Precise evaluation of the state of the cornea is vital to refractive surgeons.1 New technology in the field of corneal topography has resulted in more accurate and reliable measurement of corneal curvature, thickness, and elevation.2,3
Since 1995, the Orbscan (Bausch & Lomb) has been the most widely used method for measurement of corneal characteristics. This system assesses the corneal curvature, thickness, and elevation using a scanning-slit beam across the cornea.4 The newer Orbscan IIz system incorporates Placido disk and scanning-slit systems.5 In addition, advanced measurement systems to assess corneal thickness, curvature, and elevation have become available.2,3,6–12 In particular, dual rotating Scheimpflug–Placido (Galilei G2, Ziemer Ophthalmic Systems AG) and swept-source optical coherence tomography (OCT) (Casia SS-1000, Tomey Corp.) systems enable direct imaging of the posterior cornea.6,10,12
Apart from conventional methods using a Placido-disk system, swept-source OCT detects corneal structures using axial images in 16 radial scans within 0.3 second.10 Comparability of keratometry (K) values between the swept-source OCT system and the conventional Placido-disk system have been on the rise because there is increasing interest in applying new techniques to evaluate the cornea.
Several studies report different values and agreement between the newly developed systems and Placido–scanning-slit system. In these studies,10,12 more accurate or reliable data were obtained using the dual rotating Scheimpflug–Placido and swept-source OCT systems than using the Placido–scanning-slit system.
The present study sought to compare and calculate agreement between 3 corneal topography devices; that is, the Galilei G2 dual rotating Scheimpflug–Placido system, the Casia SS-1000 swept-source OCT system, and the Orbscan IIz Placido–scanning-slit system.
Subjects and methods
This prospective cross-sectional study comprised healthy subjects from the Department of Ophthalmology, Kangbuk Samsung Hospital, Seoul, South Korea. The study adhered to the tenets of the Declaration of Helsinki and was approved by the local ethics committee. All participants provided written informed consent.
Inclusion criteria were healthy individuals aged 18 to 40 years with a spherical equivalent ranging from +1.00 to −6.00 diopters (D). The corrected visual acuity was 0.00 logMAR in all eyes. Exclusion criteria were a history of ocular pathology, trauma, contact lens wear, pregnancy, systemic or local medications, and surgery, with the exception of laser refractive surgery for myopia. In addition, patients with astigmatism of more than 2.50 D, K values of more than 47.2 D in any axis, or inferior–superior (I–S) values (difference in I–S keratometry) of more than 1.4 D were excluded from this study.13
One eye from each subject was used for statistical analysis. Eyes were divided into 2 groups according to their history of myopic laser refractive surgery. Eyes with previous refractive surgery comprised the refractive group. Eyes that had not had previous refractive surgery comprised the normal group.
Twenty participants (20 eyes) were included in the calculation of repeatability of the devices. Three successive measurements were performed, and intraclass correlation coefficients (ICCs) were calculated for each measurement between instruments for anterior and posterior K values, posterior elevation, and pachymetry. All measurements in this study were performed by the same person who was experienced in ophthalmic examinations.
All eyes had measurements using dual rotating Scheimpflug–Placido, swept-source OCT, and Placido–scanning-slit systems. Table 1 shows the characteristics and measurement settings used for each system. The same best-fit zone was used for all devices to increase comparability. Also, the same eccentricity calculation region was used for all 3 devices to ensure better unity with the dual rotating Scheimpflug–Placido system. Ultrasound (US) pachymetry (Sonoscan 4000AP, Sonomed, Inc.) was used to obtain a reference thickness of the cornea.
All measurements were performed continuously in individual patients between 10 am and 1 pm to avoid the effects of diurnal variation in corneal indices.14 Ultrasound pachymetry was performed at the end to avoid the effects of corneal applanation on the other examinations.
Dual Rotating Scheimpflug–Placido System
The Galilei G2 uses 2 cameras in opposite positions in combination with a Placido-disk topographer to analyze the shape of the cornea. Double rotating systems prevent and compensate for error with oblique angle imaging. By detecting the edge provided by dual-Scheimpflug images, the shape of the posterior cornea can be measured. The total scanning time was approximately 0.75 second, and more than 122 000 points were scanned.
Posterior central elevation was measured at the center of the image. Maximum posterior elevation was measured at the most elevated point within the 3.0 mm central zone above the best-fit sphere (BFS).
