ARTICLE
Myopic Laser Corneal Refractive Surgery Reduces Interdevice Agreement in the Measurement of Anterior Corneal Curvature Haiying Jin,
M.D.,
Zhongmin Ou,
M.D.,
Objectives: To investigate interdevice differences and agreement in the measurement of anterior corneal curvature obtained by different technologies after laser corneal refractive surgery. Methods: The prospective study comprised 109 eyes of 109 consecutive patients who had undergone laser-assisted in situ keratomileusis (LASIK). Preoperative and postoperative corneal parameters were measured by Scheimpflug imaging (Pentacam), Placido-slit-scanning (Orbscan) and auto-keratometry (IOLMaster). Preoperative and postoperative anterior corneal curvatures (K readings) were compared between devices. Interdevice agreement was evaluated by Bland–Altman analysis. Results: Preoperatively, the difference of K reading for Pentacam–IOLMaster (0.0460.20 D) was not statistically significant (P¼0.059). The differences between Pentacam–Orbscan and Orbscan–IOLMaster were 0.2060.34 D (P,0.001) and 20.1760.29 D (P,0.001), respectively. After surgery, no difference was found for Pentacam–Orbscan (20.0560.38, P¼0.136). The differences between Pentacam–IOLMaster and Orbscan–IOLMaster were 0.1360.29 D (P,0.001) and 0.1960.34 D (P,0.001). Preoperative interdevice agreement (95% limit of agreement [LOA]) between Pentacam and Orbscan, Pentacam and IOLMaster, and Orbscan and IOLMaster were 1.31 D, 0.79 D and 1.14 D, respectively. The 95% LOAs decreased to 1.47 D, 1.14 D, and 1.34 D after refractive surgery. Conclusion: Corneal refractive surgery changed the preoperative and postoperative interdevice differences in corneal curvature measurements and reduced interdevice agreement, indicating that the devices are not interchangeable. Key Words: Laser corneal refractive surgery—Anterior corneal curvature measurement—Interdevice agreement—Corneal power. (Eye & Contact Lens 2017;0: 1–7)
From the Department of Ophthalmology (H.J., P.Z.), Xinhua Hospital Affiliated to Medical College of Shanghai Jiaotong University, Shanghai, China; Department of Physics (Z.O.), Shanghai University, Shanghai, China; and Department of Ophthalmology (H.G.), Aier Eye Hospital, Zhengzhou, China. The authors have no conflict of interests to disclose. Supported in part by National Natural Science Foundation of China (81100655). This research was not supported by the funding from any of the following organizations: National Institutes of Health (NIH); Wellcome Trust; Howard Hughes Medical Institute (HHMI); and other(s). Address correspondence to Peiquan Zhao, M.D., Department of Ophthalmology, Xinhua Hospital Affiliated to Medical College of Shanghai Jiaotong University, Kongjiang Road, Shanghai 200092, China; e-mail: [email protected] Accepted December 12, 2016. Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists. This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal. DOI: 10.1097/ICL.0000000000000364
Eye & Contact Lens Volume 0, Number 0, Month 2017
Haike Guo,
M.D.,
and Peiquan Zhao,
M.D.
M
isevaluation of corneal power is one of the principal sources of error in intraocular lens (IOL) power miscalculation after corneal laser refractive surgery.1–15 Methods for estimating actual corneal power after refractive surgery fall into two categories, the historical method1 and the nonhistorical methods.2–15 The latter includes evaluation of corneal power using devices that can measure both anterior and posterior corneal surfaces,4–6 the ray-tracing method,6–8 and corrective algorithms calculating the corneal power from measured anterior corneal curvature (or keratometric power), including the Shammas method,9,10 modified Maloney method,11–14 and Haigis-L formula.15 Because anterior corneal radius (corneal curvature) is an essential parameter in the nonhistorical methods, measurement precision of the anterior corneal curvature after refractive surgery is paramount in corneal power estimation. With alterations to the corneal anterior curvature and asphericity induced by laser ablation,16 it is necessary to evaluate the comparability and agreement of corneal curvature measured by different devices after corneal refractive surgery. A number of publications have compared corneal curvature measured by different instruments and technologies such as autokeratometry, Placido-based computerized topography, Placido-slitscanning, Scheimpflug cameras, dual Scheimpflug devices and point-source color light-emitting diode-based (LED) topography, and swept-source optical coherence tomography in patients before or after cataract surgery, in candidates before refractive surgery and in postrefractive cases.17–29 For postrefractive cases, comparisons of refractive changes and keratometric changes (with a keratometric index of 1.3375) measured by keratometry or computerized topography after refractive surgery have also been studied.30,31 However, to the best of our knowledge, very few manuscripts have reported the influence of refractive surgery on the measurement of anterior corneal curvature concerning interdevice differences and agreement based on different measuring technologies using the same group of patients by a prospective study.17 This study had two aims. First, to compare different devices and technologies, including a Scheimpflug camera, Placido-slit-scanning computerized topography and auto-keratometry, with respect to anterior corneal curvature measurements before and after corneal refractive surgery and to evaluate the variations in interdevice differences induced by refractive surgery. Second, to evaluate the influence of refractive surgery on interdevice agreement.
PATIENTS AND METHODS Consecutive patients who were scheduled for laser-assisted In situ Keratomileusis (LASIK) at Xinhua Hospital Affiliated to Medical College, Shanghai Jiaotong University, China, were invited to participate. The study followed the tenets of the Declaration of 1
Eye & Contact Lens Volume 0, Number 0, Month 2017
H. Jin et al. Helsinki was approved by the ethics committee of Xinhua Hospital Affiliated to Medical College, Shanghai Jiaotong University. Informed consent was obtained from the subjects after the nature of procedure(s) had been explained. Inclusion criteria were healthy individuals aged 18 to 40 years with a spherical equivalent ranging from 21.00 to 29.00 diopters (D). The corrected visual acuities were 20/20 in all eyes. Exclusion criteria included corneal diseases, previous corneal surgery, dry eye with poor tear film, and retinal diseases. One eye of each patient was randomly selected using a predetermined computer-generated randomization schedule generated by SPSS software (version 15.0, SPSS Inc., Chicago, IL). A single experienced examiner performed three sets of measurements per device for each eye in a random order. The three devices were rotating Scheimpflug camera (Pentacam, Oculus Optikgeräte GmbH, Wetzlar, Germany), Placido-slit-scanning (Orbscan IIz, Bausch & Lomb-Obtek, Inc., Salt Lake City, UT) and auto-keratometry (IOLMaster, Carl Zeiss Meditec AG, Jena, Germany). To avoid misevaluations because of poor quality of examinations induced by different devices, both preoperative and postoperative examinations were performed immediately after blinking and were carefully inspected before being included in the study. In the event that any device was unable to obtain a scan of acceptable quality after three attempts, the fellow eye of the patient was used instead. For Pentacam and Orbscan, simulated K values (K readings) at the 3 mm ring were obtained. K readings at the 2.3 mm ring and axial length were measured by IOLMaster. Tropicamide Phenylephrine Eye Drops (Santen, Osaka, Japan) was administrated for cycloplegic refraction determination. After examinations of autorefraction and retinoscopy, manifest refraction was obtained by refinements of spherical power, cylinder power and axis. Laserassisted in situ keratomileusis (LASIK) was performed by one of us (J.H.). Corneal parameters at postoperative 12 months were measured using the three devices. Manifest refraction 12 months after surgery was obtained. Theoretically, anterior corneal power or anterior corneal radius should be compared for corneal curvature measurement. However, keratometric power calculated using a corneal radius with a keratometric index of 1.3375 has been widely used in previous publications and is adopted in the Shammas method and modified Maloney method for corneal power estimation after refractive surgery.9–15 K readings calculated with an index of 1.3375 were therefore used for corneal curvature comparisons in our research. To analyze the accuracy of the postoperative corneal curvature measured by different devices, a theoretically calculated postoperative K reading was determined based on preoperative measured corneal power and the refractive change at the corneal plane induced by refractive surgery by the following equations:
Calculate the Sphero-Equivalent Refraction (SEQ) at the Corneal Plane From SEQ at Vertex Plane
postop SEQcorneal
plane 5
postop SEQspectacle plane 1 2 0:012 · postop SEQspectacle
plane
Calculate refractive change at the corneal plane (DSEQ): DSEQ 5 postop SEQcorneal
plane 2 preop
SEQcorneal
plane
Calculate Postoperative Anterior Corneal Power As has been proven in previous publications, the change in posterior corneal power induced by laser refractive surgery can be overlooked clinically, and the change in refraction is primarily induced by altering the anterior corneal power.12–14 The calculated postoperative anterior corneal power is calculated from preoperative anterior corneal power (376/337.5·preoperative K) and DSEQ according to the following equation: Calculated postoperative anterior corneal power 376 5 · preoperative K 2 DSEQ 337:5 The calculated postoperative K value is then determined from the calculated anterior corneal power by using an index of 337.5/ 376 according to the following equation: Calculated postoperative K¼337.5/376·calculated anterior corneal power.
Sample Size Estimation for the Comparison of Anterior Corneal Curvature A difference of more than 0.20 D for a K reading measurement was considered clinically relevant.32 In agreement with previously published studies, the standard deviation of the interdevice difference was 0.5 D. Assuming an alpha (two-sided)¼0.05 and a power of 0.9, the required sample size was determined to be 66 (Stata 10.0, Stata Corp., College Station, TX). To increase the power of the study, additional patients were enrolled in this research.
