
http://dx.doi.org/10.1016/j.clae.2015.04.001 1367-0484/© 2015 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
2
researchers have presented various indices and algorithms to assist clinicians for accurate diagnosis, such as central keratometry (K), topographic astigmatism, skewed steepest radial axis index (SRAX), the modified Rabinowitz–McDonnell indices (K and I–S values), the Maeda–Klyce (KCI% and keratoconus prediction index [KPI]) indices and the KISA% index [5–8]. In recent years, several topography based diagnostic algorithms have been developed to detect eyes with early keratoconus or forme fruste keratoconus, and the Scheimpflug technology has gained importance for the diagnosis, treatment planning and postoperative follow-up in the area of keratorefractive surgery [9–12]. In the current study, the performance of topographic and pachymetric parameters obtained from the Scheimpflug imaging system (Oculus Pentacam, Oculus Optikgerate GmbH, Wetzlar, Germany) in the diagnosis of keratoconus was measured. It was also shown that a combination of topographic and pachymetric data offers a highly sensitive and specific practical formula for distinguishing eyes with keratoconus from normal. 2. Materials and methods 2.1. Study group This retrospective controlled study involved 183 eyes of 183 patients with a confirmed diagnosis of keratoconus (keratoconus group) and 131 eyes of 131 age and sex-matched healthy controls (control group). The study followed the tenets of the Declaration of Helsinki and local ethics committee approved the methodology. All participants underwent a complete ophthalmological examination including CDVA measurement (Snellen charts), slit-lamp biomicroscopy, applanation tonometry, dilated retinoscopy and indirect fundus examination (with non-contact + 90 diopters [D] lens), and the Scheimpflug system anterior segment tomography (Oculus Pentacam, Oculus Optikgerate GmbH, Wetzlar, Germany). 2.2. Inclusion and exclusion criteria Keratoconus diagnosis was confirmed using clinical and topographical findings as follows; biomicroscopic signs (corneal thinning, Fleischer’s ring, Vogt’s striae, Munson’s sign, Rizzuti’s phenomenon), retinoscopic evidences (oil droplet and scissoring reflex), typical topographical patterns previously described for keratoconus, inferior–superior value (I–S) on topographic map >1.5. Control subjects (refractive surgery candidates) had normal biomicroscopic and topographical examination with a CDVA of 20/20 Snellen equivalent. All patients and controls were aged between 18 and 40 years and had been asked not to wear their contact lenses for at least for one month prior to the examination day. In bilateral cases, one eye was selected randomly. Subjects with history of any other corneal pathology, scarring or surgery, acute hydrops and dry eye were excluded from the study. 2.3. Corneal topography and pachymetry The Scheimpflug ocular imaging system (Oculus Pentacam, Oculus Optikgerate GmbH, Wetzlar, Germany) was used for anterior segment tomography (corneal topography and pachymetry). This device captures up to 50 sectional images of anterior segment structures in about two seconds using a slit light source and a 360◦ rotating camera. A single experienced technician performed the Scheimpflug imaging under scotopic conditions with undilated pupils (SV).
Images were captured in automatic mode and a single test with a quality score (QS) over 95% was used for the statistical analysis. Maps of sagittal curvature, corneal thickness, anterior and posterior elevation were acquired. The mean K (front), topographic astigmatism, pupil-center pachymetry, apical pachymetry, thinnest pachymetry (TP), corneal volume and maximum K (Kmax) values were recorded for each eye (Fig. 1). In the keratoconus group, disease severity was graded according to the Amsler–Krumeich classification system as follows [13]: Stage 1: Eccentric steepening; myopia, induced astigmatism, or both <5.00 D; mean central K <48 D. Stage 2: Myopia, induced astigmatism, or both from 5.00 to 8.00 D; mean central K readings <53.00 D; absence of scarring; corneal thickness >400 m. Stage 3: Myopia, induced astigmatism, or both from 8.00 to 10.00 D; mean central K readings >53.00 D, absence of scarring; corneal thickness 300–400 m. Stage 4: Refraction not measurable; mean central K readings >55.00 D; central corneal scarring, corneal thickness < 200 m. Eyes with stage 4 keratoconus were not included into the study (because of the corneal scarring). 2.4. Statistical analysis Statistical analysis was performed using the Statistical Package for Social Sciences software version 16.0 (SPSS Inc., Chicago, IL, USA) and MedCalc software version 12.6.1.0 (MedCalc Software bvba, Ostend, Belgium). The sample size in this study was tested using “Power and Beta” tool (under Procedures–Diagnostic tests–ROC curves) of the PASS software version 11.0.1 (NSCC, LLC, Utah, USA), and power of the present study was found to be 99% at 0.05 significance level (95% confidence interval). All variables were tested for normal distribution using the Kolmogorov–Smirnov method. Values were expressed as the mean ± standard deviation (SD). Categorical variables were analyzed using the Chi square test. The independent samples t test was used to determine difference between two groups in terms of quantitative variables. A receiver operating characteristic (ROC) curve and area under the ROC curve (AUC) were assessed. The diagnostic power of the each Scheimpflug parameter was graded according to the AUC value as follows; excellent (0.90–1.00), good (0.80–0.89), fair (0.70–0.79), poor (0.60–0.69) and worthless (0.50–0.59). The ROC curve plots the true positives (sensitivity) against the false positives (100-specificity) for different threshold values [14]. Values on the ROC curve, which indicated best sensitivity–specificity pair, were accepted as the cut points [14]. At 95% confidence interval, a P value less than 0.05 was considered statistically significant. 3. Results The study cohort comprised 183 eyes of 183 patients with a confirmed diagnosis of keratoconus (keratoconus group) and 131 eyes of 131 healthy subjects (control group). In the keratoconus group, 63 eyes (34.4%) had stage 1, 75 eyes (41%) had stage 2 and 45 eyes (24.6%) had stage 3 keratoconus based on the Amsler–Krumeich classification system [13]. Table 1 presents age, gender distribution and means for the Scheimpflug imaging system parameters (mean K [front], topographic astigmatism, pupil-center pachymetry, apical pachymetry, thinnest pachymetry, corneal volume and Kmax) in the keratoconus and control groups. In the entire cohort, the ROC analyses and AUC values showed that Kmax and TP had the highest individual diagnostic ability
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
3
Fig. 1. A Scheimpflug system report demonstrates standard parameters, maps of sagittal curvature, corneal thickness, anterior and posterior elevation.
