Central Corneal Thickness After Cross-linking Using High-Definition Optical Coherence Tomography, Ultrasound, and Dual Scheimpflug Tomography: A Comparative Study Over One Year




Purpose


To assess central corneal thickness (CCT) and corneal haze in keratoconus eyes following corneal cross-linking (CXL).


Design


Prospective noncomparative case series.


Methods


Forty patients (44 eyes) with keratoconus that had serial evaluation for CCT, following CXL, using high-definition optical coherence tomography (HD-OCT), ultrasound pachymetry (USP), and dual Scheimpflug tomography. CCT was measured at baseline, then at 2 weeks and 1, 3, 6, and 12 months postoperatively.


Results


The mean baseline CCT measurements were 470.02 μm, 469.79 μm, and 466.66 μm using the HD-OCT, the USP, and dual Scheimpflug tomography, respectively ( P = .91). Following CXL, the mean CCT measurements by dual Scheimpflug, at all follow-up periods, were lower compared to mean baseline reading ( P < .003). The mean CCT measurements by the HD-OCT and USP were similar to baseline readings except for the thicker readings at the 2-week follow-up visit. Overall, the mean CCT obtained by the USP was similar to that obtained by the HD-OCT throughout the 12-month period. However, the mean CCT measurements obtained by dual Scheimpflug tomography were statistically and clinically significantly lower than those obtained by both HD-OCT and USP at all follow-up visits.


Conclusions


USP and HD-OCT showed better agreement in CCT readings at all visits, and may better estimate the true CCT following CXL as compared to dual Scheimpflug tomography. The lower mean CCT post-CXL as measured by the latter seems to be correlated with the amount of haze that develops after cross-linking.


Keratoconus is a noninflammatory, progressive, and degenerative disease of the cornea characterized by central thinning and increased corneal curvature. Currently, several treatment options are available to manage the accompanying visual disturbances of this condition, including spectacles, gas-permeable rigid contact lenses, intracorneal ring segment implantation, laser surface ablation, corneal cross-linking, and corneal transplantation. Choosing the best treatment option is tailored according to the severity of the disease at presentation and stability during follow-ups. One important parameter in staging the disease, evaluating stability, and determining subsequent treatment is central corneal thickness (CCT).


CCT is most commonly measured using ultrasound pachymetry (USP) because of its low cost, ease of use, and high repeatability; however, it can be uncomfortable and operator dependent. In that regard, many noncontact devices have been used to measure CCT; of special interest are the dual Scheimpflug and optical coherence tomography techniques, owing to their widespread availability. In keratoconus, these noncontact instruments have shown satisfactory repeatability, reproducibility, and excellent correlation with USP. However, haze formation following corneal cross-linking (CXL) in keratoconic eyes has been shown to alter the corneal optical quality in terms of light scatter and (focal) index of refraction ; the altered pathway of the reflected waves of light or sound (infrared or ultrasound) coming from the back surface of the cornea may induce errors in corneal thickness measurements calculated by the capturing software or device.


In this prospective study we compare the CCT results of spectral-domain optical coherence tomography, ultrasound pachymetry, and dual Scheimpflug tomography over a 12-month period in keratoconic eyes that underwent cross-linking. Additionally, stromal haze was evaluated clinically and by OCT, preoperatively and postoperatively throughout the follow-up period, and correlated with pachymetry measurements by the 3 devices.


Methods


Patient Selection


This was a prospective noncomparative case series study of 28 male and 12 female subjects (44 eyes) with mild to moderate keratoconus in whom CXL was performed by a single surgeon (S.T.A.), between May 2013 and November 2014, at the American University of Beirut Medical Center, Lebanon. This study was approved by the Institutional Review Board at the American University of Beirut and adhered to the Declaration of Helsinki principles. Written informed consent was obtained from each subject.


