- Email: [email protected]
Repeatability and reproducibility of ocular biometry using a new noncontact optical low-coherence interferometer
ARTICLE
Repeatability and reproducibility of ocular biometry using a new noncontact optical low-coherence interferometer Jinhai Huang, MD, Giacomo Savini, MD, Fan Wu, MD, Xinxin Yu, MD, Jing Yang, MD, Ayong Yu, MD, PhD, Ye Yu, MD, Qinmei Wang, MD
PURPOSE: To assess the precision of ocular biometry with a new noncontact optical low-coherence interferometer (Aladdin) in healthy subjects and patients with cataracts. SETTING: Eye Hospital of Wenzhou Medical University, Wenzhou, China. DESIGN: Observational cross-sectional study. METHODS: Eyes from healthy subjects and eyes from patients with cataracts were examined with the new interferometer. Axial length (AL), anterior chamber depth (ACD), keratometry (K), and white-towhite (WTW) values were measured by 2 operators. The test–retest repeatability, coefficient of variation (CoV), and intraclass correlation coefficient (ICC) were calculated to evaluate intraoperator repeatability. Different-operator comparison was analyzed with paired t tests and Bland-Altman plots to assess interoperator reproducibility. RESULTS: Ninety-eight people were enrolled for this study, of which 52 eyes were from healthy subjects and 46 eyes were from patients with cataracts. The AL was the most repeatable and reproducible parameter. The ACD, K values, and WTW measurements were highly repeatable in healthy subjects with CoV less than 0.89% and ICC more than 0.94. However, in patients with cataracts, high repeatability could only be observed for AL, ACD, and K values, whereas WTW measurement had test–retest repeatability of 0.80 mm and an ICC of 0.795. Bland-Altman analysis also showed good agreement between the 2 operators for ocular component measurements, except that the WTW in patients with cataracts had wider 95% limits of agreement (range 0.88 to 0.95 mm). CONCLUSION: The new biometer showed excellent intraoperator repeatability and interoperator reproducibility for AL, ACD, and K values measurements in both groups. The precision of WTW measurements was lower in patients with cataracts. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2015; 41:2233–2241 Q 2015 ASCRS and ESCRS
Achieving high accuracy and precision with ocular biometry has several practical and theoretical applications in ophthalmic practice. Axial length (AL), keratometry (K), anterior chamber depth (ACD), and white-to-white (WTW) measurements are used for intraocular lens (IOL) power calculation and phakic IOL implantation.1–4 The ACD measurement can also be used in studies to screen glaucoma risk factors.5 Since 1999, with the introduction of the IOLMaster (Carl Zeiss Meditec AG),6 optical biometry has become the gold standard to achieve these measurements. Recently, other instruments have been introduced, Q 2015 ASCRS and ESCRS Published by Elsevier Inc.
namely, the Lenstar (Haag-Streit Diagnostics),7 ALScan (Nidek Co., Ltd.),8,9 Galilei G6 (Ziemer Ophthalmic Systems AG), OA-2000/OA-1000 (Tomey Corp.),10,11 and Aladdin (Topcon Corp.). The new noncontact device Aladdin, based on optical low-coherence interferometry (OLCI) and with Placido-disk corneal topography, is simultaneously able to measure AL, ACD, K values, corneal astigmatism, higher-order aberrations, and horizontal WTW. The Aladdin has been shown to provide biometric values that are close to the IOLMaster.12 Before a new instrument can be widely accepted for use in ophthalmic examination, http://dx.doi.org/10.1016/j.jcrs.2015.10.062 0886-3350
2233
2234
RELIABILITY OF BIOMETRY USING OLCI
it is critical to evaluate its precision. It is not possible to achieve reliable accuracy or agreement in individual measurements without good repeatability.13 Assessing its repeatability and reproducibility are fundamental to determine the actual change in longitudinal studies due to the passage of time or an intervention. However, to our knowledge, there are no studies assessing intraoperator repeatability. Hence, the present study aimed to evaluate both the intraoperator repeatability and interoperator reproducibility of the Aladdin measurements in healthy subjects and patients with cataracts. SUBJECTS AND METHODS This prospective study comprised eyes of healthy subjects and eyes of patients with cataracts. Patients were recruited at the Eye Hospital of Wenzhou Medical University, Wenzhou, China. The healthy subjects had no ocular abnormalities such as keratoconus, cataract, previous ocular surgery, or history of wearing contact lens other than for refractive error. The exclusion criteria for the cataract group included a history of ocular surgery, keratoconus, fundus disease, contact lens usage other than for refractive error, and any other ocular disease except for cataracts. All measurements were performed in the right eye of each subject to avoid methodology bias from using both eyes. The research protocol was reviewed and approved by the Ethics Committee of the Eye Hospital of Wenzhou Medical University. An informed consent document was obtained from each subject after he or she received thorough explanation about the study, per Declaration of Helsinki principles.
Instruments The Aladdin combines an optical biometer and a Placidoring topographer. Optical biometry relies on OLCI, based on an 830 nm super-luminescent diode that is used to Submitted: January 6, 2015. Final revision submitted: March 6, 2015. Accepted: March 9, 2015. From the School of Ophthalmology and Optometry and Eye Hospital (Huang, Wu, X. Yu, Yang, A. Yu, Y. Yu, Wang), Wenzhou Medical University, Wenzhou, and the Key Laboratory of Vision Science (Huang, Yang, A. Yu, Y. Yu, Wang), Ministry of Health, P.R. China, Wenzhou, Zhejiang, China; the G.B. Bietti Foundation IRCCS (Savini), Rome, Italy. Supported in part by the Natural Science Foundation of China (81300807), the Foundation of Wenzhou City Science & Technology Bureau (J20140014), the Health Bureau of Zhejiang Province (2012KYB135), and the Zhejiang Provincial and Ministry of Health Research Fund for Medical Sciences (WKJ-ZJ-1530). The contribution of the GB Bietti Foundation IRCCS was supported by Fondazione Roma and the Italian Ministry of Health. Corresponding author: Qinmei Wang, MD, Eye Hospital of Wenzhou Medical University, 270 West Xueyuan Road, Wenzhou, Zhejiang 325027, China. E-mail: [email protected].
measure the AL of the eye (range 15 to 38 mm). This ACD measurement is performed using the reflection principle of a 473 nm blue light–emitting diode horizontal slit light projected onto the anterior chamber. The WTW measurements obtained are similar to the AL-Scan by fitting the best circle with the lowest error square to the detected edge.14 Corneal topography is based on the reflection of 24 Placido disk rings with a diameter of 8.0 mm. However, K values are not derived from simulated K but from automated K. Keratometry values are generated from the reflection of 4 dedicated Placido rings (for a total of 1024 points), whose diameter ranges between 2.4 mm and 3.4 mm. Corneal curvature data are converted to corneal power by means of the standard keratometric index (1.3375). In this study, the flattest K, steepest K, and mean K (average of flat K and steep K) were recorded and used to assess the corneal power measurement of the new biometer.