followed by apical pachymetry, pupil-center pachymetry, mean K, corneal volume and topographic astigmatism in discrimination of keratoconic eyes from normals (Figs. 2 and 3 and Table 2). Based on the logic that steepening of the cornea (indicated by Kmax and positively related with keratoconus severity) and corneal thinning (indicated by TP and inversely related with keratoconus severity) are the major hallmarks of keratoconus, the ROC analysis was performed for Kmax/TP ratio and it showed 97.3%
sensitivity and 94.7% specificity at the level ≥0.08 in keratoconus diagnosis (AUC = 0.993) (Fig. 3). Moreover, when the squared Kmax (to enhance the effect of maximum keratometry [as the most powerful single parameter] on the formula) was used as the dividend (Kmax2 /TP), diagnostic power increased (99.5% sensitivity and 95.7% specificity, AUC = 0.997) at the cut point ≥4.1 (Fig. 3). Moreover, significant differences were found between the control and keratoconus groups in terms of the Kmax/TP (0.07 ± 0.01
Table 1 Comparison of the mean values in terms of age, gender and Scheimpflug parameters between the control and keratoconus groups. Control group (n = 131)
Keratoconus group (n = 183)
Age, years (mean ± SD) Gender, M/F
26.9 ± 4.3 64/67
27.8 ± 7.0 89/94
Scheimpflug parameters (mean ± SD) Mean K (front, D) Astigmatism (topographic, D) Pupil-center pachymetry (m) Apical pachymetry (m) Thinnest pachymetry (m) Corneal volume (mm3 ) Maximum K (Kmax, D)
43.47 ± 1.44 1.70 ± 1.21 566.23 ± 35.38 566.81 ± 35.09 563.34 ± 35.06 61.97 ± 3.66 44.97 ± 1.57
47.58 ± 3.34 3.78 ± 2.11 481.08 ± 39.96 476.18 ± 43.36 465.26 ± 44.20 56.81 ± 3.87 53.02 ± 5.0
P 0.169a 0.969b <0.001a <0.001a <0.001a <0.001a <0.001a <0.001a <0.001a
D, diopters; K, Keratometry; SD, standard deviation. a Independent samples t test, b Chi square test, P < 0.05 indicates statistical significance between two groups, bold and italic values emphasize statistical significance.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
4
Table 2 The AUC values and cut points for the Scheimpflug parameters, Kmax/TP and Kmax2 /TP in discriminating keratoconic eyes from healthy eyes. AUC value (at 95% CI lower–upper bounds)
P
Cut points (sensitivity–specificity)
Scheimpflug parameters (mean ± SD) Mean K (front, D) Astigmatism (topographic, D) Pupil-center pachymetry (m) Apical pachymetry (m) Thinnest pachymetry (m) Corneal volume (mm3 ) Maximum K (Kmax, D)
0.908 (0.876–0.939) 0.818 (0.771–0.865) 0.942 (0.917–0.967) 0.946 (0.922–0.970) 0.956 (0.934–0.977) 0.832 (0.788–0.875) 0.981 (0.969–0.992)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
≥45.2 (76–92.4%) ≥2.4 (74.3–78.6%) ≤519 (89.1–90.8%) ≤515 (86.9–93.1%) ≤513 (89.6–93.3%) ≤58.4 (67.2–85.5%) ≥47.4 (92.9–92.4%)
Kmax/TP (arbitrary unit) Kmax2 /TP (arbitrary unit)
0.993 (0.988–0.999) 0.997 (0.994–1.000)
<0.001 <0.001
≥0.08 (97.3–94.7%) ≥4.1 (99.5–95.7%)
P < 0.05 indicates statistical significance between two groups, bold and italic values emphasize statistical significance. AUC, area under the curve; CI, confidence interval; D, diopters; K, keratometry; SD, standard deviation; TP, thinnest pachymetry.
Furthermore, a logistic regression model was created to assess the roles of Kmax, TP, Kmax/TP and Kmax2 /TP on the probability of having keratoconus. The model showed a statistical significance (P < 0.0001) and explained 96.2% (Nagelkerke R2 ) of the relation between the predictors and prediction. Moreover, this regression model accurately classified 98.4% of the subjects. Kmax (P = 0.003) and the Kmax2 /TP (P = 0.022) had significant effect on keratoconus prediction, whereas TP (P = 0.189) and the Kmax/TP (P = 0.057) were not significantly associated with increased likelihood for having keratoconus. 4. Discussion
Fig. 2. The ROC curves for pupil-center pachymetry, apical pachymetry, thinnest pachymetry and corneal volume.
vs. 0.11 ± 0.02, respectively) and Kmax2 /TP (3.6 ± 0.3 vs. 6.2 ± 1.7, respectively) values (P < 0.001). In the keratoconus group, the diagnostic performance of the Kmax2 /TP ratio was tested in different stages of the disease based on the Amsler–Krumeich classification system (Table 3) [13]. The entire study group consisted of only eyes with ≤ stage 3 keratoconus. The ROC analyses were repeated in eyes with ≤ stage 2 and stage 1 keratoconus, and the Kmax2 /TP ratio continued to show very high discriminative ability between keratoconic eyes and healthy controls even in these early stages. Table 3 demonstrates the AUC values, cut points, sensitivity and specificity values according to the keratoconus severity.
Fig. 3. The ROC curves for mean K (front), topographic astigmatism, Kmax, Kmax/TP and Kmax2 /TP.
