Objective and subjective diagnostic parameters in the fellow eye of unilateral keratoconus

Keratoconus is a progressive corneal ectatic disorder that may have variable presentations especially in earlier stages. The disease is usually bilateral but with a high prevalence of inter-eye asymmetry. In certain cases of significant asymmetric presentation (so called unilateral keratoconus), the fellow eye may show the earliest and mildest form of the disease, corresponding to the definition of forme fruste keratoconus. There have been some reports that patients with unilateral keratoconus will eventually develop keratoconus in the fellow eye due to genetic causes [1, 2].Therefore, early detection of forme fruste keratoconus is very important.

Keratoconus should be diagnosed in the early stages, as studies have suggested that UV-A mediated cross-linking may stop the progression of the disease and preserve visual function [3, 4]. Moreover, discrimination of early keratoconus among refractive candidates may reduce the risk of post-Lasik ectasia, which is considered one of the most feared postoperative complications of refractive surgery [5].

The Pentacam imaging system provides several keratometric, pachymetric, and elevation parameters as well as the Belin Ambrósio deviation index with different sensitivity and specificity [6, 7].

All parameters should be evaluated to determine the most sensitive and specific parameters of early keratoconus.

The aim of this study was to evaluate and compare different Pentacam parameters. Moreover, the mesopic vision in patients with keratoconus in one eye and forme fruste keratoconus in the fellow eye was assessed to facilitate the diagnosis of this condition.

This cross-sectional study was carried out on patients presenting keratoconus in one eye and forme fruste keratoconus in the fellow eye; the patients attended the outpatient clinic of Mansoura ophthalmic center between January 2014 and February 2015. This study was approved by the institutional research board of the Mansoura faculty of medicine and was performed in accordance with the ethical standards of the Declaration of Helsinki. All patients included in the study provided informed consent.

Ocular examinations were performed on all eyes included in the study. Clinical Keratoconus was diagnosed if all the following criteria were found:1) asymmetric bow tie with skewed radial axis (SRAX), inferior or central steepening on the topography map, 2) mean keratometry (K) > 47diopters or an inferior–superior (I-S) value >1.4diopters according to the Rabinowitz and McDonnell criteria [8], and 3) at least one clinical symptom(i.e., stromal thinning, conical protrusion of corneal apex, Fleischer ring, Vogt striae or anterior stromal scar). Forme fruste keratoconus was diagnosed if all the following criteria were found: 1) normal topography, 2) mean K < 47D and I-S ≤ 1.4, 3) normal slit lamp, and 4) keratoconus in the fellow eye.

The eyes with an evident keratoconus diagnosis were included in (Group A) and the fellow eyes were included in (Group B).Control cases(Group C) were selected from the normal database of candidates for refractive surgery with the following:1)normal topography, 2) negative rotating Scheimpflug device indices and negative topographic keratoconus classification, 3) normal slit lamp examination, and 4) no history of eye disease(only one eye of each normal candidate was included in the study depending on a computer-based randomizing program).

The exclusion criteria included previous ocular surgery or trauma, significant corneal scarring or associated ocular pathology.

All eyes were examined by rotating Scheimpflug corneal tomography (Pentacam, Oculus Optikgerate GmbH).No patients in the study wore contact lenses. During the Scheimpflug corneal tomography examination, the patient was properly positioned on the chin rest and forehead strap. The patient was asked to blink several times then open both eyes and stare at a fixed target. After obtaining proper alignment, the automatic release mode started the scan; a total of25 single Scheimpflug images were captured within 2 s for each eye. Three consecutive scans were taken of each eye by the same examiner. The average of the results of three acceptable measurements was included in the study. Each eye was required to have a corneal map with at least 9 mm of corneal coverage and no extrapolated data.

The Pentacam maps were analyzed. The following anterior and posterior corneal surface parameters were evaluated by the Scheimpflug system: corneal dioptric power at the flattest meridian in the 3 mm-central zone(K1), corneal dioptric power in the steepest meridian in the 3 mm-central zone (K2) and mean corneal power in the 3 mm-zone (mean K).

The pachymetric map was analyzed, including the central corneal thickness (CCT) at the apex of the geometric center and corneal thickness at the thinnest point (CTmin).The average progression index (PPI avg) was calculated as the progression value at the different rings referenced to the mean curve. The minimum (PPI min) and maximum (PPI max) progression indices were recorded. The Ambrósior elational thickness (ART) was calculated as follows: ARTavg = CTmin/PPIavg, ARTmin = CTmin/PPImin, ARTmax = CTmin/PPImax [9].

The posterior corneal elevation maps were evaluated; a reference best fit sphere was calculated at a fixed optical zone of 8 mm, and the posterior corneal elevation values relative to this reference were recorded. The back difference elevation and multimetric D index values were extrapolated from the difference map of the Belin/Ambrósio-enhanced ectasia display of the Pentacam system.

A total of 48 patients attended the Mansoura ophthalmic center and were diagnosed with clinical keratoconus in one eye (GroupA) and forme fruste in the fellow eye (Group B).The control cases (Group C) included 72 eyes of normal refractive candidates.

Table 1 shows the mean keratometric values and the pachymetric and posterior elevation parameters for all groups.

As shown in Table 1, the keratometric reading (i.e., steep K, flat k, mean K), difference between the central corneal thickness and thinnest location, difference between the posterior corneal elevation (PE) and back difference elevation (BDE), and D index measurement were significantly higher in eyes with keratoconus than in eyes of normal control subjects; the corneal thickness and Ambrósio relational thickness (ART avg., ART min, ART max) were significantly lower in eyes with keratoconus than in eyes of normal control subjects (P < 0.05).