Swept-Source Optical Coherence Tomography System
The Casia SS-1000 system uses 3-dimensional OCT images generated by a high-speed scanning system with a long-wave laser source. Sixteen radial scan images of less than 10 μm resolution were detected and reassembled for analysis of the anterior and posterior corneal shape. The total scan time was 0.3 second. The instrument was equipped with an auto-alignment function and an auto-shot function that automatically initiated measurement when the subject’s eyes were within the proper range. Posterior corneal elevation and maximum posterior elevation were measured in the same manner using the dual rotating Scheimpflug–Placido system.
The Orbscan IIz uses a scanning-slit image combined with a Placido disk for analysis of corneal elevation, curvature, and pachymetry. The video imaging system performs 20 vertical light slits from the right and 20 slits from the left that are then projected at a 45-degree angle. The duration of the scan was 0.75 second, and each scan was performed twice per examination. Data obtained in the 3.0 mm zone was used to maximize comparability. Posterior corneal elevation, maximum posterior elevation, and eccentricities of the cornea were measured in the same manner as with the other devices.
Sonoscan 4000AP System
Ultrasound pachymetry was performed with a calibrated US probe using the method generally used with this technique. All measurements were taken 5 times, and the average values were calculated.
Data analysis was performed using SPSS software (version 18.0, SPSS, Inc.). Normal distribution was evaluated using the Kolmogorov-Smirnov test. All corneal measurement data exhibited a normal distribution (P > .05). To compare measurements between 2 devices, paired-sample t tests were performed. The ICC was calculated to analyze the repeatability and agreement of the results. A P value less than 0.05 was considered statistically significant, and ICC values higher than 0.900 were regarded as having a high degree of agreement. Bland-Altman plots were created with Stata software (version 9.2, Stata Corp. LP) and used to evaluate the agreement in corneal eccentricity values between devices.
The study enrolled 100 eyes of 100 subjects. Fifty eyes were placed in the normal group and 50 in the refractive group. In the refractive group, 1 subject with a history of laser refractive surgery within 1 year was excluded. There was no statistical difference in sex or age between the 2 groups. As expected, refractive errors were statistically significantly smaller in the refractive group (Table 2).
Of the 20 eyes in the repeatability examinations, 12 had a history of refractive surgery for myopia.
The repeatability of the dual rotating Scheimpflug–Placido system, swept-source OCT system, and Placido–scanning-slit system were excellent for anterior average K (ICC = 0.999, ICC = 1.000, and ICC = 0.996, respectively), posterior average K (ICC = 0.982, ICC = 0.984, and ICC = 0.945, respectively), maximum posterior elevation (ICC = 0.945, ICC = 0.955, and ICC = 0.943, respectively), and pachymetry (ICC = 0.999, ICC = 1.000, and ICC = 0.995, respectively). In addition, US pachymetry showed a high degree of repeatability (ICC = 0.998).
Dual Rotating Scheimpflug–Placido and Swept-Source Optical Coherence Tomography
Table 3 compares the measurements of the corneal indices between the dual rotating Scheimpflug–Placido and swept-source OCT systems. Anterior and posterior K, BFS radius, posterior elevation, and pachymetry were statistically significantly different between the 2 devices in both groups (P < .05), with the exception of anterior average K in the normal group (P = .233). The ICC values between the 2 devices were high for anterior and posterior K, BFS radius, and pachymetry in both groups (ICC > 0.900). Posterior corneal elevation and maximum posterior elevation did not show high ICC values between the 2 systems. The mean values of maximum posterior elevation were lower with the dual rotating Scheimpflug–Placido system than with swept-source OCT system in both groups.
Dual Rotating Scheimpflug–Placido and Placido–Scanning-Slit Systems
Table 4 compares the measurements of corneal indices between the dual rotating Scheimpflug–Placido and Placido–slit-scanning systems. Anterior and posterior K, BFS radius, posterior elevation, and pachymetry were statistically significantly different between the 2 devices in both groups (P < .05). The ICC values between the 2 devices were high for anterior K, BFS radius, and pachymetry in the both groups (ICC > 0.900). Posterior K did not show a high ICC in either group. Posterior corneal elevation and maximum posterior elevation showed very low ICC values in both groups. The mean maximum posterior elevation values were much lower with the dual rotating Scheimpflug–Placido system than with the Placido–scanning-slit system in both groups.