Corneal topography is an essential tool for diagnosing corneal ectatic disorders [15–17]. With an increase in refractive surgeries, accurate interpretation of corneal topographic outcomes has become critical. The purpose of the present study was to measure the ability of topographic and pachymetric parameters obtained from the Scheimpflug imaging system in keratoconus diagnosis. Although pachymetry based diagnostic algorithms are still being investigated, corneal pachymetry was not included in the formula of indices introduced by Rabinowitz–McDonnell [5,6,8,18–20]. In the present study, ROC analysis of each Scheimpflug parameter revealed that both Kmax and TP had high diagnostic ability in discriminating keratoconic eyes from normals (with sensitivity–specificity values of 92.9–92.4% and 89.6–93.3%, respectively). However, assessment of Kmax and TP individually might be insufficient for discriminating keratoconus from other corneal pathologies. Recent advances in corneal topographic devices have not completely solved the problem that very early or unmanifested cases may be misdiagnosed. Hence, many researchers developed valuable indices and maps [5–9,18–22]. Rabinowitz and McDonnell [5] introduced the central K and dioptric difference between inferior and superior area (I–S). They suggested that a central K value >47.2 and an I–S value >1.4 were indicative of keratoconus [5]. Another index termed SRAX was presented by Rabinowitz et al. [6] and it was based on quantification of typical irregular astigmatism in keratoconus. Lastly, they reported the KISA% index, which was a combination of central K, I–S value, topographic astigmatism and SRAX [8]. A KISA% index value of 100% was reported to be highly sensitive and specific for diagnosing keratoconus, and a value between 60 and 100 was suggested to indicate keratoconus suspect [8]. Furthermore, it was suggested that the use of topography-based quantitative variables significantly improved the repeatability, reliability and reproducibility of the diagnostic indices as in above-mentioned studies [17]. It is well known that a steeper K and a thinner cornea enhance clinical detectability of keratoconus and this logic constituted the
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
5
Table 3 Diagnostic ability of the Kmax2 /TP index (AUC values and cut points) in different stages of keratoconus based on Amsler–Krumeich grading system.a Kmax2 /TP
Keratoconus stages Stage 1 (n = 63) ≤ Stage 2 (n = 138)
AUC value (at 95% CI lower–upper bounds)
P
Cut points (sensitivity–specificity)
Mean ± SD
0.994 (0.987–1.000) 0.996 (0.992–1.000)
<0.001 <0.001
≥4.1 (99.4–94.7%) ≥4.1 (99.3–95.4%)
4.8 ± 0.4 5.3 ± 0.5
P < 0.05 indicates statistical significance between two groups, bold and italic values emphasize statistical significance. AUC, area under the curve; CI, confidence interval; D, diopters; K, keratometry; SD, standard deviation; TP, thinnest pachymetry. a Amsler–Krumeich classification for grading keratoconus. Stage 1: Eccentric steepening; myopia, induced astigmatism, or both <5.00 D; mean central K <48 D. Stage 2: Myopia, induced astigmatism, or both from 5.00 to 8.00 D; mean central K readings <53.00 D; absence of scarring; corneal thickness >400 m. Stage 3: Myopia, induced astigmatism, or both from 8.00 to 10.00 D; mean central K readings >53.00 D, absence of scarring; corneal thickness 300–400 m. Stage 4: Refraction not measurable; mean central K readings >55.00 D; central corneal scarring, corneal thickness < 200 m.
basis of the proportional (Kmax/TP and Kmax2 /TP) analysis in the present study. The Kmax2 /TP ratio showed the highest performance (99.5% sensitivity and 95.7% specificity) in keratoconus diagnosis when the cut point was ≥4.1. In the study of Rabinowitz and Rasheed [8], the KISA% index was tested in additional 8 eyes with keratoconus suspect and 12 eyes with early keratoconus. They found that 6/8 eyes with keratoconus suspect topography had a KISA% value between 60% and 100%, and 11/12 eyes with early keratoconus had a value of KISA% >100%. In the current study, the Kmax2 /TP ratio showed satisfying performance in discrimination of keratoconus from normal even in early cases (>99% sensitivity and >94% specificity). However, relatively small numbers of early cases in our study (n = 63) and the study of Rabinowitz et al. can be considered as a limitation and further studies in larger groups are needed [23]. In diagnosis of keratoconus, Maeda et al. [7] presented an automated expert classifier system and it was derived from discriminant analysis of eight indices obtained from TMS-1 videokeratoscope data. Another comparative study by them showed that the expert system classifier was significantly better than keratometry concerning sensitivity values (98% and 84%, respectively) [24]. In terms of specificity, the expert system classifier was reported to be significantly better than keratometry and modified Rabinowitz–McDonnell index (99%, 86% and 84%, respectively) for distinguishing keratoconus from normal [24]. The sensitivity–specificity values of previously introduced keratoconus indices might be important to show that the Kmax2 /TP ratio has a very high diagnostic power in keratoconus diagnosis. However, it should be mentioned that this study was not designed as a comparative study. Therefore, the formula should be tested in discrimination of keratoconus from other corneal diseases in a larger population, and this index should be compared with the previous keratoconus indices. On the other hand, the performance of the Kmax2 /TP ratio was analyzed in the early stages of keratoconus based on the Amsler–Krumeich classification system, whereas validity of the formula should be checked in other keratoconus grading systems, and particularly in diagnosis of forme fruste keratoconus. In conclusion, the performances of topographic and pachymetric parameters of the Scheimpflug system in diagnosis of keratoconus has been presented. Furthermore, a highly sensitive and specific practical formula (Kmax2 /TP) has been introduced for keratoconus diagnosis, which may be useful as an additional criterion for discriminating keratoconic eyes from normals in clinical practice. Acknowledgements No author has a financial or proprietary interest in any product, material, or method mentioned. No financial support was received
for this study. The abstract of this study has been accepted as a poster presentation by the program committee of the congress of European Society of Ophthalmology – SOE 2015 (6–9 June 2015, Vienna, Austria).