Statistical Analysis Statistical analysis was performed using SPSS for Windows software. The distributions of the dataset (anterior corneal curvature) were checked for normality using Kolmogorov–Smirnov tests. The results indicated that the data were normally distributed (P.0.05). K readings obtained by different methods were compared by paired sample t tests. P values of 0.05 or less were considered statistically significant. Correlation between different methods was determined by linear regression. The Bland–Altman method33 performed by MedCalc (version 15, MedCalc Software bvba, Ostend, Belgium) was used to determine variability in the differences between K values obtained by different methods. Differences between methods (y-axis) were plotted against their mean (x-axis), and 95% limits of agreement (LOA) were determined. The preoperative and postoperative Bland–Altman analyses were plotted in the same coordinate system to identify the change of LOA induced by refractive surgery.
Calculate preoperative spherical equivalent (SEQ¼S+1/2 C) at the corneal plane: preop SEQcorneal
plane
5
preop SEQspectacle plane 1 2 0:012 · preop SEQspectacle
0.012 is the vertex distance in meters (m) Calculate postoperative SEQ at the corneal plane: 2
RESULTS plane
Study Population for Anterior Corneal Curvature Measurement Overall, 109 eyes of 109 patients were enrolled in this study. The demographics of the study population are presented in Table 1. Eye & Contact Lens Volume 0, Number 0, Month 2017
Eye & Contact Lens Volume 0, Number 0, Month 2017 TABLE 1.
Postrefractive Corneal Curvature Measurement
Patient Demographic and Clinical Information Mean6SD/ Number (%)
Parameter
Range
P
Eyes (patients) 109 (109) — Male gender 52 (47.7%) — Right eye 51 (53%) — Age, yrs 25.564.7 19 to 39 Preoperative axial length, mm 25.6461.01 24.01 to 28.46 ,0.001 Postoperative axial length, mm 25.5260.99 23.78 to 28.28 Preoperative refraction 24.5661.76 D 21.0 to 29.0 D ,0.001 Postoperative refraction 0.2160.46 D 20.88 to 1.25 D
D, diopters; mm, millimeters, P value, P value of paired sample t test; SD, standard deviation.
No subject with the fellow eye was used instead because of acceptable quality scanning obtained by the three devices in the measurement.
Preoperative K Readings Measured by Different Devices The mean preoperative K values measured by Pentacam, Orbscan and IOLMaster are presented in Table 2. The difference (0.0460.20 D) between Pentacam and IOLMaster was not statistically significant (P¼0.056, independent t test). The difference between Pentacam and Orbscan was 0.2060.34 D, which was significantly different from zero (P,0.001, independent t test). The difference between Orbscan and IOLMaster was 0.1760.29 D (P,0.001, independent t test) (Table 3). Relatively high correlations between K reading measurements were found for different devices. Correlation coefficients (r) between Pentacam and Orbscan, Pentacam and IOLMaster, and Orbscan and IOLMaster were 0.963 (P,0.001), 0.986 (P,0.001) and 0.973 (P,0.001), respectively (Table 3). Bland–Altman analysis of the 95% LOA showed that Pentacam and IOLMaster exhibited the highest agreement (20.36 D to 0.43 D, size 0.79 D). The 95% LOA between Pentacam and Orbscan ranged from 20.45 D to 0.86 D, size 1.31 D. The 95% LOA between IOLMaster and Orbscan ranged from 20.40 D to 0.74 D, size 1.14 D (Fig. 1).
Postoperative K Readings Measured by Different Devices The mean postoperative K readings measured by Pentacam, Orbscan and IOLMaster were 38.6961.99 D (range 33.83–42.95 D), TABLE 2.
38.7561.91 D (33.55–43.00 D) and 38.5662.03 D (range 34.17–43.32 D), respectively (Table 2). IOLMaster provided flatter K values than Pentacam and Orbscan. The difference between Pentacam and IOLMaster was 0.1360.29 D (P,0.001, independent t test). The difference between Orbscan and IOLMaster was 0.1960.34 (P,0.001, independent t test). The difference between Pentacam and Orbscan was 0.0560.38 D, which was not significantly different (P¼0.136, paired samples t test) (Table 3). Changes of interdevice differences were observed for PentacamOrbscan (0.25 D), Pentacam-IOLMaster (0.09 D), and OrbscanIOLMaster (0.36 D) (Fig. 1). Relatively high correlations were found between different technologies. The correlation coefficients (r) between Pentacam and Orbscan, Pentacam and IOLMaster, and Orbscan and IOLMaster were 0.981 (P,0.001), 0.989 (P,0.001) and 0.987 (P,0.001), respectively (Table 3). However, interdevice agreement was reduced compared with preoperative measurements. Bland–Altman analysis showed that agreement between Pentacam and Orbscan was moderate (95% LOA 20.79 to 0.68 D, size 1.47 D), Pentacam and IOLMaster generated higher agreement (95% LOA 20.44 to 0.79 D, size 1.14 D), and the 95% LOA between Orbscan and IOLMaster ranged from 20.48 to 0.86 D (size, 1.34 D) (Fig. 1).
Comparison of Measured K and Calculated K After Refractive Surgery To evaluate the reliability of postoperative K readings measured by different devices, the theoretically calculated postoperative K value was determined by using preoperative corneal data and refractive change at the corneal plane. The calculated K values for Pentacam, Orbscan and IOLMaster were 38.7361.95 D (range 34.14–43.54 D), 38.5361.95 D (33.84–43.31 D) and 38.6961.94 D (range 34.17–43.32 D), respectively. Paired t test showed that no difference was found between the measured and calculated Pentacam K values (P¼0.355). Differences were found between the measured and calculated K values for Orbscan (P,0.001, paired t test) and IOLMaster (P¼0.001, paired samples t test) (Table 3). Relatively high correlations were revealed between the calculated and measured K values in different devices. Bland–Altman analysis showed that the 95% LOA between the measured and calculated Pentacam K values ranged from 20.88 to 0.81 D (size 1.69 D), and the 95% LOA between the measured and calculated IOLMaster K values ranged from 20.95 to 0.68 D (size 1.63 D). The agreement between the measured and calculated Orbscan K values (20.79 to 1.22 D size 2.01 D) were lower than those of the other two devices.
K Values Obtained by Different Methods
K Value Preoperative measured K value Pentacam Orbscan IOLMaster Postoperative measured K value Pentacam Orbscan IOLMaster Postoperative calculated K value Pentacam Orbscan IOLMaster
Mean6SD (D)
Range (D)
43.2561.20 43.0461.25 43.2161.21
39.98–46.02 39.65–45.80 39.95–45.81
38.6961.99 38.7461.91 38.5662.03
33.83–42.95 33.55–43.00 33.37–42.67
38.7361.95 38.5361.95 38.6961.94
34.14–43.54 33.84–43.31 34.17–43.32
D, diopters; SD, standard deviation.
DISCUSSION Precise measurements of corneal parameters are of vital clinical importance in both refractive and cataract surgeries. Several studies have compared different devices in the measurement of corneal parameters in preoperative and postoperative cataract patients, healthy participants and candidates before and after refractive surgery17–29 However, very few studies have evaluated the impact of refractive surgery on interdevice differences and agreement based on different measuring technologies.17 A summary of previous studies and the present research comparing K reading measurements with enrolled sample size is provided in Table 4. In the
Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists.3
Eye & Contact Lens Volume 0, Number 0, Month 2017
H. Jin et al. TABLE 3.
Comparisons and Agreements of K Values Obtained by Different Methods Difference Between Devices
K Value Preoperative measured K Pentacam-Orbscan Pentacam-IOLMaster Orbscan-IOLMaster Postoperative measured K Pentacam-Orbscan Pentacam-IOLMaser Orbscan-IOLMaster Postoperative calculated K-measured K Calculated K-measured K (Pentacam) Calculated K-measured K (Orbscan) Calculated K-measured K (IOLMaster)
Agreement Between Devices (95% LOA)
Correlation Between Devices
Mean6SD
Range
P
Range
Size
r
P
0.2060.34 0.0460.20 20.1760.29
20.81 to 1.18 20.54 to 0.52 20.81 to 0.61
,0.001 0.059 ,0.001
20.45 to 0.86 20.36 to 0.43 20.74 to 0.40
1.31 0.79 1.14
0.963 0.986 0.973
,0.001 ,0.001 ,0.001
20.0560.38 0.1360.29 0.1960.34
21.13 to 1.04 20.58 to 1.13 20.61 to 1.24
0.136 ,0.001 ,0.001
20.79 to 0.68 20.44 to 0.70 20.48 to 0.86
1.47 1.14 1.34
0.981 0.989 0.987
,0.001 ,0.001 ,0.001
20.0460.43 0.2260.51 20.1360.42
20.90 to 0.99 20.88 to 1.47 21.42 to 0.91
0.35 ,0.001 0.001
20.88 to 0.81 20.79 to 1.22 20.95 to 0.68
1.69 2.01 1.63
0.975 0.965 0.979
,0.001 ,0.001 ,0.001
D, diopters; LOA, limit of agreement; SD, standard deviation.