Measurements The precision of the new biometer measurements was assessed on the basis of the definitions adopted by the British Standards Institution as recommended by Bland and Altman.13 The subject was seated with his or her chin on a chinrest and forehead against the forehead strap and asked to fixate straight ahead on a fixation target. Each subject was consecutively measured by 2 experienced operators in random order; the entire process took no more than 10 minutes. The machine provided 6 AL measurements with each click, which were averaged to a single value; 1 K measurement for the flat and steep meridians; and ACD and WTW measurements. After each measurement, the device was moved backward and realigned for the next scan to eliminate interdependence of the successive acquisition, and 3 valid scans were performed by both operators. The scans could be influenced by several factors such as eye movements and eyelid blinking. According to the manufacturer, the yellow mark in the acquisition window indicated correctly acquired measurements, whereas the red mark indicated wrong measurements, which were discarded and repeated again. If the obtained scans did not meet the required specifications, they were considered as subqualified scans and repeated. The AL, ACD, K, and WTW values for every qualified measurement were recorded to estimate the repeatability and reproducibility of the new biometer.
Statistical Analysis All data were analyzed using SPSS for Windows software (version 21, SPSS, Inc.). The results are presented as the mean G standard deviations (SD). The distributions of the datasets were checked for normality using the Kolmogorov-Smirnov test, and the results indicated that the data were normally distributed (P O .05). To assess intraoperator repeatability, the similar degree between the outcomes of multiple measurements was performed by a single operator. The within-subject SD, coefficient of variation (CoV), test–retest repeatability, and intraclass correlation coefficient (ICC) were measured. The within-subject SD was the square root of the residual mean square in the 1-way analysis of variance. The test–retest was calculated by multiplying the within-subject value by 2.77, which represented the interval within which 95% of the differences in the measurements were expected to lie.15 The percentage of CoV was calculated as the ratio of the within-subject to the overall mean (the lower the CoV, the
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2235
RELIABILITY OF BIOMETRY USING OLCI
Table 1. Intraoperator repeatability outcomes for biometric measurements obtained in the health group. Parameter
Operator
Mean G SD
Sw
TRT
CoV (%)
ICC (95% CI)
AL (mm)
1 2 1 2 1 2 1 2 1 2 1 2
25.14 G 0.99 25.14 G 0.99 3.73 G 0.24 3.73 G 0.24 43.22 G 1.33 43.23 G 1.33 44.21 G 1.55 44.21 G 1.56 43.72 G 1.42 43.72 G 1.42 11.61 G 0.42 11.63 G 0.42
0.02 0.02 0.03 0.03 0.10 0.08 0.14 0.10 0.10 0.08 0.09 0.10
0.06 0.05 0.09 0.09 0.27 0.22 0.38 0.28 0.28 0.22 0.24 0.28
0.09 0.08 0.82 0.88 0.23 0.19 0.31 0.23 0.23 0.18 0.74 0.88
0.999 (0.999-1.000) 1.000 (0.999-1.000) 0.984 (0.975-0.990) 0.982 (0.971-0.989) 0.995 (0.992-0.997) 0.996 (0.994-0.998) 0.992 (0.988-0.995) 0.996 (0.993-0.997) 0.995 (0.992-0.997) 0.997 (0.995-0.998) 0.959 (0.936-0.975) 0.942 (0.911-0.964)
ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; CoV Z within-subject coefficient of variation; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; Sw Z within-subject standard deviation; TRT Z test–retest repeatability (2.77 Sw); WTW Z white to white
higher the repeatability). The ICC represented the consistency in data measurement, and its value ranged from 0 to 1; a value higher than 0.90 indicated high agreement.16 To assess interoperator reproducibility, the similar degree between the outcomes of measurements was performed by 2 different operators, and the mean of the 3 readings from each session was calculated. The paired t test was used to compare the mean ocular components obtained with the new biometer by 2 operators. Furthermore, agreement between these 2 operators' measurements was assessed per the method described by the BlandAltman plot, and the 95% limits of agreement (LoA) were calculated by the mean difference G1.96 SD. The ICC was also used to evaluate the consistency of measurement by different operators. A P value less than 0.05 was considered statistically significant. The sample size was calculated per Bland and Altman, who recommended at least 24 subjects when 3 repeated measurements are taken and a confidence interval with 20% in the estimate is considered.A
RESULTS The healthy group comprised 52 eyes of 52 healthy subjects (18 men and 34 women) with a mean age of 25.33 years G 1.99 (SD) (range 22 to 33 years). The cataract group comprised 46 eyes of 46 patients with cataract (14 men and 32 women) with a mean age of 65.07 G 8.01 years (range 50 to 82 years).
Intraoperator Repeatability Measurements of the AL, ACD, K values, and WTW with the new biometer showed high intraoperator repeatability with both operators (Table 1). The CoV of all ocular components was lower than 0.89%, and the ICC was higher than 0.94. Measurement of the AL provided the highest repeatability
Healthy Group
Table 2. Intraoperator repeatability outcomes for biometric measurements obtained using Aladdin optical low-coherence interferometer in the cataract group. Parameter
Operator
Mean G SD
Sw
TRT
CoV (%)
ICC (95% CI)
AL (mm)
1 2 1 2 1 2 1 2 1 2 1 2
23.52 G 1.38 23.52 G 1.37 3.20 G 0.40 3.19 G 0.40 43.90 G 1.55 43.93 G 1.55 44.81 G 1.47 44.78 G 1.46 44.35 G 1.47 44.36 G 1.47 11.28 G 0.52 11.24 G 0.60
0.02 0.02 0.05 0.04 0.12 0.18 0.13 0.17 0.09 0.13 0.24 0.29
0.06 0.07 0.15 0.12 0.33 0.51 0.36 0.46 0.25 0.37 0.66 0.80
0.09 0.11 1.70 1.38 0.27 0.42 0.29 0.37 0.21 0.30 2.11 2.58
1.000 (1.000-1.000) 1.000 (0.999-1.000) 0.982 (0.971-0.989) 0.988 (0.980-0.993) 0.994 (0.990-0.997) 0.986 (0.977-0.992) 0.992 (0.987-0.995) 0.987 (0.979-0.992) 0.996 (0.994-0.998) 0.992 (0.987-0.995) 0.814 (0.719-0.885) 0.795 (0.693-0.872)
ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; CoV Z within-subject coefficient of variation; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; Sw Z within-subject standard deviation; TRT Z test-retest repeatability (2.77 Sw); WTW Z white to white
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2236
RELIABILITY OF BIOMETRY USING OLCI
Table 3. Mean difference, paired t test, 95% LOA, and ICC for differences between the 2 operators in the healthy group. Parameter
Mean Difference G SD
AL (mm) ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
0.00 G 0.02 0.00 G 0.03 0.00 G 0.08 0.00 G 0.12 0.00 G 0.08 0.02 G 0.09
P Value
95% LoA
.148 .684 .829 .831 .987 .080
0.03, 0.04 0.06, 0.07 0.17, 0.16 0.22, 0.23 0.17, 0.17 0.19, 0.15
ICC (95% CI) 1.000 (1.000-1.000) 0.990 (0.983-0.994) 0.998 (0.997-0.999) 0.997 (0.995-0.998) 0.998 (0.997-0.999) 0.978 (0.962-0.988)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; LoA Z limits of agreement; WTW Z white to white.
as the test–retest repeatability was lower than 0.06 mm and the ICC between 0.999 and 1.0. Cataract Group Measurements of the AL, ACD, and K
values with the new biometer showed high intraoperator repeatability with both operators. The ICC was higher than 0.98 (Table 2). In contrast to the healthy group, a lower repeatability was detected for WTW measurements; the ICC was about 0.8 and the test–retest repeatability values were 0.66 and 0.8 mm for the 2 operators. Interoperator Reproducibility Healthy Group There was no statistically significant
difference between the 2 operators' measurements (P O .05), and the interoperator reproducibility was excellent for all parameters (Table 3). Bland-Altman plots showed narrow 95% LoA for the AL, ACD, K values, and WTW measurements, and a fixed bias was not detected between the 2 operators (Figures 1, A to 6, A).