All eyes enrolled in this study had progression of the disease in the 12 months prior to recruitment, as documented by serial differential corneal topographies and by pachymetry analysis. Keratoconus was graded according to the Amsler-Krumeich classification, based on patients’ refraction, mean central keratometry (K) reading, corneal signs, and central corneal thickness. As per prior studies, progression was defined as an increase in maximum K (Kmax) of at least 1.00 diopter (D) in 1 year, and/or a subjective reporting of visual deterioration due to corneal reasons. All patients underwent a complete ophthalmic examination including refraction, slit-lamp microscopy, applanation tonometry, dilated retinal examination, and corneal tomography. Exclusion criteria included intraocular pressure of >21 mm Hg, active ocular pathology, corneal thickness <400 μm at the thinnest point, corneal opacification/scars, history of keratitis (any form), peripheral marginal degeneration, previous corneal and/or intraocular surgeries, and autoimmune and/or connective tissue disease.


Corneal Collagen Cross-linking Technique


The eye to be treated was anesthetized by applying proparacaine hydrochloride 0.5% drops on 2 occasions at 5-minute intervals. After the patient was positioned under the operating microscope, an eyelid speculum was inserted and the central 9-mm corneal epithelium was removed with a blunt spatula. A mixed riboflavin 0.1%–20% dextran solution (IROC Innocross AG, Zurich, Switzerland) was instilled every 2 minutes until the riboflavin penetrated the corneal stroma (ie, approximately 30 minutes). An ultraviolet lamp (UV-X; IROC AG, Zurich, Switzerland) was then focused on the apex of the cornea at a distance of 5 cm for a total of 30 minutes, providing a mean of 3.0 mW/cm 2 radiant energy. The required irradiance was calibrated prior to each treatment. Riboflavin drops were applied to the cornea every 2 minutes during the total period of ultraviolet irradiation.


The thinnest and central corneal thickness was measured by USP to ensure a reading of at least 400 μm, before the UV lamp was focused on the apex of the cornea. In eyes with corneal stromal thickness (without epithelium) of less than 400 μm, swelling of the cornea was induced by instilling a hypo-osmolar riboflavin 0.1% solution, generated by diluting vitamin B2-riboflavin-5-phosphate 0.5% with physiologic sodium chloride 0.9% solution. Subsequently, ultrasound pachymetry was repeated to confirm that the stroma thickness had reached 400 μm or more.


After treatment, the eye surface was copiously washed with balanced salt solution and 2 drops of gatifloxacin 0.3% were instilled, followed by placement of a bandage soft contact lens. Postoperatively, patients received acetaminophen 500 mg every 8 hours for pain for 3 days, 1 drop of gatifloxacin 0.3% 4 times daily for 2 weeks with 1 drop of tobramycin–dexamethasone 0.1% 4 times daily for 1 week, and then 1 drop of fluorometholone 0.1% 4 times daily, slowly tapered over 6 weeks. The bandage soft contact lens was kept for at least 4 days postoperatively and the eye examined by slit-lamp microscopy to confirm complete corneal epithelialization before contact lens removal.


Corneal Thickness Measurements


Central corneal thickness measurements were performed at baseline and at 2 weeks and 1, 3, 6, and 12 months following CXL. A single operator experienced in using all 3 instruments performed the measurements. The measurements by caliper on the OCT were performed in another setting by a different investigator who was masked to the results of the measurements by the USP and dual Scheimpflug.


Cirrus high-definition optical coherence tomography (Cirrus HD-OCT; Carl Zeiss Meditec, Inc, Dublin, California, USA) was employed using the anterior segment cube 512 × 128 mode. Central corneal thickness was measured manually with the caliper tool in the cross-line scan at the corneal apex; the vertical distance between the 2 indicators of the caliper tool was considered to represent the CCT. The CCT measurements were always manually performed at the corneal apex. Two repeated measurements were performed by 2 masked observers pre- and post-surgery on all subjects.


The Galilei Dual Scheimpflug Analyzer (Ziemer, Port, Switzerland), set at 25 images per scan, performs a full scan within 1–2 seconds and combines 2 technologies—a dual Scheimpflug camera and a Placido disk—to measure both surfaces of the cornea. The dual camera captures and averages Scheimpflug slit images from opposite sides of the illuminated slit, all while tracking decentration due to eye movements.