present study, the required sample size (66 eyes) was calculated. Moreover, one eye of each patient should be used in a study to avoid bias induced by the symmetry of both eyes. Our research strictly adhered to the study protocol and had a relatively larger sample size that met the required sample size (Table 4). The present study found that before refractive surgery, Pentacam and IOLMaster provided similar corneal curvature measurements. The difference between the two devices was 0.0460.20 D (P¼0.059). The difference has no clinical significance. These two devices also had the highest agreement with a 95% LOA of 0.79 D. The observations in our study resonate with those of most previous studies. Several studies have compared K values measured by Pentacam and IOLMaster and have found similar mean K values and relatively good agreement between the technologies.20–22 Lee et al.23 found that the mean difference (Pentacam-IOLMaster) in K values was 0.05 D, with 95% of measurement differences falling within roughly 61 D (95% LOA from 21.02 to +1.13 D). Other studies have found that the difference between Pentacam and IOLMaster is not statistically significant.20–22 Savini et al.24 found that IOLMaster provided steeper K readings than Pentacam, and the difference (0.3 D) was statistically significant. Several factors may account for the discrepancies between that study and our research and other studies. First, the sample size was 41 eyes, which was smaller than the required sample size. Second, the enrolled patients were elderly cataract patients (Mean age 76.568.4 years), which may have caused the difference in imaging quality in the measurement owing to poor fixation because of an existing cataract and poor tear film because of elder age.34–36 The present study also showed that K readings measured by Orbscan were flatter than Pentacam and IOLMaster before refractive surgery. The difference between Pentacam and Orbscan was 0.2060.34 D (P,0.001) with a 95% LOA ranging from 20.45 to 0.86 (size 1.31 D). The difference between IOLMaster and Orbscan was 0.1760.29 (P,0.001) with a 95% LOA ranging from 20.40 to 0.74 (size 1.14 D). These results are in agreement with a study performed by Tajbakhsh et al.,18 in which a relatively larger sample size (115 eyes of 115 patients) showed that K readings measured by Orbscan were 0.37 D flatter than those measured by Pentacam. The difference was statically significant, and the 95% LOA value was 1.046 D. Crawford et al.19 compared K readings of steep and flat meridians measured by Orbscan and Pentacam sep4
arately with a sample size of 30 eyes (30 patients). K readings measured by Orbscan were flatter than those measured by Pentacam with values of 0.260.5 D (steep axis) and 0.160.5 D (flat axis). The difference in K readings for the steep axis approached statistical significance (0.078). Two other researchers showed dissimilar results. Hashemi and Mehravaran17 found that K readings by Orbscan were steeper than those of Pentacam. Whang et al.25 compared the K values measured by Orbscan and IOLMaster and found that Orbscan provided steeper K readings than IOLMaster (0.6360.78, P¼0.000). The disparities between the two studies and our results and those of other studies may be due to the sample size and characteristics of the investigated populations. Both eyes of 23 patients were enrolled in Hashemi and Mehravaran’s research.17 The sample size was low, and biases may have been introduced with both eyes enrolled. The sample size of Whang et al.’s25 study (69 eyes of 69 patients) was also smaller than in our research. Moreover, the research was a retrospective study and enrolled eyes of patients undergoing cataract surgery. The importance of this study was to compare anterior corneal curvature measurements obtained by different devices and to evaluate the agreement between devices after refractive surgery based on the evaluation of corneal power for IOL calculation after refractive surgery. It is well accepted that misevaluation of corneal power is one of the principal sources of error in IOLpower miscalculation after corneal laser refractive surgery.1–15 The precision of anterior corneal curvature measurement is paramount in the estimation of actual corneal power after refractive surgery in the nonhistorical methods. As there are various instruments based on different technologies for measuring the anterior corneal curvature, it is not clear whether these devices are interchangeable or whether interdevice differences and agreement are similar to preoperative parameters. The present study showed that interdevice differences and agreement in anterior corneal measurement were influenced by refractive surgery. Reduced interdevice agreement was observed after surgery, indicating that corneal measurement precision might decrease because of alteration of the corneal curvature and asphericity induced by refractive surgery. This might be one source of error in IOL power calculations after refractive surgery. Pentacam and IOLMaster showed greater agreement both before and after refractive surgery. The 95% LOA value was 0.79 D before surgery and was 1.14 D after surgery. Although relatively higher agreement between Pentacam and IOLMaster Eye & Contact Lens Volume 0, Number 0, Month 2017
Eye & Contact Lens Volume 0, Number 0, Month 2017
Postrefractive Corneal Curvature Measurement
FIG. 1. Bland–Altman analysis evaluating agreement of measured K values between different devices before and corneal refractive surgery. (A) Agreement between Pentacam and Orbscan; (B) Agreement between Pentacam and IOLMaster; (C) Agreement between Orbscan and IOLMaster. Dots: preoperative interdevice agreement; triangles: Postperative interdevice agreement. The value between mean values shows the change of preoperative and postoperative interdevice difference. The difference between preoperative and postoperative LOAs shows the reduced interdevice LOA induced by refractive surgery.
Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists.5
Eye & Contact Lens Volume 0, Number 0, Month 2017
H. Jin et al. TABLE 4.
First Author Hashemi and Mehravaran17
Tajbakhsh et al.18 Crawford et al.19
No. of Eyes (No. Patients)
Summary of Researches Comparing K Readings Measured by Different Technologies
Instruments
46 (23)
Orbscan vs Pentacam (axial power, 3 mm zone) Orbscan vs Pentacam, (tangential power, 3 mm zone) Orbscan vs Pentacam (after refractive surgery, axial power, 3 mm zone) Orbscan vs Pentacam (after refractive surgery, tangential power, 3 mm zone) 115 (115) Orbscan vs Pentacam
K Value Measured by Device 2 (D) Mean6SD (D)
Difference Between Technologies (D)
P
Range (D)
Size (D)
Correlation (r)
44.5261.67
43.8461.60
0.68
,0.01
20.08 to 1.45
1.53
0.972
44.4361.73
43.7761.68
0.66
,0.01
20.93 to 2.24
3.17
0.888
40.6961.86
40.0561.78
0.63
,0.01
20.23 to 1.5
1.73
0.972
41.3461.69
40.8061.84
0.55
0.02
21.64 to 2.73
4.37
0.804
43.5261.48
43.7961.50
20.37
,0.001
1.046
NA
44.361.9
44.561.8
20.260.5
0.078
20.790 to 0.256 21 to 0.7
1.7
NA
43.361.8
43.461.6
20.160.5
No statistical significance 0.118
21 to 0.9
1.9
NA
20.15 to 0.92
1.07
0.931
NA
NA
NA
0.946
0.180
NA
NA
NA
0.102
NA
NA
NA
0.49 ,0.001
21.02 to 1.13 20.59 to 1.18
2.15 1.77
0.94 20.6870
0.000 ,0.001
NA 20.45 to 0.86
NA 1.31
NA 0.963
Symes and Ursell.20
63 (49)
Orbscan vs Pentacam (Kvalue of steep meridian) Orbscan vs Pentacam (Kvalue of flat meridian) Pentacam vs IOLMaster
Symes et al.21
29 (29)
Pentacam vs IOLMaster
43.5361.736
Saad et al.22
50 (50)
IOLMaster vs Pentacam 3 mm ring (simulated K) IOLMaster vs Pentacam 3 mm zone Pentacam vs IOLMaster IOLMaster vs Pentacam
43.6861.28
43.9361.415 20.11 calculated by author 43.4761.564 0.06 calculated by author 43.7661.31 20.0860.27
43.6861.28
43.7761.33
20.0860.35
43.3161.62 43.9761.44
43.2661.59 43.6761.49
44.6761.53 43.2561.20
44.0361.41 43.0461.25
0.0560.55 0.30 calculated by author 0.6360.78 0.2060.34
43.2561.20
43.2161.21
0.0460.20
0.06
20.36 to 0.43
0.79
0.986
43.0461.25
43.2161.21
20.1760.29
,0.001
20.74 to 0.40
1.14
0.973
38.6961.99
38.7561.91
20.0560.38
0.136
20.79 to 0.68
1.47
0.981
38.6961.99
38.5662.03
0.1360.29
,0.001
20.44 to 0.70
1.14
0.989
38.7561.91
38.5662.03
0.1960.34
,0.001
20.48 to 0.86
1.34
0.987
Lee et al.23 Savini et al.24 Whang et al.25 The present study
30 (30)
95% LOA
K Value Measured by Device 1 Mean6SD (D)
49 (41) 41 (41)
69 (69) Orbscan vs IOLMaster 109 (109) Pentacam vs Orbscan (before refractive surgery) Pentacam vs IOLMaster (before refractive surgery) Orbscan vs IOLMaster (before refractive surgery) Pentacam vs Orbscan (after refractive surgery) Pentacam vs IOLMaster (after refractive surgery) Orbscan vs IOLMaster (after refractive surgery)
43.8261.439
D, diopters; SD, standard deviation.