Cataract Group There was no statistically significant difference between the 2 operators' measurements. The interoperator reproducibility was high with both operators for all measurements except for WTW (Table 4). Bland-Altman plots showed narrow 95% LoA for the AL, ACD, and K values and wider 95% LoA for WTW (Figures 1, B to 6, B). However, a fixed bias was not detected in WTW measurements between the 2 operators.
Difference Between Healthy and Cataract Groups The 95% LoA of AL, ACD, flat K, steep K, and mean K measurements were slightly narrower in the healthy group than in the cataract group. However, eyes with cataract produced wider 95% LoA for WTW measurements than healthy eyes (Tables 3 and 4). DISCUSSION Refractive cataract surgery requires the most accurate and precise biometric measurements possible. Several
Figure 1. Bland-Altman plots showing agreement in AL measurements between the 2 operators in healthy group (A) and patients in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA. J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
RELIABILITY OF BIOMETRY USING OLCI
2237
Figure 2. Bland-Altman plots showing agreement in ACD measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA.
devices are currently available to fulfill this goal. The Aladdin is a new optical biometer based on OLCI. This is the first study to investigate the intraoperator repeatability and interoperator reproducibility of AL, K values, ACD, and WTW measurements using the new biometer in healthy subjects and patients with cataracts. Excellent repeatability and reproducibility were found in all measured parameters except for WTW in patients with cataracts. In AL and ACD measurements using the new biometer, the study found that test–retest values were slightly lower in the healthy and cataract groups; at the same time, the ICC of repeatability was up to 1 for AL and to 0.98 for ACD measurements in both groups. Another optical device, the IOLMaster, which
measures AL, K values, and ACD, has been compared with the new biometer.12 The IOLMaster is accepted as a gold standard for ocular biometry, and its repeatability and reproducibility have been assessed in several studies.6,17–20 According to Mandal et al.,12 the Aladdin provided repeatable measurements similar to those of the IOLMaster. The interoperator reproducibility of AL and ACD measurements was also found to be excellent. Compared with patients with cataracts, healthy subjects showed slightly narrow 95% LoA. Several authors had already evaluated the reproducibility by other optical biometers. Huang et al.14 assessed the interoperator reproducibility of the AL-Scan and reported that highly reproducible AL and ACD measurements with the ICC
Figure 3. Bland-Altman plots showing agreement in flat K measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA. J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2238
RELIABILITY OF BIOMETRY USING OLCI
Figure 4. Bland-Altman plots showing agreement in steep K measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA.
were higher than 0.99 and the CoV was lower than 0.43%. A more recent study21 showed that the AL-Scan achieved excellent reliability for AL measurements in keratoconic eyes (test–retest values ranged from 0.06 to 0.09 mm for both operators). Bjelos Roncevic et al.22 did similar research of the reproducibility with the Lenstar. It provided reproducible measurements of AL and ACD in patients with cataracts with 95% LoA within a range from 0.02 to 0.02 mm and 0.11 to 0.14 mm, respectively. Buckhurst et al.,7 Srivannaboon et al.8 Goebels et al.,10,11 and S¸ahin et al.23 also reported good reproducibility with the Lenstar, AL-Scan, OA-1000, and IOLMaster. All of these results were similar to the present study results.
Unlike most ocular biometers, the Aladdin can also provide Placido-disk corneal topography. The K values are obtained from 1024 points reflected from rings with an average diameter of 3.0 mm. Previous studies evaluated the repeatability and reproducibility of Placido systems. Tang et al.24 compared 3 Placidodisk videokeratoscopes including the Keratron (Alliance Medical Marketing), Medmont (Medmont Pty Ltd.), and Topographic Modeling System (TMS, Computed Anatomy Inc.) to assess the precision in measuring test surfaces and determined that the Keratron and Medmont had good precision, whereas the TMS showed worse results. In this study, the test– retest values were lower than 0.51 diopter (D) in healthy subjects and patients with cataracts and the
Figure 5. Bland-Altman plots showing agreement in mean K measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA. J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2239
RELIABILITY OF BIOMETRY USING OLCI
Figure 6. Bland-Altman plots showing agreement in WTW measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA.
95% reproducibility LoA ranged from 0.28 to 0.33 D. A previous study reported that a 1.0 D K value measurement error would contribute to 0.8 to 1.3 D IOL power calculation changes.25 The repeatability of these values measured by the Aladdin device was excellent and similar to that in previous studies of other devices. Therefore, the Aladdin demonstrated clinically acceptable repeatability and reproducibility in K value measurement. The repeatability and reproducibility of WTW in the healthy group were good and similar to the other measured parameters. On the contrary, the values were lower in patients with cataracts as the 95% the LoA for interoperator reproducibility was as wide as 0.95 mm. Several studies have assessed the precision of WTW measurement with other optical biometry devices and obtained outcomes similar to those in the present study. Huang et al.14 found that the test–retest repeatability with the AL-scan in patients with cataract was 0.53 mm, and the ICC of reproducibility was 0.897. However, Bjelos Roncevic et al.22 showed that the Lenstar had a lower reproducibility of WTW
measurements (95% LoA, 1.13 to 1.05 mm) in patients with cataracts. Srivannaboon et al.8 also reported the low repeatability and reproducibility of WTW with the AL-Scan and IOLMaster in patients with cataracts. The results showed that the 95% LoA of repeatability and reproducibility were 0.13 to 1.21 mm and 0.94 to 1.06 mm for the AL-Scan, respectively, and 0.27 to 0.45 mm and 0.44 to 0.37 mm for the IOLMaster, respectively. Exact measurement of WTW is important to estimate the horizontal diameter of the anterior chamber for phakic IOL implantation (eg, angle- and sulcussupported phakic IOLs). In these circumstances, selecting accurate IOL size based on WTW value is critical to reduce postoperative complications such as pupil ovalization, endothelial damage, iris retraction, chronic inflammation, and cataract formation.21–23 The Aladdin biometer calculates the WTW by distinguishing the light and shade interface between cornea and sclera; the high prevalence of arcus senilis in elderly patients with cataracts may influence the identification of this edge. Another
Table 4. Mean difference, paired t test, 95% LOA, and ICC for differences between the 2 operators in the cataract group. Parameter AL (mm) ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
Mean Difference G SD 0.01 G 0.03 0.01 G 0.04 0.03 G 0.13 0.02 G 0.16 0.00 G 0.11 0.04 G 0.47
P Value
95% LoA
.091 .103 .131 .372 .857 .587
0.05, 0.07 0.06, 0.08 0.28, 0.22 0.28, 0.33 0.23, 0.22 0.88, 0.95
ICC (95% CI) 1.000 (1.000-1.000) 0.996 (0.993-0.998) 0.997 (0.994-0.998) 0.994 (0.990-0.997) 0.997 (0.995-0.998) 0.653 (0.449-0.792)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; LoA Z limits of agreement; WTW Z white to white
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2240
RELIABILITY OF BIOMETRY USING OLCI
reason is poor cooperation from older patients who possibly could not follow the operator's instructions. There were several limitations in the present study. First, the study investigated the reliability in healthy subjects and in patients with cataracts. However, the study findings cannot be applied to patients with other diseases such as keratoconus or contact lens usage or those who had corneal refractive surgery. Measurements in eyes with irregular or postsurgical corneas and ocular diseases could be the focus of future research. Moreover, the repeatability and reproducibility parameters derived from corneal topography such as asphericity and aberration have not been investigated. In conclusion, the new noncontact OLCI device Aladdin showed excellent intraoperator repeatability and interoperator reproducibility for all measurements except WTW in eyes with cataract. This suggests that the Aladdin can be used for routine clinical practice to acquire AL, ACD, and K values for IOL power calculation. WHAT WAS KNOWN Currently available optical biometers provide repeatable and reproducible measurements of ocular parameters required to calculate IOL power. WHAT THIS PAPER ADDS The new biometer provided highly repeatable and reproducible measurements of corneal power, ACD, and AL. The precision of WTW measurements was lower in eyes with cataract.