Sonogage ultrasound pachymetry (CORNEO-GAGE; Sonogage, Cleveland, Ohio, USA) was used at the frequency of 50 MHz to measure CCT.


For all volunteers, USP was performed last, because it required topical anesthesia and contact with the ultrasound probe, which would have skewed the Cirrus and Galilei measurements. In addition, all eye measurements were performed without dilation in a dim room according to the manufacturer’s guidelines. For both the Galilei and Cirrus imaging devices, the subjects were positioned with a headrest and asked to fixate on an internal fixation within each device. As this was an inter-method agreement study, the repeatability of each instrument was not assessed. A single measurement for both the Galilei dual Scheimpflug and Cirrus HD-OCT devices with no replicates was recorded. For USP, 5 measurements were obtained and the thinnest was recorded.


Haze Assessment


The criteria used by Fantes and associates was adopted for the clinical grading of haze at baseline and at each follow-up visit: 0 = no haze; +0.5 = trace haze on oblique illumination; +1 = corneal cloudiness not interfering with the visibility of fine iris details; +2 = mild effacement of fine iris details; +3 = moderate effacement of iris details; and +4 = complete obscuration of the iris details. In addition, the corneal densitometry analysis, which is provided as an add-on feature to the standard software of the Galilei Dual Scheimpflug Analyzer, was used to assess the amount of light scatter at baseline and at each follow-up visit. The output is expressed in grayscale units (GSU). The GSU scale is calibrated by proprietary software, which defines a minimum light scatter of 0 (maximum transparency) and maximum light scatter of 100 (minimum transparency). The images at all visits were performed in the same windowless clinical assessment room. Only the maximum GSU within the central 8-mm zone was used for analysis.


Statistical Analysis


Statistical analysis was performed using SPSS version 20.0 (SPSS Inc, Chicago, Illinois, USA) and Microsoft Office Excel was used for data management and analyses. Descriptive statistics were reported as means and standard deviations for continuous variables. One-way analysis of variance (ANOVA) was used to compare the baseline CCT obtained by the 3 different devices. Paired t test was used to compare postoperative change in measured CCT for each device. Two-way repeated-measures ANOVA with Bonferroni correction for post hoc analysis was used to compare the change in CCT following CXL between the devices. Reliability analysis with 2-way mixed model selected intraclass correlation coefficient (ICC) was calculated to assess interobserver reproducibility on OCT measurements. The correlation between Scheimpflug densitometry and CCT values was estimated by the Pearson correlation coefficient. A P value < .003 was considered to be statistically significant unless otherwise stated.




Results


In this prospective study 44 eyes of 40 patients (28 male and 12 female; mean age 22.0 ± 6.1 years) were analyzed. All patients completed the 12 months of follow-up. According to the Amsler-Krumeich classification, 10 eyes (22.7%) had grade 1, 28 eyes (63.6%) had grade 2, and 6 eyes (13.6%) had grade 3 keratoconus.


Change in Central Corneal Thickness Over Time


At baseline, there were no statistically significant differences for CCT as measured by the 3 instruments. CCT was 470.02 μm, 469.79 μm, and 466.66 μm for the Cirrus HD-OCT, USP, and Galilei dual Scheimpflug, respectively.


The mean CCT measurements by the dual Scheimpflug, at all follow-up periods, were statistically significantly lower compared with the mean baseline reading ( Table 1 ). The mean CCT readings were progressively lower following CXL, reaching the lowest at 3 months follow-up (434.65 ± 30.84 compared with 466.66 ± 43.70 at baseline; P < .001). Thereafter, the CCT readings increased but remained lower than baseline at 12 months (456.33 ± 45.34 compared with 466.66 ± 43.70 at baseline; P < .001).