Swept-Source Optical Coherence Tomography and Placido–Scanning-Slit Systems
Table 5 compares the measurements of corneal indices between the swept-source OCT and Placido–scanning-slit systems. Anterior and posterior K, BFS radius, posterior elevation, and pachymetry showed statistically significant differences between the 2 devices in both groups (P < .05), with the exception of thinnest pachymetry in the normal and refractive groups (P =.138 and P =.586, respectively). The ICC values were high for anterior K, BFS radius, and pachymetry (ICC > 0.900) in the refractive group and the normal group. Posterior K did not show high ICC values in either group. Posterior corneal elevation and maximum posterior elevation had very low ICC values in both groups. The mean maximum posterior elevation values were much lower with the swept-source OCT system than with Placido–scanning-slit system in both groups.
Posterior Average Keratometry and Maximum Posterior Elevation
Table 6 shows the posterior average K ICC values. Table 7 shows the ICC values for the maximum posterior elevation of the cornea above the BFS. The dual rotating Scheimpflug–Placido and swept-source OCT systems showed high ICC values for posterior average K, although the Placido–scanning-slit system showed much steeper values than those 2 devices as well as low ICC values. The dual rotating Scheimpflug–Placido system showed slightly steeper values than the swept-source OCT system; however, the difference was within ±0.1 D (Figure 1). Maximum posterior elevation was statistically significantly different and ICC values were lower than 0.900 between the 3 devices. The mean maximum posterior elevation was highest for the Placido–scanning-slit system followed by the swept-source OCT system and, finally, the dual rotating Scheimpflug–Placido system (Figure 2).
Corneal Center Thickness
Table 8 shows the P values and ICC values for the central corneal thickness. In the refractive group and the normal group, the corneal center was thickest with the dual rotating Scheimpflug–Placido system followed by US pachymetry, then the Placido–scanning-slit system, and finally the SS-OCT system. The dual rotating Scheimpflug–Placido system, swept-source OCT system, Placido–scanning-slit system, and US pachymetry system showed excellent agreement for corneal center thickness measurements (all ICC > 0.900) (Figure 3).
Table 9 shows the corneal eccentricity measurements of the 3 systems. Anterior corneal eccentricity showed high ICC values between the 3 devices in the normal group (ICC > 0.900) but not in the refractive group (ICC ≤ 0.900). Posterior corneal eccentricity showed high ICC values between the dual rotating Scheimpflug–Placido system and the swept-source OCT system. The Placido–scanning-slit system showed low ICC values with the dual rotating Scheimpflug–Placido and swept-source OCT systems in both groups. Bland-Altman plots showed better agreement between the dual rotating Scheimpflug–Placido and swept-source OCT systems than with the Placido–scanning-slit system (Figure 4).
It is important to compare and evaluate new measurement techniques to determine whether there is sufficient agreement between new methods and old methods.15 It is also essential to identify characteristics and evaluate measurement results from various types of topography before refractive surgery.2,6,9,11,12 Different results from the same cornea can influence the diagnosis of ocular disease or the measurement of other ocular parameters.1,13,16–19
This prospective cross-sectional study directly examined corneal topography with dual rotating Scheimpflug–Placido (Galilei G2), swept-source OCT (Casia SS-1000), and Placido–scanning-slit (Orbscan IIz) systems. As in previous studies,10,12 corneal measurements obtained using the different devices showed statistically significant differences. However, one should use caution when comparing mean values between 2 devices because these values might appear to be statistically different even with less than 0.1 D, which is smaller than the diurnal cornea variation14,20 or the minimum measurement scale. Second, because there is not a sufficient standard reference, it is hard to determine the accuracy of the devices.15,21 For these reasons, we evaluated and compared quantitative agreement by calculating the ICC between the 3 devices.21
Agreement was high for anterior K, BFS radius, and pachymetry between the 3 systems. Anterior eccentricity was similar in normal eyes but not in post-refractive surgery eyes. Posterior K and eccentricity showed high agreement between the dual rotating Scheimpflug–Placido system and swept-source OCT system but not with the Placido–scanning-slit system.
Sy et al.12 found that the BFS radius, K, and posterior elevation were significantly different between the dual rotating Scheimpflug–Placido and Placido–scanning-slit systems in 78 patients before and after refractive surgery. Posterior K was steeper and posterior elevation was higher with the Placido–scanning-slit system than the dual rotating Scheimpflug–Placido system. Moreover, dual rotating Scheimpflug–Placido measurements suggested a more reliable maximum posterior elevation than the Placido–scanning-slit system because the changes in maximum posterior elevation before and after refractive surgery were much lower with the dual rotating Scheimpflug–Placido device. They wrote that because the Placido–scanning-slit system inferred the posterior cornea through the anterior surface, it could be relatively imprecise. As a result, the same patient could appear to have a much steeper posterior elevation after refractive surgery.