There was no statistically significant difference in the steep K, flat K, difference between pachymetry at the thinnest point and the apex, PPIavg, and PPI min between the forme fruste eyes (fellow eyes) and the control group. The mean K, corneal pachymetry, BDE, PE, PPImax, Ambrósio relational thickness (ARTmax, ARTmin), and multimetric D were statistically significant in the forme fruste eyes compared to the control group.

The receiver operating characteristic curve analysis included the AUC, standard error,95%CI,significance level, best cutoff points, sensitivity and specificity for each parameter to differentiate between the forme fruste and control groups; these data are shown in Table 2.

As shown in Table 2, there was a large AUC for the mean K, thinnest pachymetry, ART max, BDE, D index, and PPI max. However, the AUC was the greatest for the ART max (0.88), Fig. 1.

The ROC curve analysis of the Scheimpflug parameters to differentiate the clinical keratoconus group from the control group was shown in Table 3 and Fig. 2.

In discriminating between the keratoconus eyes and control group, almost all parameters had a large AUC. The D index had the greatest AUC, followed by posterior elevation, BDE, and ART max.

The visual acuity, spherical equivalent and outcome of the contrast and sensitivity test are illustrated in Table 4.For the contrast sensitivity test with glare, there were statistically significant differences between the forme fruste eyes and the control group (P < 0.01); for the contrast sensitivity test without glare, there were no significant differences between the forme fruste eyes and the control group (P < 0.06).For both contrast sensitivity tests, there were significant differences between the clinical keratoconus eyes and the control group.

It has been established that keratoconus is a bilateral progressive disease with strong genetic association [7].It is difficult to make a diagnosis of early keratoconus, as there is no clear definition of these entities in the literature. Many terms have been employed to describe the preclinical stages of the KC condition, including subclinical keratoconus, keratoconus suspect and forme fruste keratoconus [9, 10].We based our study on the Saad and Gatinel [10]definition of forme fruste keratoconus, which defines it as the contralateral eye of unilateral keratoconus without any ectatic changes clinically or topographically using Placido disk-based topography; the definition has been followed by many researchers [10‐13].

The introduction of the rotating Scheimpflug camera has introduced many diagnostic parameters, such as keratometric, pachymetric or elevation parameters. In this study, ROC curves were analyzed to detect the specificity and sensitivity of each parameter; moreover, the quality of mesopic vision was subjectively assessed to identify the sensitive and specific detectors, which support early and accurate diagnoses of forme fruste cases. The ART max had the largest area under the curve in forme fruste eyes compared to the control group, followed by the multimetric D index. This result does not agree with that of Muftuoglu et al., [14] who reported that the multimetric D had the largest AUC in cases of forme fruste keratoconus (used as simulation group of the subclinical cases).The discrepancy may be explained by different statistical analyses as the D index became more diagnostic when increased; however, the ARTmax became more diagnostic when decreased. We found an ARTmax cutoff value of 404 with a sensitivity and specificity of 82 and 70%, respectively. However, the D index cutoff was 1.2 with a sensitivity and specificity of 71and 60%, respectively, in cases of forme fruste keratoconus. In other words, the false negative results are rare when applying ART max; however, there is limited sensitivity and specificity when applying the multimetric D index, with possible false-positive and false-negative results in forme fruste cases. In clinical cases, the multimetric D had a larger AUC compared to the ART max; this result may be due to the multimetric D index being a multimetric parameter composed of keratometric, pachymetric, pachymetric progression and back elevation parameters [14], which change significantly in clinical keratoconus cases. For the forme fruste cases in this study, the largest AUC was found for the mean K, followed by PPI and back difference elevation, while the posterior elevation had a low AUC. This is similar to the findings of previous studies [6, 10], which indicated that the back difference elevation is better for diagnosing subtle cases of keratoconus in normal eyes compared to the back elevation. In our study, the sensitivity and specificity of the posterior elevation and back difference elevation were limited; this result is supported by the findings of Muftuoglu et al. [6], who stated that there was significant overlap of the back difference elevation and posterior elevation between forme fruste eyes and control eyes.

In our study, the posterior elevation was not found to be a sensitive parameter in detecting the forme fruste cases; this result disagrees with reports of Muftuoglo et al. [6], which may be due to most of the keratoconus eyes being mild to moderate cases, and the fellow eyes showed minimal changes in the early stage.

The corneal thickness and thinnest corneal points had large AUCs for diagnosing keratoconus, which agrees with previous studies [15‐17]; however, both parameters had low AUCs in forme fruste eyes. This result may be explained by the thickness not being significantly decreased in early keratoconus, so they had higher levels and lower sensitivity.

In this study, the contrast sensitivity test with glare was found to be an important diagnostic test for early forme fruste cases. This finding is in accordance with that of Bilen et al., [18] who reported that contrast sensitivity(CS) was affected more than corrected distant visual acuity (CDVA) in its correlation with refractive, topographic, pachymetric, and aberrometric changes. This report concluded that CS is a more sensitive indicator of visual function in keratoconus follow-up than CDVA; however, Bilen et al. assessed CS by the Hamilton veal chart [19].

Maeda et al. [20] found a significantly greater loss in CS in a keratoconus group compared to a normal control group. They concluded that subtle visual deteriorations, which are detected by CS testing, maybe predicted objectively by corneal topographic indices.

There is no single Pentacam parameter that can be used to diagnose forme fruste keratoconus with high specificity and sensitivity, and further studies should be conducted to determine more efficient parameters. The ART max seems to increase our detection rate of forme fruste keratoconus; however, the sensitivity and specificity of all Pentacam parameters are limited in diagnosing forme fruste keratoconus, including the ARTmax. Early detection of mesopic vision may be an easy outpatient subjective test to identify a forme fruste case and differentiate it from the true unilateral keratoconus, which despite being very rare should still be considered.