might indicate the superiority of the two devices in the measurement of anterior corneal curvature after refractive surgery, the change in interdevice difference and reduced agreement indicated that the three devices were not interchangeable. Before surgery, Pentacam and IOLMaster showed similar results, and Orbscan provided flatter K readings than Pentacam and IOLMaster. After refractive surgery, the difference between Pentacam and Orbscan was not statistically significant, and IOLMaster provided slightly steeper K readings than Pentacam and Orbscan. To further investigate the reliability of different devices in corneal curvature measurement after surgery, comparisons between measured and calculated K values after refractive surgery were performed for each device. No difference was found between the measured and calculated K value for Pentacam, whereas the differences between the measured and calculated K values were significantly different 6
for IOLMaster and Orbscan. The differences in K value measurements obtained by different devices may originate from the different measuring principles of the three devices. Several studies have compared K readings measured by different technologies after corneal refractive surgery.26,27 Ventura et al.26 compared K readings measured by a novel point-source color LED-based topographer, a Placido-disk topographer, and a combined Placido-based and dual Scheimpflug device in normal eyes and in postrefractive eyes. No interdevice difference was found between the devices. In another study performed by Lee et al.,27 dual rotating Scheimpflug–Placido, swept-source optical coherence tomography, and Placido-scanning-slit systems were used to compare corneal parameters in normal eyes and in postrefractive cases. Interdevice differences were revealed in both groups. In contrast to our study, the normal and postrefractive eyes were enrolled from Eye & Contact Lens Volume 0, Number 0, Month 2017
Eye & Contact Lens Volume 0, Number 0, Month 2017 two groups of populations in the two studies. Therefore, the influence of refractive surgery on interdevice agreement could not be evaluated. To the best of our knowledge, only one study has compared parameters measured by Pentacam and Orbscan before and after refractive surgery using the same group of patients.17 Changes in both anterior and posterior corneal powers were studied. The research revealed that Orbscan yielded steeper anterior corneal curvature measurements than Pentacam both before and after refractive surgery, which is not in agreement with our result. As has been discussed, the possible reasons for the disparity may be due to the study design. In that study, both eyes of 23 patients were enrolled. Moreover, although preoperative and postoperative interdevice agreements were analyzed respectively, a reduced agreement between them was not observed after surgery. The present research suggests for the first time that myopic laser corneal refractive surgery reduces interdevice agreement in the measurement of anterior corneal curvature. In conclusion, to the best of our knowledge, we believe that this study is the first to observe reduced interdevice agreement after corneal refractive surgery, indicating that devices for anterior corneal curvature measurement are not interchangeable. The reduced agreement might be one source of error in IOL power calculation after corneal refractive surgery. Further studies with more corneal curvature measurement devices and comparisons of their roles in IOL power calculation accuracy after refractive surgery are necessary in our future studies. REFERENCES 1. Holladay JT. Cataract surgery in patients with previous keratorefractive surgery (RK, PRK, and LASIK). Ophthalmic Pract 1997;15: 238–244. 2. Hoffer KJ. Intraocular lens power calculation after previous laser refractive surgery. J Cataract Refract Surg 2009;35:759–765. 3. Seitz B, Langenbucher A. Intraocular lens power calculation in eyes after corneal refractive surgery. J Refract Surg 2000;16:349–361. 4. Langenbucher A, Torres F, Behrens A, et al. Consideration of the posterior corneal curvature for assessment of corneal power after myopic LASIK. Acta Ophthalmol Scand 2004;82(3 Pt 1):264–269. 5. Savini G, Barboni P, Profazio V, et al. Corneal power measurements with the Pentacam Scheimpflug camera after myopic excimer laser surgery. J Cataract Refract Surg 2008;34:809–813. 6. Jin H, Rabsilber T, Ehmer A, et al. Comparison of ray-tracing method and thin-lens formula in intraocular lens power calculations. J Cataract Refract Surg 2009;35:650–662. 7. Canovas C, van der Mooren M, Rosén R, et al. Effect of the equivalent refractive index on intraocular lens power prediction with ray tracing after myopic laser in situ keratomileusis. J Cataract Refract Sur 2015;41: 1030–1037. 8. Savini G, Bedei A, Barboni P, et al. Intraocular lens power calculation by ray-tracing after myopic excimer laser surgery. Am J Ophthalmol 2014;157: 150–153. 9. Shammas HJ, Shammas MC, Garabet A, et al. Correcting the corneal power measurements for intraocular lens power calculations after myopic laser in situ keratomileusis. Am J Ophthalmol 2003;136:426–432. 10. Shammas HJ, Shammas MC. No-history method of intraocular lens power calculation for cataract surgery after myopic laser in situ keratomileusis. J Cataract Refract Surg 2007;33:31–36. 11. Wang L, Booth MA, Koch DD. Comparison of intraocular lens power calculation methods in eyes that have undergone LASIK. Ophthalmology 2004;111:1825–1831. 12. Wang L, Hill WE, Koch DD. Evaluation of intraocular lens power prediction methods using the American society of cataract and refractive surgeons post-keratorefractive intraocular lens power calculator. J Cataract Refract Surg 2010;36:1466–1473.
Postrefractive Corneal Curvature Measurement 13. Jin H, Holzer MP, Rabsilber T, et al. Intraocular lens power calculation after laser refractive surgery: corrective algorithm for corneal power estimation. J Cataract Refract Surg 2010;36:87–96. 14. Jin H, Auffarth GU, Guo H, et al. Corneal power estimation for intraocular lens power calculation after corneal laser refractive surgery in Chinese eyes. J Cataract Refract Surg 2012;38:1749–1757. 15. Haigis W. Intraocular lens calculation after refractive surgery for myopia: Haigis-L formula. J Cataract Refract Surg 2008;34:1658–1663. 16. Gatinel D, Hoang-Xuan T, Azar DT. Determination of corneal asphericity after myopia surgery with the excimer laser: a mathematical model. Invest Ophthalmol Vis Sci 2001;42:1736–1742. 17. Hashemi H, Mehravaran S. Corneal changes after laser refractive surgery for myopia: comparison of Orbscan II and Pentacam findings. J Cataract Refract Surg 2007;33:841–847. 18. Tajbakhsh Z, Salouti R, Nowroozzadeh MH, et al. Comparison of keratometry measurements using the Pentacam HR, the Orbscan IIz, and the TMS-4 topographer. Ophthalmic Physiol Opt 2012;32:539–546. 19. Crawford AZ, Patel DV, McGhee CN. Comparison and repeatability of keratometric and corneal power measurements obtained by Orbscan II, Pentacam, and Galilei corneal tomography systems. Am J Ophthalmol 2013;156:53–60. 20. Symes RJ, Ursell PG. Automated keratometry in routine cataract surgery: comparison of Scheimpflug and conventional values. J Cataract Refract Surg 2011;37:295–301. 21. Symes RJ, Say MJ, Ursell PG. Scheimpflug keratometry versus conventional automated keratometry in routine cataract surgery. J Cataract Refract Surg 2010;36:1107–1114. 22. Saad E, Shammas MC, Shammas HJ. Scheimpflug corneal power measurements for intraocular lens power calculation in cataract surgery. Am J Ophthalmol 2013;156:460–467. 23. Lee BW, Galor A, Feuer WJ, et al. Agreement between Pentacam and IOL master in patients undergoing toric IOL implantation. J Refract Surg 2013; 29:114–120. 24. Savini G, Barboni P, Carbonelli M, et al. Accuracy of Scheimpflug corneal power measurements for intraocular lens power calculation. Cataract Refract Surg 2009;35:1193–1197. 25. Whang WJ, Byun YS, Joo CK. Comparison of refractive outcomes using five devices for the assessment of preoperative corneal power. Clin Exp Ophthalmol 2012;40:425–432. 26. Ventura BV, Wang L, Ali SF, et al. Comparison of corneal power, astigmatism, and wavefront aberration measurements obtained by a point-source color light-emitting diode-based topographer, a Placido-disk topographer, and a combined Placido and dual Scheimpflug device. J Cataract Refract Surg 2015;41:1658–1671. 27. Lee YW, Choi CY, Yoon GY. Comparison of dual rotating ScheimpflugPlacido, swept-source optical coherence tomography, and Placido-scanningslit systems. J Cataract Refract Surg 2015;41:1018–1029. 28. Karimian F, Feizi S, Doozandeh A, et al. Comparison of corneal tomography measurements using Galilei, Orbscan II, and Placido disk-based topographer systems. J Refract Surg 2011;27:502–508. 29. Wang Q, Savini G, Hoffer KJ, et al. A comprehensive assessment of the precision and agreement of anterior corneal power measurements obtained using 8 different devices. PLoS One 2012;7:e45607. 30. Hugger P, Kohnen T, La Rosa FA, et al. Comparison of changes in manifest refraction and corneal power after photorefractive keratectomy. Am J Ophthalmol 2000;129:68–75. 31. Moshirfar M, Christiansen SM, Kim G. Comparison of the ratio of keratometric change to refractive change induced by myopic ablation. J Refract Surg 2012;28:675–682. 32. Visser N, Berendschot TT, Verbakel F, et al. Comparability and repeatability of corneal astigmatism measurements using different measurement technologies. J Cataract Refract Surg 2012;38:1764–1770. 33. Bland J, Altman D. Measuring agreement in method comparison studies. Stat Methods Med Res 1999;8:135–160. 34. Hashemi H, Asgari S, Emamian MH, et al. Age-Related changes in corneal curvature and shape: the shahroud eye cohort study. Cornea 2015;34:1456–1458. 35. Gipson IK. Age-related changes and diseases of the ocular surface and cornea. Invest Ophthalmol Vis Sci 2013;54:ORSF48–ORSF53. 36. Viso E, Rodriguez-Ares MT, Gude F. Prevalence of and associated factors for dry eye in a Spanish adult population (the Salnes Eye Study). Ophthalmic Epidemiol 2009;16:15–21.
Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists.7
Myopic Laser Corneal Refractive Surgery Reduces Interdevice Agreement in the Measurement of Anterior Corneal Curvature Haiying Jin,
M.D.,
Zhongmin Ou,
M.D.,
Objectives: To investigate interdevice differences and agreement in the measurement of anterior corneal curvature obtained by different technologies after laser corneal refractive surgery. Methods: The prospective study comprised 109 eyes of 109 consecutive patients who had undergone laser-assisted in situ keratomileusis (LASIK). Preoperative and postoperative corneal parameters were measured by Scheimpflug imaging (Pentacam), Placido-slit-scanning (Orbscan) and auto-keratometry (IOLMaster). Preoperative and postoperative anterior corneal curvatures (K readings) were compared between devices. Interdevice agreement was evaluated by Bland–Altman analysis. Results: Preoperatively, the difference of K reading for Pentacam–IOLMaster (0.0460.20 D) was not statistically significant (P¼0.059). The differences between Pentacam–Orbscan and Orbscan–IOLMaster were 0.2060.34 D (P,0.001) and 20.1760.29 D (P,0.001), respectively. After surgery, no difference was found for Pentacam–Orbscan (20.0560.38, P¼0.136). The differences between Pentacam–IOLMaster and Orbscan–IOLMaster were 0.1360.29 D (P,0.001) and 0.1960.34 D (P,0.001). Preoperative interdevice agreement (95% limit of agreement [LOA]) between Pentacam and Orbscan, Pentacam and IOLMaster, and Orbscan and IOLMaster were 1.31 D, 0.79 D and 1.14 D, respectively. The 95% LOAs decreased to 1.47 D, 1.14 D, and 1.34 D after refractive surgery. Conclusion: Corneal refractive surgery changed the preoperative and postoperative interdevice differences in corneal curvature measurements and reduced interdevice agreement, indicating that the devices are not interchangeable. Key Words: Laser corneal refractive surgery—Anterior corneal curvature measurement—Interdevice agreement—Corneal power. (Eye & Contact Lens 2017;0: 1–7)
From the Department of Ophthalmology (H.J., P.Z.), Xinhua Hospital Affiliated to Medical College of Shanghai Jiaotong University, Shanghai, China; Department of Physics (Z.O.), Shanghai University, Shanghai, China; and Department of Ophthalmology (H.G.), Aier Eye Hospital, Zhengzhou, China. The authors have no conflict of interests to disclose. Supported in part by National Natural Science Foundation of China (81100655). This research was not supported by the funding from any of the following organizations: National Institutes of Health (NIH); Wellcome Trust; Howard Hughes Medical Institute (HHMI); and other(s). Address correspondence to Peiquan Zhao, M.D., Department of Ophthalmology, Xinhua Hospital Affiliated to Medical College of Shanghai Jiaotong University, Kongjiang Road, Shanghai 200092, China; e-mail: [email protected] Accepted December 12, 2016. Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists. This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal. DOI: 10.1097/ICL.0000000000000364
Eye & Contact Lens Volume 0, Number 0, Month 2017
Haike Guo,
M.D.,
and Peiquan Zhao,
M.D.
M
isevaluation of corneal power is one of the principal sources of error in intraocular lens (IOL) power miscalculation after corneal laser refractive surgery.1–15 Methods for estimating actual corneal power after refractive surgery fall into two categories, the historical method1 and the nonhistorical methods.2–15 The latter includes evaluation of corneal power using devices that can measure both anterior and posterior corneal surfaces,4–6 the ray-tracing method,6–8 and corrective algorithms calculating the corneal power from measured anterior corneal curvature (or keratometric power), including the Shammas method,9,10 modified Maloney method,11–14 and Haigis-L formula.15 Because anterior corneal radius (corneal curvature) is an essential parameter in the nonhistorical methods, measurement precision of the anterior corneal curvature after refractive surgery is paramount in corneal power estimation. With alterations to the corneal anterior curvature and asphericity induced by laser ablation,16 it is necessary to evaluate the comparability and agreement of corneal curvature measured by different devices after corneal refractive surgery. A number of publications have compared corneal curvature measured by different instruments and technologies such as autokeratometry, Placido-based computerized topography, Placido-slitscanning, Scheimpflug cameras, dual Scheimpflug devices and point-source color light-emitting diode-based (LED) topography, and swept-source optical coherence tomography in patients before or after cataract surgery, in candidates before refractive surgery and in postrefractive cases.17–29 For postrefractive cases, comparisons of refractive changes and keratometric changes (with a keratometric index of 1.3375) measured by keratometry or computerized topography after refractive surgery have also been studied.30,31 However, to the best of our knowledge, very few manuscripts have reported the influence of refractive surgery on the measurement of anterior corneal curvature concerning interdevice differences and agreement based on different measuring technologies using the same group of patients by a prospective study.17 This study had two aims. First, to compare different devices and technologies, including a Scheimpflug camera, Placido-slit-scanning computerized topography and auto-keratometry, with respect to anterior corneal curvature measurements before and after corneal refractive surgery and to evaluate the variations in interdevice differences induced by refractive surgery. Second, to evaluate the influence of refractive surgery on interdevice agreement.
PATIENTS AND METHODS Consecutive patients who were scheduled for laser-assisted In situ Keratomileusis (LASIK) at Xinhua Hospital Affiliated to Medical College, Shanghai Jiaotong University, China, were invited to participate. The study followed the tenets of the Declaration of 1
Eye & Contact Lens Volume 0, Number 0, Month 2017
H. Jin et al. Helsinki was approved by the ethics committee of Xinhua Hospital Affiliated to Medical College, Shanghai Jiaotong University. Informed consent was obtained from the subjects after the nature of procedure(s) had been explained. Inclusion criteria were healthy individuals aged 18 to 40 years with a spherical equivalent ranging from 21.00 to 29.00 diopters (D). The corrected visual acuities were 20/20 in all eyes. Exclusion criteria included corneal diseases, previous corneal surgery, dry eye with poor tear film, and retinal diseases. One eye of each patient was randomly selected using a predetermined computer-generated randomization schedule generated by SPSS software (version 15.0, SPSS Inc., Chicago, IL). A single experienced examiner performed three sets of measurements per device for each eye in a random order. The three devices were rotating Scheimpflug camera (Pentacam, Oculus Optikgeräte GmbH, Wetzlar, Germany), Placido-slit-scanning (Orbscan IIz, Bausch & Lomb-Obtek, Inc., Salt Lake City, UT) and auto-keratometry (IOLMaster, Carl Zeiss Meditec AG, Jena, Germany). To avoid misevaluations because of poor quality of examinations induced by different devices, both preoperative and postoperative examinations were performed immediately after blinking and were carefully inspected before being included in the study. In the event that any device was unable to obtain a scan of acceptable quality after three attempts, the fellow eye of the patient was used instead. For Pentacam and Orbscan, simulated K values (K readings) at the 3 mm ring were obtained. K readings at the 2.3 mm ring and axial length were measured by IOLMaster. Tropicamide Phenylephrine Eye Drops (Santen, Osaka, Japan) was administrated for cycloplegic refraction determination. After examinations of autorefraction and retinoscopy, manifest refraction was obtained by refinements of spherical power, cylinder power and axis. Laserassisted in situ keratomileusis (LASIK) was performed by one of us (J.H.). Corneal parameters at postoperative 12 months were measured using the three devices. Manifest refraction 12 months after surgery was obtained. Theoretically, anterior corneal power or anterior corneal radius should be compared for corneal curvature measurement. However, keratometric power calculated using a corneal radius with a keratometric index of 1.3375 has been widely used in previous publications and is adopted in the Shammas method and modified Maloney method for corneal power estimation after refractive surgery.9–15 K readings calculated with an index of 1.3375 were therefore used for corneal curvature comparisons in our research. To analyze the accuracy of the postoperative corneal curvature measured by different devices, a theoretically calculated postoperative K reading was determined based on preoperative measured corneal power and the refractive change at the corneal plane induced by refractive surgery by the following equations:
Calculate the Sphero-Equivalent Refraction (SEQ) at the Corneal Plane From SEQ at Vertex Plane
postop SEQcorneal
plane 5
postop SEQspectacle plane 1 2 0:012 · postop SEQspectacle
plane
Calculate refractive change at the corneal plane (DSEQ): DSEQ 5 postop SEQcorneal
plane 2 preop
SEQcorneal
plane
Calculate Postoperative Anterior Corneal Power As has been proven in previous publications, the change in posterior corneal power induced by laser refractive surgery can be overlooked clinically, and the change in refraction is primarily induced by altering the anterior corneal power.12–14 The calculated postoperative anterior corneal power is calculated from preoperative anterior corneal power (376/337.5·preoperative K) and DSEQ according to the following equation: Calculated postoperative anterior corneal power 376 5 · preoperative K 2 DSEQ 337:5 The calculated postoperative K value is then determined from the calculated anterior corneal power by using an index of 337.5/ 376 according to the following equation: Calculated postoperative K¼337.5/376·calculated anterior corneal power.
Sample Size Estimation for the Comparison of Anterior Corneal Curvature A difference of more than 0.20 D for a K reading measurement was considered clinically relevant.32 In agreement with previously published studies, the standard deviation of the interdevice difference was 0.5 D. Assuming an alpha (two-sided)¼0.05 and a power of 0.9, the required sample size was determined to be 66 (Stata 10.0, Stata Corp., College Station, TX). To increase the power of the study, additional patients were enrolled in this research.