References [1] Rabinowitz YS. Keratoconus. Surv Ophthalmol 1998;42:297–319. [2] Kenney MC, Nesburn AB, Burgeson RE, Butkowski RJ, Ljubimov AV. Abnormalities of the extracellular matrix in keratoconus corneas. Cornea 1997;16:345–51. [3] Rabinowitz YS. Tangential vs sagittal videokeratographs in the early detection of keratoconus. Am J Ophthalmol 1996;122:887–9. [4] Barkana Y, Gerber Y, Elbaz U, Schwartz S, Ken-Dror G, Avni I, et al. Central corneal thickness measurement with the Pentacam Scheimpflug system, optical low-coherence reflectometry pachymeter, and ultrasound pachymetry. J Cataract Refract Surg 2005;31:1729–35. [5] Rabinowitz YS, McDonnell PJ. Computer-assisted corneal topography in keratoconus. Refract Corneal Surg 1989;5:400–8. [6] Rabinowitz YS. Videokeratographic indices to aid in screening for keratoconus. J Refract Surg 1995;11:371–9. [7] Maeda N, Klyce SD, Smolek MK, Thompson HW. Automated keratoconus screening with corneal topography analysis. Invest Ophthalmol Vis Sci 1994;35:2749–57. [8] Rabinowitz YS, Rasheed K. KISA% index: a quantitative videokeratography algorithm embodying minimal topographic criteria for diagnosing keratoconus. J Cataract Refract Surg 1999;25:1327–35. [9] Belin MW, Ambrósio R. Scheimpflug imaging for keratoconus and ectatic disease. Indian J Ophthalmol 2013;61:401–6. [10] Kozobolis V, Sideroudi H, Giarmoukakis A, Gkika M, Labiris G. Corneal biomechanical properties and anterior segment parameters in forme fruste keratoconus. Eur J Ophthalmol 2012;22:920–30. [11] Saad A, Gatinel D. Topographic and tomographic properties of forme fruste keratoconus corneas. Invest Ophthalmol Vis Sci 2010;51:5546–55. [12] Bühren J, Kook D, Yoon G, Kohnen T. Detection of subclinical keratoconus by using corneal anterior and posterior surface aberrations and thickness spatial profiles. Invest Ophthalmol Vis Sci 2010;51:3424–32. [13] Kamiya K, Ishii R, Shimizu K, Igarashi A. Evaluation of corneal elevation, pachymetry and keratometry in keratoconic eyes with respect to the stage of Amsler–Krumeich classification. Br J Ophthalmol 2014;98: 459–63. [14] Elbendary AM, Mohamed Helal R. Discriminating ability of spectral domain optical coherence tomography in different stages of glaucoma. Saudi J Ophthalmol 2013;27:19–24. [15] Wilson SE, Lin DT, Klyce SD. Corneal topography of keratoconus. Cornea 1991;10:2–8. [16] Tang M, Shekhar R, Miranda D, Huang D. Characteristics of keratoconus and pellucid marginal degeneration in mean curvature maps. Am J Ophthalmol 2005;140:993–1001. [17] Li X, Yang H, Rabinowitz YS. Keratoconus: classification scheme based on videokeratography and clinical signs. J Cataract Refract Surg 2009;35: 1597–603. [18] Prakash G, Agarwal A, Mazhari AI, Kumar G, Desai P, Kumar DA, et al. A new, pachymetry-based approach for diagnostic cutoffs for normal, suspect and keratoconic cornea. Eye (Lond) 2012;26:650–7. [19] Fukuda S, Yamanari M, Lim Y, Hoshi S, Beheregaray S, Oshika T, et al. Keratoconus diagnosis using anterior segment polarization-sensitive optical coherence tomography. Invest Ophthalmol Vis Sci 2013;54: 1384–91. [20] Qin B, Chen S, Brass R, Li Y, Tang M, Zhang X, et al. Keratoconus diagnosis with optical coherence tomography-based pachymetric scoring system. J Cataract Refract Surg 2013;39:1864–71.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
[21] Schweitzer C, Roberts CJ, Mahmoud AM, Colin J, Maurice-Tison S, Kerautret J. Screening of forme fruste keratoconus with the ocular response analyzer. Invest Ophthalmol Vis Sci 2010;51:2403–10. [22] Kocamıs¸ SI, Cakmak HB, Ca˘gıl N, Toklu Y. Investigation of the efficacy of the cone location and magnitude index in the diagnosis of keratoconus. Semin Ophthalmol 2014, http://dx.doi.org/10.3109/08820538.2014.914234.2014.05.19.
[23] Sedghipour MR, Sadigh AL, Motlagh BF. Revisiting corneal topography for the diagnosis of keratoconus: use of Rabinowitz’s KISA% index. Clin Ophthalmol 2012;6:181–4. [24] Maeda N, Klyce SD, Smolek MK. Comparison of methods for detecting keratoconus using videokeratography. Arch Ophthalmol 1995;113: 870–4.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
ARTICLE IN PRESS
CLAE-799; No. of Pages 6
Contact Lens & Anterior Eye xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Contact Lens & Anterior Eye journal homepage: www.elsevier.com/locate/clae
A combination of topographic and pachymetric parameters in keratoconus diagnosis Ibrahim Toprak a,∗ , Volkan Yaylalı b , Cem Yildirim b a b
Department of Ophthalmology, Servergazi State Hospital, Denizli, Turkey Department of Ophthalmology, Faculty of Medicine, Pamukkale University, Denizli, Turkey
a r t i c l e
i n f o
Article history: Received 27 December 2014 Received in revised form 30 March 2015 Accepted 3 April 2015 Keywords: Diagnosis Index Keratoconus Maximum keratometry Pachymetry
a b s t r a c t Purpose: To evaluate the utility of topographic and pachymetric parameters of Scheimpflug system in keratoconus diagnosis. Methods: This study included 183 eyes of 183 patients with keratoconus (keratoconus group) and 131 eyes of 131 age and sex-matched healthy subjects (control group). Mean keratometry (K, front), topographic astigmatism, pupil-center pachymetry, apical pachymetry, thinnest pachymetry (TP), corneal volume and maximum K (Kmax) were obtained from the Scheimpflug imaging system. A receiver operating characteristic (ROC) analysis was performed and area under the curve (AUC) was calculated to determine the diagnostic ability of each parameter in eyes with ≤ stage 3, ≤ stage 2 and stage 1 keratoconus based on the Amsler–Krumeich grading system. Results: The Kmax and TP showed the highest individual performance (with sensitivity–specificity of 92.9–92.4% and 89.6–93.3%, respectively) in diagnosis of keratoconus. The AUCs and sensitivity–specificity values for the Kmax/TP and Kmax2 /TP were calculated to improve the diagnostic performance. As expected, sensitivity–specificity values significantly increased by using Kmax/TP (97.3–94.7% at the level ≥0.08) and Kmax2 /TP (99.5–95.7% at the level ≥4.1) in discrimination of keratoconic eyes from normals. Moreover, Kmax2 /TP had very high sensitivity (>99%) and specificity (>94%) in diagnosis of stage 1 and stage 2 keratoconus. Conclusions: Although Kmax and TP appear to have high diagnostic ability in keratoconus, the use of either single parameter in isolation might be unsatisfactory in differential diagnosis. Therefore, the Kmax2 /TP ratio has been introduced, which reflects major characteristics of keratoconus and might be used as an important criterion in keratoconus diagnosis. © 2015 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
1. Introduction Keratoconus is a non-inflammatory corneal disease with progressive thinning and apical protrusion, which result in visual deterioration secondary to high myopia and irregular astigmatism [1]. The incidence of keratoconus in the general population is 1 per 2000 and no gender predominance was reported. The disease begins at puberty and shows progression until the third-fourth decade of life [1]. Clinical symptoms vary on the stage of the disease. A detailed patient history, refraction, keratometry and typical findings such as stromal thinning, iron deposits within the corneal epithelium (Fleischer’s ring), vertical stress lines in the deep stroma
∗ Corresponding author at: Department of Ophthalmology, Servergazi State Hospital, Bereketli Beldesi, Denizli 20070, Turkey. Tel.: +90 505 495 37 91; fax: +90 258 361 31 01. E-mail address: [email protected] (I. Toprak).