Table 1

Means and Standard Deviations of Central Corneal Thickness Measured by Ultrasound Pachymetry, High-Definition Optical Coherence Tomography, and Dual Scheimpflug Devices at Baseline and at All Follow-up Periods After Corneal Cross-linking






































Modality CCT (μm)
Baseline 2 Weeks 1 Month 3 Months 6 Months 12 Months
Ultrasound pachymetry 469.79 ± 40.30 486.16 ± 40.74 (<.001) a 471.53 ± 41.02 (.078) 469.03 ± 36.88 (.115) 470.21 ± 39.76 (.105) 474.67± 40.80 (.109)
HD-OCT 470.02 ± 37.65 479.71 ± 36.15 (.019) 470.50 ± 36.42 (.035) 465.49 ± 33.51 (.593) 470.04 ± 38.32 (.183) 472.3 ± 37.32 (.564)
Dual Scheimpflug 466.66 ± 43.70 450.53 ± 42.51 (<.001) a 438.71 ± 38.76 (<.001) a 434.65 ± 30.84 (<.001) a 448.00 ± 38.52 (<.001) a 456.33 ± 45.34 (.002) a

CCT = central corneal thickness; HD-OCT = high-definition optical coherence tomography.

P values are shown in parentheses.

a P value <.003 is significant after Bonferroni correction.



As for the Cirrus HD-OCT and USP devices, the mean CCT readings were similar to baseline readings except for the 2-week follow-up visit ( Table 1 ). At that visit, mean CCT was 479.71 ± 36.15 μm for the Cirrus and 486.16 ± 40.74 μm for the USP ( P = .019 and P < .001, respectively; P < .003 is statistically significant after Bonferroni correction) ( Table 1 ). Excellent agreement was found at baseline and post surgery between the 2 measuring observers, with ICC of 0.988 (95% confidence interval [CI] 0.978–0.995) and 0.991 (95% CI 0.984–0.995), respectively.


Difference in Central Corneal Thickness Between the Devices


In comparison to the CCT reading by USP at baseline, the mean CCT by the Cirrus HD-OCT was 0.22 μm more, whereas it was 3.10 μm less using the Galilei dual Scheimpflug ( Table 2 ). Overall, the mean CCT readings obtained by the USP, following CXL, were similar to those obtained by the Cirrus HD-OCT throughout the 12-month period, except for the 2-week follow-up visit. The disparity of 6.45 μm less for the Cirrus HD-OCT was statistically significant ( P = .002), albeit with mild clinical relevance. On the other hand, the mean CCT measurements obtained by the Galilei were statistically and clinically significantly lower than those obtained by both the Cirrus HD-OCT and the USP at all follow-up visits ( Table 2 ).



Table 2

Differences in Mean Central Corneal Thicknesses Between Ultrasound Pachymetry, High-Definition Optical Coherence Tomography, and Dual Scheimpflug Calculated at Each Time Point






















































































Modalities CCT Difference (μm)
Baseline 2 Weeks 1 Month 3 Months 6 Months 12 Months
USP minus HD-OCT −0.22 6.45 1.03 3.55 0.19 2.35
95% LoA −15.90 to 15.46 −12.79 to 25.69 [38.48] −18.35 to 20.41 [38.76] −18.72 to 25.80 [44.52] −22.86 to 22.48 [45.34] −17.91 to 22.61 [40.52]
(.002) a (.9) (.19) (.52) (0.12)
USP minus dual Scheimpflug 3.10 35.63 32.82 34.21 22.22 18.30
95% LoA −25.85 to 32.11 −12.26 to 83.5 [95.76] −5.90 to 71.54 [77.54] −4.17 to 72.59 [76.76] −16.96 to 61.39 [78.35] −26.67 to 63.33 [90]
(<.001) a (<.001) a (<.001) a (<.001) a (<.001) a
HD-OCT minus dual Scheimpflug 3.30 29.18 31.79 30.67 22.41 15.97
95% LoA −29.41 to 36.13 −14.73 to 73.07 [87.8] −0.51 to 64.09 [64.6] −6.07 to 67.40 [73.47] −21.91 to 66.72 [88.63] −30.05 to 61.99 [92.04]
(<.001) a (<.001) a (<.001) a (<.001) a (<.001) a

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 6, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Central Corneal Thickness After Cross-linking Using High-Definition Optical Coherence Tomography, Ultrasound, and Dual Scheimpflug Tomography: A Comparative Study Over One Year

Full access? Get Clinical Tree

Get Clinical Tree app for offline access