In the present study, we set the same BFS radius region at 10.0 mm and compared the values measured with each device. Although there were statistically significant differences in posterior BFS radius values between the dual rotating Scheimpflug–Placido device and the Placido–scanning-slit system, high agreement (ICC = 0.925 in refractive group; ICC = 0.957 in normal group) was observed. Similar to the previous study,12 posterior K values were steeper and posterior corneal elevation values (and maximum posterior elevation) were higher with the Placido–scanning-slit system than with the dual rotating Scheimpflug–Placido device (P < .001; ICC ≤ 0.900).
In a study by Jhanji et al.,10 corneal thickness and elevation measurements were significantly different between the swept-source OCT and the Placido–scanning-slit systems, with swept-source OCT having better reproducibility coefficients and ICC values. According to this study, swept-source OCT measured approximately 8 μm thinner pachymetry in normal eyes than the Placido–scanning-slit system. Also, the BFS radius in a 9.0 mm region with swept-source OCT was 0.1 mm shorter than that of the Placido–scanning-slit system. Although there was no reference value for topography, the authors directly analyzed the anterior and posterior surfaces with shorter examination times and better reproducibility (P < .001); thus, swept-source OCT could be regarded as a more reliable and useful technology.
In the present study, results were similar between the swept-source OCT and the Placido–scanning-slit systems. The swept-source OCT system measured approximately 2 to 7 μm thinner pachymetry. Repeatability was slightly higher with the swept-source OCT device than with the Placido–scanning-slit device. In addition, posterior corneal measurements (posterior K, elevation, eccentricity) showed poor agreement between the swept-source OCT system and the Placido–scanning-slit system.
In comparing values between the dual rotating Scheimpflug–Placido device and swept-source OCT, most indices showed statistically significant differences, while these 2 systems showed excellent agreement (ICC > 0.900), with the exception of posterior elevation (posterior corneal elevation and maximum posterior elevation).
There was poor agreement in posterior elevation measurements between the dual rotating Scheimpflug–Placido device and swept-source OCT device. The maximum posterior elevation value that was greater by approximately 1.6 μm (1.56 μm in refractive group; 1.76 μm in normal group) was obtained with the swept-source OCT system. However, compared with the Placido–scanning-slit system, the dual rotating Scheimpflug–Placido and swept-source OCT systems gave very similar results for the maximum posterior elevation. The Placido–scanning-slit system showed an approximately 30 μm difference in maximum posterior elevation compared with the other 2 devices.
With respect to posterior keratometry, the dual rotating Scheimpflug–Placido device and swept-source OCT device showed very slight differences (<0.1 D) and high agreement in the flat, steep, and average values (ICC > 0.900). Considering the minimum measurement scale of refraction is typically 0.25 D, this small difference can be regarded as clinically irrelevant. In addition to posterior K, posterior cornea eccentricity also showed high agreement between the dual rotating Scheimpflug–Placido system and swept-source OCT system (ICC = 0.911 in refractive group; ICC = 0.906 in normal group). Thus, the dual rotating Scheimpflug–Placido device and swept-source OCT device detected an equivalent posterior cornea shape and could be considered interchangeable.22
Overall, while the dual rotating Scheimpflug–Placido device and swept-source OCT showed differences in posterior elevation values, eccentricity and K showed high agreement even in different mechanical properties within the same cornea as reference line measurements for each device. Therefore, we concluded that the measurements obtained using the 2 devices were highly comparable.
The major limitation of this study was that the measuring zone of keratometry could not be standardized due to restricted settings in the dual rotating Scheimpflug–Placido device. However, to increase comparability of corneal indices, we unified all configurable settings including the keratometric index, BFS measurement region size and mode, and eccentricity measurement region size.
In conclusion, we believe that this study is the first to compare dual rotating Scheimpflug–Placido and swept-source OCT systems with the Placido–scanning-slit system. In contrast with the Placido–scanning-slit system, the dual rotating Scheimpflug–Placido and swept-source OCT systems showed high agreement, especially for posterior corneal measurements. Maximum posterior elevation values were lower with the dual rotating Scheimpflug–Placido and swept-source OCT systems than with the Placido–scanning-slit system. These results were similar in post-refractive surgery eyes and normal eyes. The swept-source OCT system provided reliable corneal evaluation without the use of a conventional Placido disk or using dual rotating Scheimpflug–Placido technology. Our results may help interpret characteristics and measurement results from various types of topography.
What Was Known
- Topography devices with different techniques are widely used in clinics. Previous studies report different measured values between different devices.
What This Paper Adds
- Comparisons of newly developed devices with high technology are essential.