(Vogt’s striae), anterior stromal scars, Munson’s sign, Rizzuti’s phenomenon, scissoring reflex and oil droplet sign on retinoscopy sign on retinoscopy guide the clinician in diagnosis of moderate to advanced keratoconus [1–3]. However, in early cases and forme fruste (or subclinical) keratoconus, corrected distance visual acuity (CDVA) is generally 20/20 (Snellen equivalent) and typical signs are absent [1–3]. Corneal topography devices are widely used, and accepted as the reference method for keratoconus diagnosis [2,3]. A relatively new technology termed as Scheimpflug imaging, captures highresolution slit images of the anterior segment structures from the anterior corneal surface to the posterior of the lens using a 360◦ rotating camera. The Scheimpflug system allows anterior chamber and lens assessment, pachymetric mapping of the entire cornea and corneal wavefront, as well as corneal topography [2,4]. Although topography devices provide valuable corneal data and typical patterns for keratoconus, it can be challenging to discriminate very early or subclinical keratoconus from normal. Thus,
http://dx.doi.org/10.1016/j.clae.2015.04.001 1367-0484/© 2015 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
2
researchers have presented various indices and algorithms to assist clinicians for accurate diagnosis, such as central keratometry (K), topographic astigmatism, skewed steepest radial axis index (SRAX), the modified Rabinowitz–McDonnell indices (K and I–S values), the Maeda–Klyce (KCI% and keratoconus prediction index [KPI]) indices and the KISA% index [5–8]. In recent years, several topography based diagnostic algorithms have been developed to detect eyes with early keratoconus or forme fruste keratoconus, and the Scheimpflug technology has gained importance for the diagnosis, treatment planning and postoperative follow-up in the area of keratorefractive surgery [9–12]. In the current study, the performance of topographic and pachymetric parameters obtained from the Scheimpflug imaging system (Oculus Pentacam, Oculus Optikgerate GmbH, Wetzlar, Germany) in the diagnosis of keratoconus was measured. It was also shown that a combination of topographic and pachymetric data offers a highly sensitive and specific practical formula for distinguishing eyes with keratoconus from normal. 2. Materials and methods 2.1. Study group This retrospective controlled study involved 183 eyes of 183 patients with a confirmed diagnosis of keratoconus (keratoconus group) and 131 eyes of 131 age and sex-matched healthy controls (control group). The study followed the tenets of the Declaration of Helsinki and local ethics committee approved the methodology. All participants underwent a complete ophthalmological examination including CDVA measurement (Snellen charts), slit-lamp biomicroscopy, applanation tonometry, dilated retinoscopy and indirect fundus examination (with non-contact + 90 diopters [D] lens), and the Scheimpflug system anterior segment tomography (Oculus Pentacam, Oculus Optikgerate GmbH, Wetzlar, Germany). 2.2. Inclusion and exclusion criteria Keratoconus diagnosis was confirmed using clinical and topographical findings as follows; biomicroscopic signs (corneal thinning, Fleischer’s ring, Vogt’s striae, Munson’s sign, Rizzuti’s phenomenon), retinoscopic evidences (oil droplet and scissoring reflex), typical topographical patterns previously described for keratoconus, inferior–superior value (I–S) on topographic map >1.5. Control subjects (refractive surgery candidates) had normal biomicroscopic and topographical examination with a CDVA of 20/20 Snellen equivalent. All patients and controls were aged between 18 and 40 years and had been asked not to wear their contact lenses for at least for one month prior to the examination day. In bilateral cases, one eye was selected randomly. Subjects with history of any other corneal pathology, scarring or surgery, acute hydrops and dry eye were excluded from the study. 2.3. Corneal topography and pachymetry The Scheimpflug ocular imaging system (Oculus Pentacam, Oculus Optikgerate GmbH, Wetzlar, Germany) was used for anterior segment tomography (corneal topography and pachymetry). This device captures up to 50 sectional images of anterior segment structures in about two seconds using a slit light source and a 360◦ rotating camera. A single experienced technician performed the Scheimpflug imaging under scotopic conditions with undilated pupils (SV).