Statistical Analysis Statistical analysis was performed using SPSS for Windows software. The distributions of the dataset (anterior corneal curvature) were checked for normality using Kolmogorov–Smirnov tests. The results indicated that the data were normally distributed (P.0.05). K readings obtained by different methods were compared by paired sample t tests. P values of 0.05 or less were considered statistically significant. Correlation between different methods was determined by linear regression. The Bland–Altman method33 performed by MedCalc (version 15, MedCalc Software bvba, Ostend, Belgium) was used to determine variability in the differences between K values obtained by different methods. Differences between methods (y-axis) were plotted against their mean (x-axis), and 95% limits of agreement (LOA) were determined. The preoperative and postoperative Bland–Altman analyses were plotted in the same coordinate system to identify the change of LOA induced by refractive surgery.
Calculate preoperative spherical equivalent (SEQ¼S+1/2 C) at the corneal plane: preop SEQcorneal
plane
5
preop SEQspectacle plane 1 2 0:012 · preop SEQspectacle
0.012 is the vertex distance in meters (m) Calculate postoperative SEQ at the corneal plane: 2
RESULTS plane
Study Population for Anterior Corneal Curvature Measurement Overall, 109 eyes of 109 patients were enrolled in this study. The demographics of the study population are presented in Table 1. Eye & Contact Lens Volume 0, Number 0, Month 2017
Eye & Contact Lens Volume 0, Number 0, Month 2017 TABLE 1.
Postrefractive Corneal Curvature Measurement
Patient Demographic and Clinical Information Mean6SD/ Number (%)
Parameter
Range
P
Eyes (patients) 109 (109) — Male gender 52 (47.7%) — Right eye 51 (53%) — Age, yrs 25.564.7 19 to 39 Preoperative axial length, mm 25.6461.01 24.01 to 28.46 ,0.001 Postoperative axial length, mm 25.5260.99 23.78 to 28.28 Preoperative refraction 24.5661.76 D 21.0 to 29.0 D ,0.001 Postoperative refraction 0.2160.46 D 20.88 to 1.25 D
D, diopters; mm, millimeters, P value, P value of paired sample t test; SD, standard deviation.
No subject with the fellow eye was used instead because of acceptable quality scanning obtained by the three devices in the measurement.
Preoperative K Readings Measured by Different Devices The mean preoperative K values measured by Pentacam, Orbscan and IOLMaster are presented in Table 2. The difference (0.0460.20 D) between Pentacam and IOLMaster was not statistically significant (P¼0.056, independent t test). The difference between Pentacam and Orbscan was 0.2060.34 D, which was significantly different from zero (P,0.001, independent t test). The difference between Orbscan and IOLMaster was 0.1760.29 D (P,0.001, independent t test) (Table 3). Relatively high correlations between K reading measurements were found for different devices. Correlation coefficients (r) between Pentacam and Orbscan, Pentacam and IOLMaster, and Orbscan and IOLMaster were 0.963 (P,0.001), 0.986 (P,0.001) and 0.973 (P,0.001), respectively (Table 3). Bland–Altman analysis of the 95% LOA showed that Pentacam and IOLMaster exhibited the highest agreement (20.36 D to 0.43 D, size 0.79 D). The 95% LOA between Pentacam and Orbscan ranged from 20.45 D to 0.86 D, size 1.31 D. The 95% LOA between IOLMaster and Orbscan ranged from 20.40 D to 0.74 D, size 1.14 D (Fig. 1).
Postoperative K Readings Measured by Different Devices The mean postoperative K readings measured by Pentacam, Orbscan and IOLMaster were 38.6961.99 D (range 33.83–42.95 D), TABLE 2.
38.7561.91 D (33.55–43.00 D) and 38.5662.03 D (range 34.17–43.32 D), respectively (Table 2). IOLMaster provided flatter K values than Pentacam and Orbscan. The difference between Pentacam and IOLMaster was 0.1360.29 D (P,0.001, independent t test). The difference between Orbscan and IOLMaster was 0.1960.34 (P,0.001, independent t test). The difference between Pentacam and Orbscan was 0.0560.38 D, which was not significantly different (P¼0.136, paired samples t test) (Table 3). Changes of interdevice differences were observed for PentacamOrbscan (0.25 D), Pentacam-IOLMaster (0.09 D), and OrbscanIOLMaster (0.36 D) (Fig. 1). Relatively high correlations were found between different technologies. The correlation coefficients (r) between Pentacam and Orbscan, Pentacam and IOLMaster, and Orbscan and IOLMaster were 0.981 (P,0.001), 0.989 (P,0.001) and 0.987 (P,0.001), respectively (Table 3). However, interdevice agreement was reduced compared with preoperative measurements. Bland–Altman analysis showed that agreement between Pentacam and Orbscan was moderate (95% LOA 20.79 to 0.68 D, size 1.47 D), Pentacam and IOLMaster generated higher agreement (95% LOA 20.44 to 0.79 D, size 1.14 D), and the 95% LOA between Orbscan and IOLMaster ranged from 20.48 to 0.86 D (size, 1.34 D) (Fig. 1).
Comparison of Measured K and Calculated K After Refractive Surgery To evaluate the reliability of postoperative K readings measured by different devices, the theoretically calculated postoperative K value was determined by using preoperative corneal data and refractive change at the corneal plane. The calculated K values for Pentacam, Orbscan and IOLMaster were 38.7361.95 D (range 34.14–43.54 D), 38.5361.95 D (33.84–43.31 D) and 38.6961.94 D (range 34.17–43.32 D), respectively. Paired t test showed that no difference was found between the measured and calculated Pentacam K values (P¼0.355). Differences were found between the measured and calculated K values for Orbscan (P,0.001, paired t test) and IOLMaster (P¼0.001, paired samples t test) (Table 3). Relatively high correlations were revealed between the calculated and measured K values in different devices. Bland–Altman analysis showed that the 95% LOA between the measured and calculated Pentacam K values ranged from 20.88 to 0.81 D (size 1.69 D), and the 95% LOA between the measured and calculated IOLMaster K values ranged from 20.95 to 0.68 D (size 1.63 D). The agreement between the measured and calculated Orbscan K values (20.79 to 1.22 D size 2.01 D) were lower than those of the other two devices.
K Values Obtained by Different Methods
K Value Preoperative measured K value Pentacam Orbscan IOLMaster Postoperative measured K value Pentacam Orbscan IOLMaster Postoperative calculated K value Pentacam Orbscan IOLMaster
Mean6SD (D)
Range (D)
43.2561.20 43.0461.25 43.2161.21
39.98–46.02 39.65–45.80 39.95–45.81
38.6961.99 38.7461.91 38.5662.03
33.83–42.95 33.55–43.00 33.37–42.67
38.7361.95 38.5361.95 38.6961.94
34.14–43.54 33.84–43.31 34.17–43.32
D, diopters; SD, standard deviation.
DISCUSSION Precise measurements of corneal parameters are of vital clinical importance in both refractive and cataract surgeries. Several studies have compared different devices in the measurement of corneal parameters in preoperative and postoperative cataract patients, healthy participants and candidates before and after refractive surgery17–29 However, very few studies have evaluated the impact of refractive surgery on interdevice differences and agreement based on different measuring technologies.17 A summary of previous studies and the present research comparing K reading measurements with enrolled sample size is provided in Table 4. In the
Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists.3
Eye & Contact Lens Volume 0, Number 0, Month 2017
H. Jin et al. TABLE 3.
Comparisons and Agreements of K Values Obtained by Different Methods Difference Between Devices
K Value Preoperative measured K Pentacam-Orbscan Pentacam-IOLMaster Orbscan-IOLMaster Postoperative measured K Pentacam-Orbscan Pentacam-IOLMaser Orbscan-IOLMaster Postoperative calculated K-measured K Calculated K-measured K (Pentacam) Calculated K-measured K (Orbscan) Calculated K-measured K (IOLMaster)
Agreement Between Devices (95% LOA)
Correlation Between Devices
Mean6SD
Range
P
Range
Size
r
P
0.2060.34 0.0460.20 20.1760.29
20.81 to 1.18 20.54 to 0.52 20.81 to 0.61
,0.001 0.059 ,0.001
20.45 to 0.86 20.36 to 0.43 20.74 to 0.40
1.31 0.79 1.14
0.963 0.986 0.973
,0.001 ,0.001 ,0.001
20.0560.38 0.1360.29 0.1960.34
21.13 to 1.04 20.58 to 1.13 20.61 to 1.24
0.136 ,0.001 ,0.001
20.79 to 0.68 20.44 to 0.70 20.48 to 0.86
1.47 1.14 1.34
0.981 0.989 0.987
,0.001 ,0.001 ,0.001
20.0460.43 0.2260.51 20.1360.42
20.90 to 0.99 20.88 to 1.47 21.42 to 0.91
0.35 ,0.001 0.001
20.88 to 0.81 20.79 to 1.22 20.95 to 0.68
1.69 2.01 1.63
0.975 0.965 0.979
,0.001 ,0.001 ,0.001
D, diopters; LOA, limit of agreement; SD, standard deviation.