REFERENCES 1. Hoffer KJ. The Hoffer Q formula: a comparison of theoretic and regression formulas. J Cataract Refract Surg 1993; 19:700–712; errata, 1994; 20:677 2. Baumeister M, Terzi E, Ekici Y, Kohnen T. Comparison of manual and automated methods to determine horizontal corneal diameter. J Cataract Refract Surg 2004; 30:374–380 3. Hoffer KJ. Clinical results using the Holladay 2 intraocular lens power formula. J Cataract Refract Surg 2000; 26:1233–1237 4. Retzlaff JA, Sanders DR, Kraff MC. Development of the SRK/T intraocular lens implant power calculation formula. J Cataract Refract Surg 1990; 16:333–340; erratum, 528 5. Congdon NG, Youlin Q, Quigley H, Hung PT, Wang TH, Ho TC, Tielsch JM. Biometry and primary angle-closure glaucoma among Chinese, white, and black populations. Ophthalmology 1997; 104:1489–1495 6. Santodomingo-Rubido J, Mallen EA, Gilmartin B, Wolffsohn JS. A new non-contact optical device for ocular biometry. Br J Ophthalmol 2002; 86:458–462. Available at: http://www.ncbi.nlm. nih.gov/pmc/articles/PMC1771084/pdf/bjo08600458.pdf. Accessed August 12, 2015
7. Buckhurst PJ, Wolffsohn JS, Shah S, Naroo SA, Davies LN, Berrow EJ. A new optical low coherence reflectometry device for ocular biometry in cataract patients. Br J Ophthalmol 2009; 93:949–953 8. Srivannaboon S, Chirapapaisan C, Chonpimai P, Koodkaew S. Comparison of ocular biometry and intraocular lens power using a new biometer and a standard biometer. J Cataract Refract Surg 2014; 40:709–715 9. Kola M, Duran H, Turk A, Mollamehmetoglu S, Kalkisim A, Erdol H. Evaluation of the repeatability and the reproducibility of AL-Scan measurements obtained by residents. J Ophthalmol 2014 article ID:739652. Available at: http://downloads.hindawi. com/journals/joph/2014/739652.pdf. Accessed August 12, 2015 10. Goebels SC, Seitz B, Langenbucher A. Reproducibility of the optical biometer OA-1000 (Tomey). Biomed Res Int 2014 article ID:814761. Available at: http://downloads.hindawi.com/journals/ bmri/2014/814761.pdf. Accessed August 12, 2015 11. Goebels SC, Seitz B, Langenbucher A. Comparison of the new biometer OA-1000 with IOLMaster and Tomey AL-3000. Curr Eye Res 2013; 38:910–916 12. Mandal P, Berrow EJ, Naroo SA, Wolffsohn JS, Uthoff D, Holland D, Shah S. Validity and repeatability of the Aladdin ocular biometer. Br J Ophthalmol 2014; 98:256–258 13. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310. Available at: http://www-users.york.ac.uk/ wmb55/meas/ba.pdf. Accessed August 12, 2015 14. Huang J, Savini G, Li J, Lu W, Wu F, Wang J, Li Y, Feng Y, Wang Q. Evaluation of a new optical biometry device for measurements of ocular components and its comparison with IOLMaster. Br J Ophthalmol 2014; 98:1277–1281 15. Bland JM, Altman DG. Measurement error [Statistics notes]. BMJ 1996; 313:744. Available at: http://www.ncbi.nlm.nih.gov/ pmc/articles/PMC2352111/pdf/bmj00560-0056b.pdf. Accessed August 12, 2015 €ller R, Bu €ttner P. A critical discussion of intraclass correlation 16. Mu coefficients. Stat Med 1994; 13:2465–2476. Available at: http:// www.en.msc-epidemiologie.med.uni-muenchen.de/download/ winter_term_2011_2012/epiresearchdesigns_articles/mueller buttner02_12.pdf. Accessed August 12, 2015 17. Kielhorn I, Rajan MS, Tesha PM, Subryan VR, Bell JA. Clinical assessment of the Zeiss IOLMaster. J Cataract Refract Surg 2003; 29:518–522 18. Sheng H, Bottjer CA, Bullimore MA. Ocular component measurement using the Zeiss IOLMaster. Optom Vis Sci 2004; 81:27–34. Available at: http://journals.lww.com/optvissci/Fulltext/2004/ 01000/The_Contact_Lens_and_Myopia_Progression__CLAMP_ .7.aspx. Accessed August 12, 2015 19. Carkeet A, Saw S-M, Gazzard G, Tang W, Tan DT. Repeatability of IOLMaster biometry in children. Optom Vis Sci 2004; 81:829– 834. Available at: http://journals.lww.com/optvissci/Fulltext/2004/ 11000/Repeatability_of_IOLMaster_Biometry_in_Children.8.aspx. Accessed August 12, 2015 20. Lam AK, Chan R, Pang PC. The repeatability and accuracy of axial length and anterior chamber depth measurements from the IOLMasterä. Ophthalmic Physiol Opt 2001; 21:477–483 cı R, Gu an BD, Balcı M, Heps‚en _IF. €ler E, Kulak AE, Erdog 21. Yag Repeatability and reproducibility of a new optical biometer in normal and keratoconic eyes. J Cataract Refract Surg 2015; 41:171–177 22. Bjelos Roncevic M, Busic M, Cima I, Kuzmanovic Elabjer B, Bosnar D, Miletic D. Comparison of optical low-coherence reflectometry and applanation ultrasound biometry on intraocular lens power calculation. Graefes Arch Clin Exp Ophthalmol 2011; 249:69–75
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2241
RELIABILITY OF BIOMETRY USING OLCI
€ rsoy H, Bas‚mak H, Yildirim N, Usalp Z, C‚olak E. 23. S‚ahin A, Gu Reproducibility of ocular biometry with a new noncontact optical low-coherence reflectometer in children. Eur J Ophthalmol 2011; 21:194–198 24. Tang W, Collins MJ, Carney L, Davis B. The accuracy and precision performance of four videokeratoscopes in measuring test surfaces. Optom Vis Sci 2000; 77:483–491. Available at: http:// journals.lww.com/optvissci/Fulltext/2000/09000/The_Accuracy_ and_Precision_Performance_of_Four.9.aspx. Accessed August 12, 2015 25. Mehdizadeh M. Effect of axial length and keratometry measurement error on intraocular lens implant power prediction formulas in pediatric patients [letter]. J AAPOS 2008; 12:425; reply by M Eibschitz-Tsimhoni, O Tsimhoni, SM Archer, MA Del Monte, 425 426
OTHER CITED MATERIAL A. Bland JM. Frequently asked questions on the design and analysis of measurement studies. How can I decide the sample size for a repeatability study? Available at: https://www-users.york. ac.uk/wmb55/meas/sizerep.htm. Accessed August 12, 2015.