Images were captured in automatic mode and a single test with a quality score (QS) over 95% was used for the statistical analysis. Maps of sagittal curvature, corneal thickness, anterior and posterior elevation were acquired. The mean K (front), topographic astigmatism, pupil-center pachymetry, apical pachymetry, thinnest pachymetry (TP), corneal volume and maximum K (Kmax) values were recorded for each eye (Fig. 1). In the keratoconus group, disease severity was graded according to the Amsler–Krumeich classification system as follows [13]: Stage 1: Eccentric steepening; myopia, induced astigmatism, or both <5.00 D; mean central K <48 D. Stage 2: Myopia, induced astigmatism, or both from 5.00 to 8.00 D; mean central K readings <53.00 D; absence of scarring; corneal thickness >400 m. Stage 3: Myopia, induced astigmatism, or both from 8.00 to 10.00 D; mean central K readings >53.00 D, absence of scarring; corneal thickness 300–400 m. Stage 4: Refraction not measurable; mean central K readings >55.00 D; central corneal scarring, corneal thickness < 200 m. Eyes with stage 4 keratoconus were not included into the study (because of the corneal scarring). 2.4. Statistical analysis Statistical analysis was performed using the Statistical Package for Social Sciences software version 16.0 (SPSS Inc., Chicago, IL, USA) and MedCalc software version 12.6.1.0 (MedCalc Software bvba, Ostend, Belgium). The sample size in this study was tested using “Power and Beta” tool (under Procedures–Diagnostic tests–ROC curves) of the PASS software version 11.0.1 (NSCC, LLC, Utah, USA), and power of the present study was found to be 99% at 0.05 significance level (95% confidence interval). All variables were tested for normal distribution using the Kolmogorov–Smirnov method. Values were expressed as the mean ± standard deviation (SD). Categorical variables were analyzed using the Chi square test. The independent samples t test was used to determine difference between two groups in terms of quantitative variables. A receiver operating characteristic (ROC) curve and area under the ROC curve (AUC) were assessed. The diagnostic power of the each Scheimpflug parameter was graded according to the AUC value as follows; excellent (0.90–1.00), good (0.80–0.89), fair (0.70–0.79), poor (0.60–0.69) and worthless (0.50–0.59). The ROC curve plots the true positives (sensitivity) against the false positives (100-specificity) for different threshold values [14]. Values on the ROC curve, which indicated best sensitivity–specificity pair, were accepted as the cut points [14]. At 95% confidence interval, a P value less than 0.05 was considered statistically significant. 3. Results The study cohort comprised 183 eyes of 183 patients with a confirmed diagnosis of keratoconus (keratoconus group) and 131 eyes of 131 healthy subjects (control group). In the keratoconus group, 63 eyes (34.4%) had stage 1, 75 eyes (41%) had stage 2 and 45 eyes (24.6%) had stage 3 keratoconus based on the Amsler–Krumeich classification system [13]. Table 1 presents age, gender distribution and means for the Scheimpflug imaging system parameters (mean K [front], topographic astigmatism, pupil-center pachymetry, apical pachymetry, thinnest pachymetry, corneal volume and Kmax) in the keratoconus and control groups. In the entire cohort, the ROC analyses and AUC values showed that Kmax and TP had the highest individual diagnostic ability
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
3
Fig. 1. A Scheimpflug system report demonstrates standard parameters, maps of sagittal curvature, corneal thickness, anterior and posterior elevation.
followed by apical pachymetry, pupil-center pachymetry, mean K, corneal volume and topographic astigmatism in discrimination of keratoconic eyes from normals (Figs. 2 and 3 and Table 2). Based on the logic that steepening of the cornea (indicated by Kmax and positively related with keratoconus severity) and corneal thinning (indicated by TP and inversely related with keratoconus severity) are the major hallmarks of keratoconus, the ROC analysis was performed for Kmax/TP ratio and it showed 97.3%
sensitivity and 94.7% specificity at the level ≥0.08 in keratoconus diagnosis (AUC = 0.993) (Fig. 3). Moreover, when the squared Kmax (to enhance the effect of maximum keratometry [as the most powerful single parameter] on the formula) was used as the dividend (Kmax2 /TP), diagnostic power increased (99.5% sensitivity and 95.7% specificity, AUC = 0.997) at the cut point ≥4.1 (Fig. 3). Moreover, significant differences were found between the control and keratoconus groups in terms of the Kmax/TP (0.07 ± 0.01
Table 1 Comparison of the mean values in terms of age, gender and Scheimpflug parameters between the control and keratoconus groups. Control group (n = 131)
Keratoconus group (n = 183)
Age, years (mean ± SD) Gender, M/F
26.9 ± 4.3 64/67
27.8 ± 7.0 89/94
Scheimpflug parameters (mean ± SD) Mean K (front, D) Astigmatism (topographic, D) Pupil-center pachymetry (m) Apical pachymetry (m) Thinnest pachymetry (m) Corneal volume (mm3 ) Maximum K (Kmax, D)
43.47 ± 1.44 1.70 ± 1.21 566.23 ± 35.38 566.81 ± 35.09 563.34 ± 35.06 61.97 ± 3.66 44.97 ± 1.57
47.58 ± 3.34 3.78 ± 2.11 481.08 ± 39.96 476.18 ± 43.36 465.26 ± 44.20 56.81 ± 3.87 53.02 ± 5.0
P 0.169a 0.969b <0.001a <0.001a <0.001a <0.001a <0.001a <0.001a <0.001a
D, diopters; K, Keratometry; SD, standard deviation. a Independent samples t test, b Chi square test, P < 0.05 indicates statistical significance between two groups, bold and italic values emphasize statistical significance.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
4
Table 2 The AUC values and cut points for the Scheimpflug parameters, Kmax/TP and Kmax2 /TP in discriminating keratoconic eyes from healthy eyes. AUC value (at 95% CI lower–upper bounds)
P
Cut points (sensitivity–specificity)
Scheimpflug parameters (mean ± SD) Mean K (front, D) Astigmatism (topographic, D) Pupil-center pachymetry (m) Apical pachymetry (m) Thinnest pachymetry (m) Corneal volume (mm3 ) Maximum K (Kmax, D)
0.908 (0.876–0.939) 0.818 (0.771–0.865) 0.942 (0.917–0.967) 0.946 (0.922–0.970) 0.956 (0.934–0.977) 0.832 (0.788–0.875) 0.981 (0.969–0.992)
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
≥45.2 (76–92.4%) ≥2.4 (74.3–78.6%) ≤519 (89.1–90.8%) ≤515 (86.9–93.1%) ≤513 (89.6–93.3%) ≤58.4 (67.2–85.5%) ≥47.4 (92.9–92.4%)
Kmax/TP (arbitrary unit) Kmax2 /TP (arbitrary unit)
0.993 (0.988–0.999) 0.997 (0.994–1.000)
<0.001 <0.001
≥0.08 (97.3–94.7%) ≥4.1 (99.5–95.7%)
P < 0.05 indicates statistical significance between two groups, bold and italic values emphasize statistical significance. AUC, area under the curve; CI, confidence interval; D, diopters; K, keratometry; SD, standard deviation; TP, thinnest pachymetry.