present study, the required sample size (66 eyes) was calculated. Moreover, one eye of each patient should be used in a study to avoid bias induced by the symmetry of both eyes. Our research strictly adhered to the study protocol and had a relatively larger sample size that met the required sample size (Table 4). The present study found that before refractive surgery, Pentacam and IOLMaster provided similar corneal curvature measurements. The difference between the two devices was 0.0460.20 D (P¼0.059). The difference has no clinical significance. These two devices also had the highest agreement with a 95% LOA of 0.79 D. The observations in our study resonate with those of most previous studies. Several studies have compared K values measured by Pentacam and IOLMaster and have found similar mean K values and relatively good agreement between the technologies.20–22 Lee et al.23 found that the mean difference (Pentacam-IOLMaster) in K values was 0.05 D, with 95% of measurement differences falling within roughly 61 D (95% LOA from 21.02 to +1.13 D). Other studies have found that the difference between Pentacam and IOLMaster is not statistically significant.20–22 Savini et al.24 found that IOLMaster provided steeper K readings than Pentacam, and the difference (0.3 D) was statistically significant. Several factors may account for the discrepancies between that study and our research and other studies. First, the sample size was 41 eyes, which was smaller than the required sample size. Second, the enrolled patients were elderly cataract patients (Mean age 76.568.4 years), which may have caused the difference in imaging quality in the measurement owing to poor fixation because of an existing cataract and poor tear film because of elder age.34–36 The present study also showed that K readings measured by Orbscan were flatter than Pentacam and IOLMaster before refractive surgery. The difference between Pentacam and Orbscan was 0.2060.34 D (P,0.001) with a 95% LOA ranging from 20.45 to 0.86 (size 1.31 D). The difference between IOLMaster and Orbscan was 0.1760.29 (P,0.001) with a 95% LOA ranging from 20.40 to 0.74 (size 1.14 D). These results are in agreement with a study performed by Tajbakhsh et al.,18 in which a relatively larger sample size (115 eyes of 115 patients) showed that K readings measured by Orbscan were 0.37 D flatter than those measured by Pentacam. The difference was statically significant, and the 95% LOA value was 1.046 D. Crawford et al.19 compared K readings of steep and flat meridians measured by Orbscan and Pentacam sep4
arately with a sample size of 30 eyes (30 patients). K readings measured by Orbscan were flatter than those measured by Pentacam with values of 0.260.5 D (steep axis) and 0.160.5 D (flat axis). The difference in K readings for the steep axis approached statistical significance (0.078). Two other researchers showed dissimilar results. Hashemi and Mehravaran17 found that K readings by Orbscan were steeper than those of Pentacam. Whang et al.25 compared the K values measured by Orbscan and IOLMaster and found that Orbscan provided steeper K readings than IOLMaster (0.6360.78, P¼0.000). The disparities between the two studies and our results and those of other studies may be due to the sample size and characteristics of the investigated populations. Both eyes of 23 patients were enrolled in Hashemi and Mehravaran’s research.17 The sample size was low, and biases may have been introduced with both eyes enrolled. The sample size of Whang et al.’s25 study (69 eyes of 69 patients) was also smaller than in our research. Moreover, the research was a retrospective study and enrolled eyes of patients undergoing cataract surgery. The importance of this study was to compare anterior corneal curvature measurements obtained by different devices and to evaluate the agreement between devices after refractive surgery based on the evaluation of corneal power for IOL calculation after refractive surgery. It is well accepted that misevaluation of corneal power is one of the principal sources of error in IOLpower miscalculation after corneal laser refractive surgery.1–15 The precision of anterior corneal curvature measurement is paramount in the estimation of actual corneal power after refractive surgery in the nonhistorical methods. As there are various instruments based on different technologies for measuring the anterior corneal curvature, it is not clear whether these devices are interchangeable or whether interdevice differences and agreement are similar to preoperative parameters. The present study showed that interdevice differences and agreement in anterior corneal measurement were influenced by refractive surgery. Reduced interdevice agreement was observed after surgery, indicating that corneal measurement precision might decrease because of alteration of the corneal curvature and asphericity induced by refractive surgery. This might be one source of error in IOL power calculations after refractive surgery. Pentacam and IOLMaster showed greater agreement both before and after refractive surgery. The 95% LOA value was 0.79 D before surgery and was 1.14 D after surgery. Although relatively higher agreement between Pentacam and IOLMaster Eye & Contact Lens Volume 0, Number 0, Month 2017
Eye & Contact Lens Volume 0, Number 0, Month 2017
Postrefractive Corneal Curvature Measurement
FIG. 1. Bland–Altman analysis evaluating agreement of measured K values between different devices before and corneal refractive surgery. (A) Agreement between Pentacam and Orbscan; (B) Agreement between Pentacam and IOLMaster; (C) Agreement between Orbscan and IOLMaster. Dots: preoperative interdevice agreement; triangles: Postperative interdevice agreement. The value between mean values shows the change of preoperative and postoperative interdevice difference. The difference between preoperative and postoperative LOAs shows the reduced interdevice LOA induced by refractive surgery.
Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists.5
Eye & Contact Lens Volume 0, Number 0, Month 2017
H. Jin et al. TABLE 4.
First Author Hashemi and Mehravaran17
Tajbakhsh et al.18 Crawford et al.19
No. of Eyes (No. Patients)
Summary of Researches Comparing K Readings Measured by Different Technologies
Instruments
46 (23)
Orbscan vs Pentacam (axial power, 3 mm zone) Orbscan vs Pentacam, (tangential power, 3 mm zone) Orbscan vs Pentacam (after refractive surgery, axial power, 3 mm zone) Orbscan vs Pentacam (after refractive surgery, tangential power, 3 mm zone) 115 (115) Orbscan vs Pentacam
K Value Measured by Device 2 (D) Mean6SD (D)
Difference Between Technologies (D)
P
Range (D)
Size (D)
Correlation (r)
44.5261.67
43.8461.60
0.68
,0.01
20.08 to 1.45
1.53
0.972
44.4361.73
43.7761.68
0.66
,0.01
20.93 to 2.24
3.17
0.888
40.6961.86
40.0561.78
0.63
,0.01
20.23 to 1.5
1.73
0.972
41.3461.69
40.8061.84
0.55
0.02
21.64 to 2.73
4.37
0.804
43.5261.48
43.7961.50
20.37
,0.001
1.046
NA
44.361.9
44.561.8
20.260.5
0.078
20.790 to 0.256 21 to 0.7
1.7
NA
43.361.8
43.461.6
20.160.5
No statistical significance 0.118
21 to 0.9
1.9
NA
20.15 to 0.92
1.07
0.931
NA
NA
NA
0.946
0.180
NA
NA
NA
0.102
NA
NA
NA
0.49 ,0.001
21.02 to 1.13 20.59 to 1.18
2.15 1.77
0.94 20.6870
0.000 ,0.001
NA 20.45 to 0.86
NA 1.31
NA 0.963
Symes and Ursell.20
63 (49)
Orbscan vs Pentacam (Kvalue of steep meridian) Orbscan vs Pentacam (Kvalue of flat meridian) Pentacam vs IOLMaster
Symes et al.21
29 (29)
Pentacam vs IOLMaster
43.5361.736
Saad et al.22
50 (50)
IOLMaster vs Pentacam 3 mm ring (simulated K) IOLMaster vs Pentacam 3 mm zone Pentacam vs IOLMaster IOLMaster vs Pentacam
43.6861.28
43.9361.415 20.11 calculated by author 43.4761.564 0.06 calculated by author 43.7661.31 20.0860.27
43.6861.28
43.7761.33
20.0860.35
43.3161.62 43.9761.44
43.2661.59 43.6761.49
44.6761.53 43.2561.20
44.0361.41 43.0461.25
0.0560.55 0.30 calculated by author 0.6360.78 0.2060.34
43.2561.20
43.2161.21
0.0460.20
0.06
20.36 to 0.43
0.79
0.986
43.0461.25
43.2161.21
20.1760.29
,0.001
20.74 to 0.40
1.14
0.973
38.6961.99
38.7561.91
20.0560.38
0.136
20.79 to 0.68
1.47
0.981
38.6961.99
38.5662.03
0.1360.29
,0.001
20.44 to 0.70
1.14
0.989
38.7561.91
38.5662.03
0.1960.34
,0.001
20.48 to 0.86
1.34
0.987
Lee et al.23 Savini et al.24 Whang et al.25 The present study
30 (30)
95% LOA
K Value Measured by Device 1 Mean6SD (D)
49 (41) 41 (41)
69 (69) Orbscan vs IOLMaster 109 (109) Pentacam vs Orbscan (before refractive surgery) Pentacam vs IOLMaster (before refractive surgery) Orbscan vs IOLMaster (before refractive surgery) Pentacam vs Orbscan (after refractive surgery) Pentacam vs IOLMaster (after refractive surgery) Orbscan vs IOLMaster (after refractive surgery)
43.8261.439
D, diopters; SD, standard deviation.