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
First author: Jinhai Huang, MD School of Ophthalmology and Optometry and Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, China
Repeatability and reproducibility of ocular biometry using a new noncontact optical low-coherence interferometer Jinhai Huang, MD, Giacomo Savini, MD, Fan Wu, MD, Xinxin Yu, MD, Jing Yang, MD, Ayong Yu, MD, PhD, Ye Yu, MD, Qinmei Wang, MD
PURPOSE: To assess the precision of ocular biometry with a new noncontact optical low-coherence interferometer (Aladdin) in healthy subjects and patients with cataracts. SETTING: Eye Hospital of Wenzhou Medical University, Wenzhou, China. DESIGN: Observational cross-sectional study. METHODS: Eyes from healthy subjects and eyes from patients with cataracts were examined with the new interferometer. Axial length (AL), anterior chamber depth (ACD), keratometry (K), and white-towhite (WTW) values were measured by 2 operators. The test–retest repeatability, coefficient of variation (CoV), and intraclass correlation coefficient (ICC) were calculated to evaluate intraoperator repeatability. Different-operator comparison was analyzed with paired t tests and Bland-Altman plots to assess interoperator reproducibility. RESULTS: Ninety-eight people were enrolled for this study, of which 52 eyes were from healthy subjects and 46 eyes were from patients with cataracts. The AL was the most repeatable and reproducible parameter. The ACD, K values, and WTW measurements were highly repeatable in healthy subjects with CoV less than 0.89% and ICC more than 0.94. However, in patients with cataracts, high repeatability could only be observed for AL, ACD, and K values, whereas WTW measurement had test–retest repeatability of 0.80 mm and an ICC of 0.795. Bland-Altman analysis also showed good agreement between the 2 operators for ocular component measurements, except that the WTW in patients with cataracts had wider 95% limits of agreement (range 0.88 to 0.95 mm). CONCLUSION: The new biometer showed excellent intraoperator repeatability and interoperator reproducibility for AL, ACD, and K values measurements in both groups. The precision of WTW measurements was lower in patients with cataracts. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2015; 41:2233–2241 Q 2015 ASCRS and ESCRS
Achieving high accuracy and precision with ocular biometry has several practical and theoretical applications in ophthalmic practice. Axial length (AL), keratometry (K), anterior chamber depth (ACD), and white-to-white (WTW) measurements are used for intraocular lens (IOL) power calculation and phakic IOL implantation.1–4 The ACD measurement can also be used in studies to screen glaucoma risk factors.5 Since 1999, with the introduction of the IOLMaster (Carl Zeiss Meditec AG),6 optical biometry has become the gold standard to achieve these measurements. Recently, other instruments have been introduced, Q 2015 ASCRS and ESCRS Published by Elsevier Inc.
namely, the Lenstar (Haag-Streit Diagnostics),7 ALScan (Nidek Co., Ltd.),8,9 Galilei G6 (Ziemer Ophthalmic Systems AG), OA-2000/OA-1000 (Tomey Corp.),10,11 and Aladdin (Topcon Corp.). The new noncontact device Aladdin, based on optical low-coherence interferometry (OLCI) and with Placido-disk corneal topography, is simultaneously able to measure AL, ACD, K values, corneal astigmatism, higher-order aberrations, and horizontal WTW. The Aladdin has been shown to provide biometric values that are close to the IOLMaster.12 Before a new instrument can be widely accepted for use in ophthalmic examination, http://dx.doi.org/10.1016/j.jcrs.2015.10.062 0886-3350
2233
2234
RELIABILITY OF BIOMETRY USING OLCI
it is critical to evaluate its precision. It is not possible to achieve reliable accuracy or agreement in individual measurements without good repeatability.13 Assessing its repeatability and reproducibility are fundamental to determine the actual change in longitudinal studies due to the passage of time or an intervention. However, to our knowledge, there are no studies assessing intraoperator repeatability. Hence, the present study aimed to evaluate both the intraoperator repeatability and interoperator reproducibility of the Aladdin measurements in healthy subjects and patients with cataracts. SUBJECTS AND METHODS This prospective study comprised eyes of healthy subjects and eyes of patients with cataracts. Patients were recruited at the Eye Hospital of Wenzhou Medical University, Wenzhou, China. The healthy subjects had no ocular abnormalities such as keratoconus, cataract, previous ocular surgery, or history of wearing contact lens other than for refractive error. The exclusion criteria for the cataract group included a history of ocular surgery, keratoconus, fundus disease, contact lens usage other than for refractive error, and any other ocular disease except for cataracts. All measurements were performed in the right eye of each subject to avoid methodology bias from using both eyes. The research protocol was reviewed and approved by the Ethics Committee of the Eye Hospital of Wenzhou Medical University. An informed consent document was obtained from each subject after he or she received thorough explanation about the study, per Declaration of Helsinki principles.
Instruments The Aladdin combines an optical biometer and a Placidoring topographer. Optical biometry relies on OLCI, based on an 830 nm super-luminescent diode that is used to Submitted: January 6, 2015. Final revision submitted: March 6, 2015. Accepted: March 9, 2015. From the School of Ophthalmology and Optometry and Eye Hospital (Huang, Wu, X. Yu, Yang, A. Yu, Y. Yu, Wang), Wenzhou Medical University, Wenzhou, and the Key Laboratory of Vision Science (Huang, Yang, A. Yu, Y. Yu, Wang), Ministry of Health, P.R. China, Wenzhou, Zhejiang, China; the G.B. Bietti Foundation IRCCS (Savini), Rome, Italy. Supported in part by the Natural Science Foundation of China (81300807), the Foundation of Wenzhou City Science & Technology Bureau (J20140014), the Health Bureau of Zhejiang Province (2012KYB135), and the Zhejiang Provincial and Ministry of Health Research Fund for Medical Sciences (WKJ-ZJ-1530). The contribution of the GB Bietti Foundation IRCCS was supported by Fondazione Roma and the Italian Ministry of Health. Corresponding author: Qinmei Wang, MD, Eye Hospital of Wenzhou Medical University, 270 West Xueyuan Road, Wenzhou, Zhejiang 325027, China. E-mail: [email protected].
measure the AL of the eye (range 15 to 38 mm). This ACD measurement is performed using the reflection principle of a 473 nm blue light–emitting diode horizontal slit light projected onto the anterior chamber. The WTW measurements obtained are similar to the AL-Scan by fitting the best circle with the lowest error square to the detected edge.14 Corneal topography is based on the reflection of 24 Placido disk rings with a diameter of 8.0 mm. However, K values are not derived from simulated K but from automated K. Keratometry values are generated from the reflection of 4 dedicated Placido rings (for a total of 1024 points), whose diameter ranges between 2.4 mm and 3.4 mm. Corneal curvature data are converted to corneal power by means of the standard keratometric index (1.3375). In this study, the flattest K, steepest K, and mean K (average of flat K and steep K) were recorded and used to assess the corneal power measurement of the new biometer.