- Three different topography devices showed excellent agreement in corneal measurements.
- The swept-source OCT system provided reliable values compared with conventional systems using a Placido disk as well as dual Scheimpflug technology.
1. Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assessment for ectasia after corneal refractive surgery. Ophthalmology
2. Konstantopoulos A, Hossain P, Anderson DF. Recent advances in ophthalmic anterior segment imaging: a new era for ophthalmic diagnosis? Br J Ophthalmol. 91, 2007, p. 551-557, Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1994765/pdf/551.pdf
. Accessed January 8, 2015.
3. Swartz T, Marten L, Wang M. Measuring the cornea: the latest developments in corneal topography. Curr Opin Ophthalmol
4. Yaylali V, Kaufman SC, Thompson HW. Corneal thickness measurements with the Orbscan Topography System and ultrasonic pachymetry. J Cataract Refract Surg
5. Cairns G, McGhee CNJ. Orbscan computerized topography: attributes, applications, and limitations. J Cataract Refract Surg
6. Aramberri J, Araiz L, Garcia A, Illarramendi I, Olmos J, Oyanarte I, Romay A, Vigara I. Dual versus single Scheimpflug camera for anterior segment analysis: precision and agreement. J Cataract Refract Surg
7. Auffarth GU, Borkenstein AFM, Ehmer A, Mannsfeld A, Rabsilber TM, Holzer MP. Scheimpflug- und Topographiesysteme in der ophthalmologischen Diagnostik [Scheimpflug and topography systems in ophthalmologic diagnostics]. Ophthalmologe
8. Fukuda R, Usui T, Miyai T, Mori Y, Miyata K, Amano S. Corneal thickness and volume measurements by swept source anterior segment optical coherence tomography in normal subjects. Curr Eye Res
9. Huang J, Feng Y, Wang Q, Pesudovs K. Assessment of corneal thickness measurement using swept-source Fourier-domain anterior segment optical coherence tomography and Scheimpflug camera [letter]. J Cataract Refract Surg
10. Jhanji V, Yang B, Yu M, Ye C, Leung CKS. Corneal thickness and elevation measurements using swept-source optical coherence tomography and slit scanning topography in normal and keratoconic eyes. Clin Exp Ophthalmol
11. Quisling S, Sjoberg S, Zimmerman B, Goins K, Sutphin J. Comparison of Pentacam and Orbscan IIz on posterior curvature topography measurements in keratoconus eyes. Ophthalmology
12. Sy ME, Ramirez-Miranda A, Zarei-Ghanavati S, Engle J, Danesh J, Hamilton DR. Comparison of posterior corneal imaging before and after LASIK using dual rotating Scheimpflug and scanning slit-beam corneal tomography systems. J Refract Surg
13. Romero-Jiménez M, Santodomingo-Rubido J, Wolffsohn JS. Keratoconus: a review. Cont Lens Anterior Eye
. 2010;33:157-166. quiz 205.
14. Lattimore MR Jr, Kaupp S, Schallhorn S, Lewis RIV. Orbscan pachymetry; implications of a repeated measures and diurnal variation analysis. Ophthalmology
15. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1, 1986, p. 307-310, Available at: http://www-users.york.ac.uk/˜mb55/meas/ba.pdf
. Accessed January 8, 2015.
16. Seiler T, Koufala K, Richter G. Iatrogenic keratectasia after laser in situ keratomileusis. J Refract Surg
17. Geggel HS, Talley AR. Delayed onset keratectasia following laser in situ keratomileusis. J Cataract Refract Surg
18. Doughty MJ, Zaman ML. Human corneal thickness and its impact on intraocular pressure measures: a review and meta-analysis approach. Surv Ophthalmol
19. Doughty MJ, Jonuscheit S, Button NF. Central corneal thickness and intraocular pressure measures in human corneas with endothelial guttata: an observational quality control study. Clin Exp Optom. 94, 2011, p. 425-432, Available at: http://onlinelibrary.wiley.com/doi/10.1111/j.1444-0938.2011.00584.x/pdf
. Accessed January 8, 2015.
20. Read SA, Collins MJ, Carney LG. The diurnal variation of corneal topography and aberrations. Cornea
21. Costa-Santos C, Bernardes J, Ayres-de-Campos D, Costa A, Costa C. The limits of agreement and the intraclass correlation coefficient may be inconsistent in the interpretation of agreement. J Clin Epidemiol
22. Gatinel D, Malet J, Hoang-Xuan T, Azar DT. Corneal elevation topography: best fit sphere, elevation distance, asphericity, toricity, and clinical implications. Cornea