Furthermore, a logistic regression model was created to assess the roles of Kmax, TP, Kmax/TP and Kmax2 /TP on the probability of having keratoconus. The model showed a statistical significance (P < 0.0001) and explained 96.2% (Nagelkerke R2 ) of the relation between the predictors and prediction. Moreover, this regression model accurately classified 98.4% of the subjects. Kmax (P = 0.003) and the Kmax2 /TP (P = 0.022) had significant effect on keratoconus prediction, whereas TP (P = 0.189) and the Kmax/TP (P = 0.057) were not significantly associated with increased likelihood for having keratoconus. 4. Discussion
Fig. 2. The ROC curves for pupil-center pachymetry, apical pachymetry, thinnest pachymetry and corneal volume.
vs. 0.11 ± 0.02, respectively) and Kmax2 /TP (3.6 ± 0.3 vs. 6.2 ± 1.7, respectively) values (P < 0.001). In the keratoconus group, the diagnostic performance of the Kmax2 /TP ratio was tested in different stages of the disease based on the Amsler–Krumeich classification system (Table 3) [13]. The entire study group consisted of only eyes with ≤ stage 3 keratoconus. The ROC analyses were repeated in eyes with ≤ stage 2 and stage 1 keratoconus, and the Kmax2 /TP ratio continued to show very high discriminative ability between keratoconic eyes and healthy controls even in these early stages. Table 3 demonstrates the AUC values, cut points, sensitivity and specificity values according to the keratoconus severity.
Fig. 3. The ROC curves for mean K (front), topographic astigmatism, Kmax, Kmax/TP and Kmax2 /TP.
Corneal topography is an essential tool for diagnosing corneal ectatic disorders [15–17]. With an increase in refractive surgeries, accurate interpretation of corneal topographic outcomes has become critical. The purpose of the present study was to measure the ability of topographic and pachymetric parameters obtained from the Scheimpflug imaging system in keratoconus diagnosis. Although pachymetry based diagnostic algorithms are still being investigated, corneal pachymetry was not included in the formula of indices introduced by Rabinowitz–McDonnell [5,6,8,18–20]. In the present study, ROC analysis of each Scheimpflug parameter revealed that both Kmax and TP had high diagnostic ability in discriminating keratoconic eyes from normals (with sensitivity–specificity values of 92.9–92.4% and 89.6–93.3%, respectively). However, assessment of Kmax and TP individually might be insufficient for discriminating keratoconus from other corneal pathologies. Recent advances in corneal topographic devices have not completely solved the problem that very early or unmanifested cases may be misdiagnosed. Hence, many researchers developed valuable indices and maps [5–9,18–22]. Rabinowitz and McDonnell [5] introduced the central K and dioptric difference between inferior and superior area (I–S). They suggested that a central K value >47.2 and an I–S value >1.4 were indicative of keratoconus [5]. Another index termed SRAX was presented by Rabinowitz et al. [6] and it was based on quantification of typical irregular astigmatism in keratoconus. Lastly, they reported the KISA% index, which was a combination of central K, I–S value, topographic astigmatism and SRAX [8]. A KISA% index value of 100% was reported to be highly sensitive and specific for diagnosing keratoconus, and a value between 60 and 100 was suggested to indicate keratoconus suspect [8]. Furthermore, it was suggested that the use of topography-based quantitative variables significantly improved the repeatability, reliability and reproducibility of the diagnostic indices as in above-mentioned studies [17]. It is well known that a steeper K and a thinner cornea enhance clinical detectability of keratoconus and this logic constituted the
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
5
Table 3 Diagnostic ability of the Kmax2 /TP index (AUC values and cut points) in different stages of keratoconus based on Amsler–Krumeich grading system.a Kmax2 /TP
Keratoconus stages Stage 1 (n = 63) ≤ Stage 2 (n = 138)
AUC value (at 95% CI lower–upper bounds)
P
Cut points (sensitivity–specificity)
Mean ± SD
0.994 (0.987–1.000) 0.996 (0.992–1.000)
<0.001 <0.001
≥4.1 (99.4–94.7%) ≥4.1 (99.3–95.4%)
4.8 ± 0.4 5.3 ± 0.5
P < 0.05 indicates statistical significance between two groups, bold and italic values emphasize statistical significance. AUC, area under the curve; CI, confidence interval; D, diopters; K, keratometry; SD, standard deviation; TP, thinnest pachymetry. a Amsler–Krumeich classification for grading keratoconus. Stage 1: Eccentric steepening; myopia, induced astigmatism, or both <5.00 D; mean central K <48 D. Stage 2: Myopia, induced astigmatism, or both from 5.00 to 8.00 D; mean central K readings <53.00 D; absence of scarring; corneal thickness >400 m. Stage 3: Myopia, induced astigmatism, or both from 8.00 to 10.00 D; mean central K readings >53.00 D, absence of scarring; corneal thickness 300–400 m. Stage 4: Refraction not measurable; mean central K readings >55.00 D; central corneal scarring, corneal thickness < 200 m.
basis of the proportional (Kmax/TP and Kmax2 /TP) analysis in the present study. The Kmax2 /TP ratio showed the highest performance (99.5% sensitivity and 95.7% specificity) in keratoconus diagnosis when the cut point was ≥4.1. In the study of Rabinowitz and Rasheed [8], the KISA% index was tested in additional 8 eyes with keratoconus suspect and 12 eyes with early keratoconus. They found that 6/8 eyes with keratoconus suspect topography had a KISA% value between 60% and 100%, and 11/12 eyes with early keratoconus had a value of KISA% >100%. In the current study, the Kmax2 /TP ratio showed satisfying performance in discrimination of keratoconus from normal even in early cases (>99% sensitivity and >94% specificity). However, relatively small numbers of early cases in our study (n = 63) and the study of Rabinowitz et al. can be considered as a limitation and further studies in larger groups are needed [23]. In diagnosis of keratoconus, Maeda et al. [7] presented an automated expert classifier system and it was derived from discriminant analysis of eight indices obtained from TMS-1 videokeratoscope data. Another comparative study by them showed that the expert system classifier was significantly better than keratometry concerning sensitivity values (98% and 84%, respectively) [24]. In terms of specificity, the expert system classifier was reported to be significantly better than keratometry and modified Rabinowitz–McDonnell index (99%, 86% and 84%, respectively) for distinguishing keratoconus from normal [24]. The sensitivity–specificity values of previously introduced keratoconus indices might be important to show that the Kmax2 /TP ratio has a very high diagnostic power in keratoconus diagnosis. However, it should be mentioned that this study was not designed as a comparative study. Therefore, the formula should be tested in discrimination of keratoconus from other corneal diseases in a larger population, and this index should be compared with the previous keratoconus indices. On the other hand, the performance of the Kmax2 /TP ratio was analyzed in the early stages of keratoconus based on the Amsler–Krumeich classification system, whereas validity of the formula should be checked in other keratoconus grading systems, and particularly in diagnosis of forme fruste keratoconus. In conclusion, the performances of topographic and pachymetric parameters of the Scheimpflug system in diagnosis of keratoconus has been presented. Furthermore, a highly sensitive and specific practical formula (Kmax2 /TP) has been introduced for keratoconus diagnosis, which may be useful as an additional criterion for discriminating keratoconic eyes from normals in clinical practice. Acknowledgements No author has a financial or proprietary interest in any product, material, or method mentioned. No financial support was received
for this study. The abstract of this study has been accepted as a poster presentation by the program committee of the congress of European Society of Ophthalmology – SOE 2015 (6–9 June 2015, Vienna, Austria).