Measurements The precision of the new biometer measurements was assessed on the basis of the definitions adopted by the British Standards Institution as recommended by Bland and Altman.13 The subject was seated with his or her chin on a chinrest and forehead against the forehead strap and asked to fixate straight ahead on a fixation target. Each subject was consecutively measured by 2 experienced operators in random order; the entire process took no more than 10 minutes. The machine provided 6 AL measurements with each click, which were averaged to a single value; 1 K measurement for the flat and steep meridians; and ACD and WTW measurements. After each measurement, the device was moved backward and realigned for the next scan to eliminate interdependence of the successive acquisition, and 3 valid scans were performed by both operators. The scans could be influenced by several factors such as eye movements and eyelid blinking. According to the manufacturer, the yellow mark in the acquisition window indicated correctly acquired measurements, whereas the red mark indicated wrong measurements, which were discarded and repeated again. If the obtained scans did not meet the required specifications, they were considered as subqualified scans and repeated. The AL, ACD, K, and WTW values for every qualified measurement were recorded to estimate the repeatability and reproducibility of the new biometer.
Statistical Analysis All data were analyzed using SPSS for Windows software (version 21, SPSS, Inc.). The results are presented as the mean G standard deviations (SD). The distributions of the datasets were checked for normality using the Kolmogorov-Smirnov test, and the results indicated that the data were normally distributed (P O .05). To assess intraoperator repeatability, the similar degree between the outcomes of multiple measurements was performed by a single operator. The within-subject SD, coefficient of variation (CoV), test–retest repeatability, and intraclass correlation coefficient (ICC) were measured. The within-subject SD was the square root of the residual mean square in the 1-way analysis of variance. The test–retest was calculated by multiplying the within-subject value by 2.77, which represented the interval within which 95% of the differences in the measurements were expected to lie.15 The percentage of CoV was calculated as the ratio of the within-subject to the overall mean (the lower the CoV, the
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2235
RELIABILITY OF BIOMETRY USING OLCI
Table 1. Intraoperator repeatability outcomes for biometric measurements obtained in the health group. Parameter
Operator
Mean G SD
Sw
TRT
CoV (%)
ICC (95% CI)
AL (mm)
1 2 1 2 1 2 1 2 1 2 1 2
25.14 G 0.99 25.14 G 0.99 3.73 G 0.24 3.73 G 0.24 43.22 G 1.33 43.23 G 1.33 44.21 G 1.55 44.21 G 1.56 43.72 G 1.42 43.72 G 1.42 11.61 G 0.42 11.63 G 0.42
0.02 0.02 0.03 0.03 0.10 0.08 0.14 0.10 0.10 0.08 0.09 0.10
0.06 0.05 0.09 0.09 0.27 0.22 0.38 0.28 0.28 0.22 0.24 0.28
0.09 0.08 0.82 0.88 0.23 0.19 0.31 0.23 0.23 0.18 0.74 0.88
0.999 (0.999-1.000) 1.000 (0.999-1.000) 0.984 (0.975-0.990) 0.982 (0.971-0.989) 0.995 (0.992-0.997) 0.996 (0.994-0.998) 0.992 (0.988-0.995) 0.996 (0.993-0.997) 0.995 (0.992-0.997) 0.997 (0.995-0.998) 0.959 (0.936-0.975) 0.942 (0.911-0.964)
ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; CoV Z within-subject coefficient of variation; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; Sw Z within-subject standard deviation; TRT Z test–retest repeatability (2.77 Sw); WTW Z white to white
higher the repeatability). The ICC represented the consistency in data measurement, and its value ranged from 0 to 1; a value higher than 0.90 indicated high agreement.16 To assess interoperator reproducibility, the similar degree between the outcomes of measurements was performed by 2 different operators, and the mean of the 3 readings from each session was calculated. The paired t test was used to compare the mean ocular components obtained with the new biometer by 2 operators. Furthermore, agreement between these 2 operators' measurements was assessed per the method described by the BlandAltman plot, and the 95% limits of agreement (LoA) were calculated by the mean difference G1.96 SD. The ICC was also used to evaluate the consistency of measurement by different operators. A P value less than 0.05 was considered statistically significant. The sample size was calculated per Bland and Altman, who recommended at least 24 subjects when 3 repeated measurements are taken and a confidence interval with 20% in the estimate is considered.A
RESULTS The healthy group comprised 52 eyes of 52 healthy subjects (18 men and 34 women) with a mean age of 25.33 years G 1.99 (SD) (range 22 to 33 years). The cataract group comprised 46 eyes of 46 patients with cataract (14 men and 32 women) with a mean age of 65.07 G 8.01 years (range 50 to 82 years).
Intraoperator Repeatability Measurements of the AL, ACD, K values, and WTW with the new biometer showed high intraoperator repeatability with both operators (Table 1). The CoV of all ocular components was lower than 0.89%, and the ICC was higher than 0.94. Measurement of the AL provided the highest repeatability
Healthy Group
Table 2. Intraoperator repeatability outcomes for biometric measurements obtained using Aladdin optical low-coherence interferometer in the cataract group. Parameter
Operator
Mean G SD
Sw
TRT
CoV (%)
ICC (95% CI)
AL (mm)
1 2 1 2 1 2 1 2 1 2 1 2
23.52 G 1.38 23.52 G 1.37 3.20 G 0.40 3.19 G 0.40 43.90 G 1.55 43.93 G 1.55 44.81 G 1.47 44.78 G 1.46 44.35 G 1.47 44.36 G 1.47 11.28 G 0.52 11.24 G 0.60
0.02 0.02 0.05 0.04 0.12 0.18 0.13 0.17 0.09 0.13 0.24 0.29
0.06 0.07 0.15 0.12 0.33 0.51 0.36 0.46 0.25 0.37 0.66 0.80
0.09 0.11 1.70 1.38 0.27 0.42 0.29 0.37 0.21 0.30 2.11 2.58
1.000 (1.000-1.000) 1.000 (0.999-1.000) 0.982 (0.971-0.989) 0.988 (0.980-0.993) 0.994 (0.990-0.997) 0.986 (0.977-0.992) 0.992 (0.987-0.995) 0.987 (0.979-0.992) 0.996 (0.994-0.998) 0.992 (0.987-0.995) 0.814 (0.719-0.885) 0.795 (0.693-0.872)
ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; CoV Z within-subject coefficient of variation; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; Sw Z within-subject standard deviation; TRT Z test-retest repeatability (2.77 Sw); WTW Z white to white
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2236
RELIABILITY OF BIOMETRY USING OLCI
Table 3. Mean difference, paired t test, 95% LOA, and ICC for differences between the 2 operators in the healthy group. Parameter
Mean Difference G SD
AL (mm) ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
0.00 G 0.02 0.00 G 0.03 0.00 G 0.08 0.00 G 0.12 0.00 G 0.08 0.02 G 0.09
P Value
95% LoA
.148 .684 .829 .831 .987 .080
0.03, 0.04 0.06, 0.07 0.17, 0.16 0.22, 0.23 0.17, 0.17 0.19, 0.15
ICC (95% CI) 1.000 (1.000-1.000) 0.990 (0.983-0.994) 0.998 (0.997-0.999) 0.997 (0.995-0.998) 0.998 (0.997-0.999) 0.978 (0.962-0.988)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; LoA Z limits of agreement; WTW Z white to white.