might indicate the superiority of the two devices in the measurement of anterior corneal curvature after refractive surgery, the change in interdevice difference and reduced agreement indicated that the three devices were not interchangeable. Before surgery, Pentacam and IOLMaster showed similar results, and Orbscan provided flatter K readings than Pentacam and IOLMaster. After refractive surgery, the difference between Pentacam and Orbscan was not statistically significant, and IOLMaster provided slightly steeper K readings than Pentacam and Orbscan. To further investigate the reliability of different devices in corneal curvature measurement after surgery, comparisons between measured and calculated K values after refractive surgery were performed for each device. No difference was found between the measured and calculated K value for Pentacam, whereas the differences between the measured and calculated K values were significantly different 6
for IOLMaster and Orbscan. The differences in K value measurements obtained by different devices may originate from the different measuring principles of the three devices. Several studies have compared K readings measured by different technologies after corneal refractive surgery.26,27 Ventura et al.26 compared K readings measured by a novel point-source color LED-based topographer, a Placido-disk topographer, and a combined Placido-based and dual Scheimpflug device in normal eyes and in postrefractive eyes. No interdevice difference was found between the devices. In another study performed by Lee et al.,27 dual rotating Scheimpflug–Placido, swept-source optical coherence tomography, and Placido-scanning-slit systems were used to compare corneal parameters in normal eyes and in postrefractive cases. Interdevice differences were revealed in both groups. In contrast to our study, the normal and postrefractive eyes were enrolled from Eye & Contact Lens Volume 0, Number 0, Month 2017
Eye & Contact Lens Volume 0, Number 0, Month 2017 two groups of populations in the two studies. Therefore, the influence of refractive surgery on interdevice agreement could not be evaluated. To the best of our knowledge, only one study has compared parameters measured by Pentacam and Orbscan before and after refractive surgery using the same group of patients.17 Changes in both anterior and posterior corneal powers were studied. The research revealed that Orbscan yielded steeper anterior corneal curvature measurements than Pentacam both before and after refractive surgery, which is not in agreement with our result. As has been discussed, the possible reasons for the disparity may be due to the study design. In that study, both eyes of 23 patients were enrolled. Moreover, although preoperative and postoperative interdevice agreements were analyzed respectively, a reduced agreement between them was not observed after surgery. The present research suggests for the first time that myopic laser corneal refractive surgery reduces interdevice agreement in the measurement of anterior corneal curvature. In conclusion, to the best of our knowledge, we believe that this study is the first to observe reduced interdevice agreement after corneal refractive surgery, indicating that devices for anterior corneal curvature measurement are not interchangeable. The reduced agreement might be one source of error in IOL power calculation after corneal refractive surgery. Further studies with more corneal curvature measurement devices and comparisons of their roles in IOL power calculation accuracy after refractive surgery are necessary in our future studies. REFERENCES 1. Holladay JT. Cataract surgery in patients with previous keratorefractive surgery (RK, PRK, and LASIK). Ophthalmic Pract 1997;15: 238–244. 2. Hoffer KJ. Intraocular lens power calculation after previous laser refractive surgery. J Cataract Refract Surg 2009;35:759–765. 3. Seitz B, Langenbucher A. Intraocular lens power calculation in eyes after corneal refractive surgery. J Refract Surg 2000;16:349–361. 4. Langenbucher A, Torres F, Behrens A, et al. Consideration of the posterior corneal curvature for assessment of corneal power after myopic LASIK. Acta Ophthalmol Scand 2004;82(3 Pt 1):264–269. 5. Savini G, Barboni P, Profazio V, et al. Corneal power measurements with the Pentacam Scheimpflug camera after myopic excimer laser surgery. J Cataract Refract Surg 2008;34:809–813. 6. Jin H, Rabsilber T, Ehmer A, et al. Comparison of ray-tracing method and thin-lens formula in intraocular lens power calculations. J Cataract Refract Surg 2009;35:650–662. 7. Canovas C, van der Mooren M, Rosén R, et al. Effect of the equivalent refractive index on intraocular lens power prediction with ray tracing after myopic laser in situ keratomileusis. J Cataract Refract Sur 2015;41: 1030–1037. 8. Savini G, Bedei A, Barboni P, et al. Intraocular lens power calculation by ray-tracing after myopic excimer laser surgery. Am J Ophthalmol 2014;157: 150–153. 9. Shammas HJ, Shammas MC, Garabet A, et al. Correcting the corneal power measurements for intraocular lens power calculations after myopic laser in situ keratomileusis. Am J Ophthalmol 2003;136:426–432. 10. Shammas HJ, Shammas MC. No-history method of intraocular lens power calculation for cataract surgery after myopic laser in situ keratomileusis. J Cataract Refract Surg 2007;33:31–36. 11. Wang L, Booth MA, Koch DD. Comparison of intraocular lens power calculation methods in eyes that have undergone LASIK. Ophthalmology 2004;111:1825–1831. 12. Wang L, Hill WE, Koch DD. Evaluation of intraocular lens power prediction methods using the American society of cataract and refractive surgeons post-keratorefractive intraocular lens power calculator. J Cataract Refract Surg 2010;36:1466–1473.
Postrefractive Corneal Curvature Measurement 13. Jin H, Holzer MP, Rabsilber T, et al. Intraocular lens power calculation after laser refractive surgery: corrective algorithm for corneal power estimation. J Cataract Refract Surg 2010;36:87–96. 14. Jin H, Auffarth GU, Guo H, et al. Corneal power estimation for intraocular lens power calculation after corneal laser refractive surgery in Chinese eyes. J Cataract Refract Surg 2012;38:1749–1757. 15. Haigis W. Intraocular lens calculation after refractive surgery for myopia: Haigis-L formula. J Cataract Refract Surg 2008;34:1658–1663. 16. Gatinel D, Hoang-Xuan T, Azar DT. Determination of corneal asphericity after myopia surgery with the excimer laser: a mathematical model. Invest Ophthalmol Vis Sci 2001;42:1736–1742. 17. Hashemi H, Mehravaran S. Corneal changes after laser refractive surgery for myopia: comparison of Orbscan II and Pentacam findings. J Cataract Refract Surg 2007;33:841–847. 18. Tajbakhsh Z, Salouti R, Nowroozzadeh MH, et al. Comparison of keratometry measurements using the Pentacam HR, the Orbscan IIz, and the TMS-4 topographer. Ophthalmic Physiol Opt 2012;32:539–546. 19. Crawford AZ, Patel DV, McGhee CN. Comparison and repeatability of keratometric and corneal power measurements obtained by Orbscan II, Pentacam, and Galilei corneal tomography systems. Am J Ophthalmol 2013;156:53–60. 20. Symes RJ, Ursell PG. Automated keratometry in routine cataract surgery: comparison of Scheimpflug and conventional values. J Cataract Refract Surg 2011;37:295–301. 21. Symes RJ, Say MJ, Ursell PG. Scheimpflug keratometry versus conventional automated keratometry in routine cataract surgery. J Cataract Refract Surg 2010;36:1107–1114. 22. Saad E, Shammas MC, Shammas HJ. Scheimpflug corneal power measurements for intraocular lens power calculation in cataract surgery. Am J Ophthalmol 2013;156:460–467. 23. Lee BW, Galor A, Feuer WJ, et al. Agreement between Pentacam and IOL master in patients undergoing toric IOL implantation. J Refract Surg 2013; 29:114–120. 24. Savini G, Barboni P, Carbonelli M, et al. Accuracy of Scheimpflug corneal power measurements for intraocular lens power calculation. Cataract Refract Surg 2009;35:1193–1197. 25. Whang WJ, Byun YS, Joo CK. Comparison of refractive outcomes using five devices for the assessment of preoperative corneal power. Clin Exp Ophthalmol 2012;40:425–432. 26. Ventura BV, Wang L, Ali SF, et al. Comparison of corneal power, astigmatism, and wavefront aberration measurements obtained by a point-source color light-emitting diode-based topographer, a Placido-disk topographer, and a combined Placido and dual Scheimpflug device. J Cataract Refract Surg 2015;41:1658–1671. 27. Lee YW, Choi CY, Yoon GY. Comparison of dual rotating ScheimpflugPlacido, swept-source optical coherence tomography, and Placido-scanningslit systems. J Cataract Refract Surg 2015;41:1018–1029. 28. Karimian F, Feizi S, Doozandeh A, et al. Comparison of corneal tomography measurements using Galilei, Orbscan II, and Placido disk-based topographer systems. J Refract Surg 2011;27:502–508. 29. Wang Q, Savini G, Hoffer KJ, et al. A comprehensive assessment of the precision and agreement of anterior corneal power measurements obtained using 8 different devices. PLoS One 2012;7:e45607. 30. Hugger P, Kohnen T, La Rosa FA, et al. Comparison of changes in manifest refraction and corneal power after photorefractive keratectomy. Am J Ophthalmol 2000;129:68–75. 31. Moshirfar M, Christiansen SM, Kim G. Comparison of the ratio of keratometric change to refractive change induced by myopic ablation. J Refract Surg 2012;28:675–682. 32. Visser N, Berendschot TT, Verbakel F, et al. Comparability and repeatability of corneal astigmatism measurements using different measurement technologies. J Cataract Refract Surg 2012;38:1764–1770. 33. Bland J, Altman D. Measuring agreement in method comparison studies. Stat Methods Med Res 1999;8:135–160. 34. Hashemi H, Asgari S, Emamian MH, et al. Age-Related changes in corneal curvature and shape: the shahroud eye cohort study. Cornea 2015;34:1456–1458. 35. Gipson IK. Age-related changes and diseases of the ocular surface and cornea. Invest Ophthalmol Vis Sci 2013;54:ORSF48–ORSF53. 36. Viso E, Rodriguez-Ares MT, Gude F. Prevalence of and associated factors for dry eye in a Spanish adult population (the Salnes Eye Study). Ophthalmic Epidemiol 2009;16:15–21.
Copyright 2017 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Contact Lens Association of Opthalmologists.7