References [1] Rabinowitz YS. Keratoconus. Surv Ophthalmol 1998;42:297–319. [2] Kenney MC, Nesburn AB, Burgeson RE, Butkowski RJ, Ljubimov AV. Abnormalities of the extracellular matrix in keratoconus corneas. Cornea 1997;16:345–51. [3] Rabinowitz YS. Tangential vs sagittal videokeratographs in the early detection of keratoconus. Am J Ophthalmol 1996;122:887–9. [4] Barkana Y, Gerber Y, Elbaz U, Schwartz S, Ken-Dror G, Avni I, et al. Central corneal thickness measurement with the Pentacam Scheimpflug system, optical low-coherence reflectometry pachymeter, and ultrasound pachymetry. J Cataract Refract Surg 2005;31:1729–35. [5] Rabinowitz YS, McDonnell PJ. Computer-assisted corneal topography in keratoconus. Refract Corneal Surg 1989;5:400–8. [6] Rabinowitz YS. Videokeratographic indices to aid in screening for keratoconus. J Refract Surg 1995;11:371–9. [7] Maeda N, Klyce SD, Smolek MK, Thompson HW. Automated keratoconus screening with corneal topography analysis. Invest Ophthalmol Vis Sci 1994;35:2749–57. [8] Rabinowitz YS, Rasheed K. KISA% index: a quantitative videokeratography algorithm embodying minimal topographic criteria for diagnosing keratoconus. J Cataract Refract Surg 1999;25:1327–35. [9] Belin MW, Ambrósio R. Scheimpflug imaging for keratoconus and ectatic disease. Indian J Ophthalmol 2013;61:401–6. [10] Kozobolis V, Sideroudi H, Giarmoukakis A, Gkika M, Labiris G. Corneal biomechanical properties and anterior segment parameters in forme fruste keratoconus. Eur J Ophthalmol 2012;22:920–30. [11] Saad A, Gatinel D. Topographic and tomographic properties of forme fruste keratoconus corneas. Invest Ophthalmol Vis Sci 2010;51:5546–55. [12] Bühren J, Kook D, Yoon G, Kohnen T. Detection of subclinical keratoconus by using corneal anterior and posterior surface aberrations and thickness spatial profiles. Invest Ophthalmol Vis Sci 2010;51:3424–32. [13] Kamiya K, Ishii R, Shimizu K, Igarashi A. Evaluation of corneal elevation, pachymetry and keratometry in keratoconic eyes with respect to the stage of Amsler–Krumeich classification. Br J Ophthalmol 2014;98: 459–63. [14] Elbendary AM, Mohamed Helal R. Discriminating ability of spectral domain optical coherence tomography in different stages of glaucoma. Saudi J Ophthalmol 2013;27:19–24. [15] Wilson SE, Lin DT, Klyce SD. Corneal topography of keratoconus. Cornea 1991;10:2–8. [16] Tang M, Shekhar R, Miranda D, Huang D. Characteristics of keratoconus and pellucid marginal degeneration in mean curvature maps. Am J Ophthalmol 2005;140:993–1001. [17] Li X, Yang H, Rabinowitz YS. Keratoconus: classification scheme based on videokeratography and clinical signs. J Cataract Refract Surg 2009;35: 1597–603. [18] Prakash G, Agarwal A, Mazhari AI, Kumar G, Desai P, Kumar DA, et al. A new, pachymetry-based approach for diagnostic cutoffs for normal, suspect and keratoconic cornea. Eye (Lond) 2012;26:650–7. [19] Fukuda S, Yamanari M, Lim Y, Hoshi S, Beheregaray S, Oshika T, et al. Keratoconus diagnosis using anterior segment polarization-sensitive optical coherence tomography. Invest Ophthalmol Vis Sci 2013;54: 1384–91. [20] Qin B, Chen S, Brass R, Li Y, Tang M, Zhang X, et al. Keratoconus diagnosis with optical coherence tomography-based pachymetric scoring system. J Cataract Refract Surg 2013;39:1864–71.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001
G Model CLAE-799; No. of Pages 6 6
ARTICLE IN PRESS I. Toprak et al. / Contact Lens & Anterior Eye xxx (2015) xxx–xxx
[21] Schweitzer C, Roberts CJ, Mahmoud AM, Colin J, Maurice-Tison S, Kerautret J. Screening of forme fruste keratoconus with the ocular response analyzer. Invest Ophthalmol Vis Sci 2010;51:2403–10. [22] Kocamıs¸ SI, Cakmak HB, Ca˘gıl N, Toklu Y. Investigation of the efficacy of the cone location and magnitude index in the diagnosis of keratoconus. Semin Ophthalmol 2014, http://dx.doi.org/10.3109/08820538.2014.914234.2014.05.19.
[23] Sedghipour MR, Sadigh AL, Motlagh BF. Revisiting corneal topography for the diagnosis of keratoconus: use of Rabinowitz’s KISA% index. Clin Ophthalmol 2012;6:181–4. [24] Maeda N, Klyce SD, Smolek MK. Comparison of methods for detecting keratoconus using videokeratography. Arch Ophthalmol 1995;113: 870–4.
Please cite this article in press as: Toprak I, et al. A combination of topographic and pachymetric parameters in keratoconus diagnosis. Contact Lens Anterior Eye (2015), http://dx.doi.org/10.1016/j.clae.2015.04.001