as the test–retest repeatability was lower than 0.06 mm and the ICC between 0.999 and 1.0. Cataract Group Measurements of the AL, ACD, and K
values with the new biometer showed high intraoperator repeatability with both operators. The ICC was higher than 0.98 (Table 2). In contrast to the healthy group, a lower repeatability was detected for WTW measurements; the ICC was about 0.8 and the test–retest repeatability values were 0.66 and 0.8 mm for the 2 operators. Interoperator Reproducibility Healthy Group There was no statistically significant
difference between the 2 operators' measurements (P O .05), and the interoperator reproducibility was excellent for all parameters (Table 3). Bland-Altman plots showed narrow 95% LoA for the AL, ACD, K values, and WTW measurements, and a fixed bias was not detected between the 2 operators (Figures 1, A to 6, A).
Cataract Group There was no statistically significant difference between the 2 operators' measurements. The interoperator reproducibility was high with both operators for all measurements except for WTW (Table 4). Bland-Altman plots showed narrow 95% LoA for the AL, ACD, and K values and wider 95% LoA for WTW (Figures 1, B to 6, B). However, a fixed bias was not detected in WTW measurements between the 2 operators.
Difference Between Healthy and Cataract Groups The 95% LoA of AL, ACD, flat K, steep K, and mean K measurements were slightly narrower in the healthy group than in the cataract group. However, eyes with cataract produced wider 95% LoA for WTW measurements than healthy eyes (Tables 3 and 4). DISCUSSION Refractive cataract surgery requires the most accurate and precise biometric measurements possible. Several
Figure 1. Bland-Altman plots showing agreement in AL measurements between the 2 operators in healthy group (A) and patients in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA. J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
RELIABILITY OF BIOMETRY USING OLCI
2237
Figure 2. Bland-Altman plots showing agreement in ACD measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA.
devices are currently available to fulfill this goal. The Aladdin is a new optical biometer based on OLCI. This is the first study to investigate the intraoperator repeatability and interoperator reproducibility of AL, K values, ACD, and WTW measurements using the new biometer in healthy subjects and patients with cataracts. Excellent repeatability and reproducibility were found in all measured parameters except for WTW in patients with cataracts. In AL and ACD measurements using the new biometer, the study found that test–retest values were slightly lower in the healthy and cataract groups; at the same time, the ICC of repeatability was up to 1 for AL and to 0.98 for ACD measurements in both groups. Another optical device, the IOLMaster, which
measures AL, K values, and ACD, has been compared with the new biometer.12 The IOLMaster is accepted as a gold standard for ocular biometry, and its repeatability and reproducibility have been assessed in several studies.6,17–20 According to Mandal et al.,12 the Aladdin provided repeatable measurements similar to those of the IOLMaster. The interoperator reproducibility of AL and ACD measurements was also found to be excellent. Compared with patients with cataracts, healthy subjects showed slightly narrow 95% LoA. Several authors had already evaluated the reproducibility by other optical biometers. Huang et al.14 assessed the interoperator reproducibility of the AL-Scan and reported that highly reproducible AL and ACD measurements with the ICC
Figure 3. Bland-Altman plots showing agreement in flat K measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA. J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2238
RELIABILITY OF BIOMETRY USING OLCI
Figure 4. Bland-Altman plots showing agreement in steep K measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA.
were higher than 0.99 and the CoV was lower than 0.43%. A more recent study21 showed that the AL-Scan achieved excellent reliability for AL measurements in keratoconic eyes (test–retest values ranged from 0.06 to 0.09 mm for both operators). Bjelos Roncevic et al.22 did similar research of the reproducibility with the Lenstar. It provided reproducible measurements of AL and ACD in patients with cataracts with 95% LoA within a range from 0.02 to 0.02 mm and 0.11 to 0.14 mm, respectively. Buckhurst et al.,7 Srivannaboon et al.8 Goebels et al.,10,11 and S¸ahin et al.23 also reported good reproducibility with the Lenstar, AL-Scan, OA-1000, and IOLMaster. All of these results were similar to the present study results.
Unlike most ocular biometers, the Aladdin can also provide Placido-disk corneal topography. The K values are obtained from 1024 points reflected from rings with an average diameter of 3.0 mm. Previous studies evaluated the repeatability and reproducibility of Placido systems. Tang et al.24 compared 3 Placidodisk videokeratoscopes including the Keratron (Alliance Medical Marketing), Medmont (Medmont Pty Ltd.), and Topographic Modeling System (TMS, Computed Anatomy Inc.) to assess the precision in measuring test surfaces and determined that the Keratron and Medmont had good precision, whereas the TMS showed worse results. In this study, the test– retest values were lower than 0.51 diopter (D) in healthy subjects and patients with cataracts and the
Figure 5. Bland-Altman plots showing agreement in mean K measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA. J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2239
RELIABILITY OF BIOMETRY USING OLCI
Figure 6. Bland-Altman plots showing agreement in WTW measurements between the 2 operators in healthy group (A) and in cataract group (B). The solid line indicates the mean difference (bias). The upper and lower dashed lines represent the 95% LoA.
95% reproducibility LoA ranged from 0.28 to 0.33 D. A previous study reported that a 1.0 D K value measurement error would contribute to 0.8 to 1.3 D IOL power calculation changes.25 The repeatability of these values measured by the Aladdin device was excellent and similar to that in previous studies of other devices. Therefore, the Aladdin demonstrated clinically acceptable repeatability and reproducibility in K value measurement. The repeatability and reproducibility of WTW in the healthy group were good and similar to the other measured parameters. On the contrary, the values were lower in patients with cataracts as the 95% the LoA for interoperator reproducibility was as wide as 0.95 mm. Several studies have assessed the precision of WTW measurement with other optical biometry devices and obtained outcomes similar to those in the present study. Huang et al.14 found that the test–retest repeatability with the AL-scan in patients with cataract was 0.53 mm, and the ICC of reproducibility was 0.897. However, Bjelos Roncevic et al.22 showed that the Lenstar had a lower reproducibility of WTW
measurements (95% LoA, 1.13 to 1.05 mm) in patients with cataracts. Srivannaboon et al.8 also reported the low repeatability and reproducibility of WTW with the AL-Scan and IOLMaster in patients with cataracts. The results showed that the 95% LoA of repeatability and reproducibility were 0.13 to 1.21 mm and 0.94 to 1.06 mm for the AL-Scan, respectively, and 0.27 to 0.45 mm and 0.44 to 0.37 mm for the IOLMaster, respectively. Exact measurement of WTW is important to estimate the horizontal diameter of the anterior chamber for phakic IOL implantation (eg, angle- and sulcussupported phakic IOLs). In these circumstances, selecting accurate IOL size based on WTW value is critical to reduce postoperative complications such as pupil ovalization, endothelial damage, iris retraction, chronic inflammation, and cataract formation.21–23 The Aladdin biometer calculates the WTW by distinguishing the light and shade interface between cornea and sclera; the high prevalence of arcus senilis in elderly patients with cataracts may influence the identification of this edge. Another
Table 4. Mean difference, paired t test, 95% LOA, and ICC for differences between the 2 operators in the cataract group. Parameter AL (mm) ACD (mm) Kf (D) Ks (D) Km (D) WTW (mm)
Mean Difference G SD 0.01 G 0.03 0.01 G 0.04 0.03 G 0.13 0.02 G 0.16 0.00 G 0.11 0.04 G 0.47
P Value
95% LoA
.091 .103 .131 .372 .857 .587
0.05, 0.07 0.06, 0.08 0.28, 0.22 0.28, 0.33 0.23, 0.22 0.88, 0.95
ICC (95% CI) 1.000 (1.000-1.000) 0.996 (0.993-0.998) 0.997 (0.994-0.998) 0.994 (0.990-0.997) 0.997 (0.995-0.998) 0.653 (0.449-0.792)
ACD Z anterior chamber depth; AL Z axial length; CI Z confidence interval; ICC Z intraclass correlation coefficient; Kf Z flat keratometry; Km Z mean keratometry; Ks Z steep keratometry; LoA Z limits of agreement; WTW Z white to white
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2240
RELIABILITY OF BIOMETRY USING OLCI
reason is poor cooperation from older patients who possibly could not follow the operator's instructions. There were several limitations in the present study. First, the study investigated the reliability in healthy subjects and in patients with cataracts. However, the study findings cannot be applied to patients with other diseases such as keratoconus or contact lens usage or those who had corneal refractive surgery. Measurements in eyes with irregular or postsurgical corneas and ocular diseases could be the focus of future research. Moreover, the repeatability and reproducibility parameters derived from corneal topography such as asphericity and aberration have not been investigated. In conclusion, the new noncontact OLCI device Aladdin showed excellent intraoperator repeatability and interoperator reproducibility for all measurements except WTW in eyes with cataract. This suggests that the Aladdin can be used for routine clinical practice to acquire AL, ACD, and K values for IOL power calculation. WHAT WAS KNOWN Currently available optical biometers provide repeatable and reproducible measurements of ocular parameters required to calculate IOL power. WHAT THIS PAPER ADDS The new biometer provided highly repeatable and reproducible measurements of corneal power, ACD, and AL. The precision of WTW measurements was lower in eyes with cataract.
REFERENCES 1. Hoffer KJ. The Hoffer Q formula: a comparison of theoretic and regression formulas. J Cataract Refract Surg 1993; 19:700–712; errata, 1994; 20:677 2. Baumeister M, Terzi E, Ekici Y, Kohnen T. Comparison of manual and automated methods to determine horizontal corneal diameter. J Cataract Refract Surg 2004; 30:374–380 3. Hoffer KJ. Clinical results using the Holladay 2 intraocular lens power formula. J Cataract Refract Surg 2000; 26:1233–1237 4. Retzlaff JA, Sanders DR, Kraff MC. Development of the SRK/T intraocular lens implant power calculation formula. J Cataract Refract Surg 1990; 16:333–340; erratum, 528 5. Congdon NG, Youlin Q, Quigley H, Hung PT, Wang TH, Ho TC, Tielsch JM. Biometry and primary angle-closure glaucoma among Chinese, white, and black populations. Ophthalmology 1997; 104:1489–1495 6. Santodomingo-Rubido J, Mallen EA, Gilmartin B, Wolffsohn JS. A new non-contact optical device for ocular biometry. Br J Ophthalmol 2002; 86:458–462. Available at: http://www.ncbi.nlm. nih.gov/pmc/articles/PMC1771084/pdf/bjo08600458.pdf. Accessed August 12, 2015
7. Buckhurst PJ, Wolffsohn JS, Shah S, Naroo SA, Davies LN, Berrow EJ. A new optical low coherence reflectometry device for ocular biometry in cataract patients. Br J Ophthalmol 2009; 93:949–953 8. Srivannaboon S, Chirapapaisan C, Chonpimai P, Koodkaew S. Comparison of ocular biometry and intraocular lens power using a new biometer and a standard biometer. J Cataract Refract Surg 2014; 40:709–715 9. Kola M, Duran H, Turk A, Mollamehmetoglu S, Kalkisim A, Erdol H. Evaluation of the repeatability and the reproducibility of AL-Scan measurements obtained by residents. J Ophthalmol 2014 article ID:739652. Available at: http://downloads.hindawi. com/journals/joph/2014/739652.pdf. Accessed August 12, 2015 10. Goebels SC, Seitz B, Langenbucher A. Reproducibility of the optical biometer OA-1000 (Tomey). Biomed Res Int 2014 article ID:814761. Available at: http://downloads.hindawi.com/journals/ bmri/2014/814761.pdf. Accessed August 12, 2015 11. Goebels SC, Seitz B, Langenbucher A. Comparison of the new biometer OA-1000 with IOLMaster and Tomey AL-3000. Curr Eye Res 2013; 38:910–916 12. Mandal P, Berrow EJ, Naroo SA, Wolffsohn JS, Uthoff D, Holland D, Shah S. Validity and repeatability of the Aladdin ocular biometer. Br J Ophthalmol 2014; 98:256–258 13. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310. Available at: http://www-users.york.ac.uk/ wmb55/meas/ba.pdf. Accessed August 12, 2015 14. Huang J, Savini G, Li J, Lu W, Wu F, Wang J, Li Y, Feng Y, Wang Q. Evaluation of a new optical biometry device for measurements of ocular components and its comparison with IOLMaster. Br J Ophthalmol 2014; 98:1277–1281 15. Bland JM, Altman DG. Measurement error [Statistics notes]. BMJ 1996; 313:744. Available at: http://www.ncbi.nlm.nih.gov/ pmc/articles/PMC2352111/pdf/bmj00560-0056b.pdf. Accessed August 12, 2015 €ller R, Bu €ttner P. A critical discussion of intraclass correlation 16. Mu coefficients. Stat Med 1994; 13:2465–2476. Available at: http:// www.en.msc-epidemiologie.med.uni-muenchen.de/download/ winter_term_2011_2012/epiresearchdesigns_articles/mueller buttner02_12.pdf. Accessed August 12, 2015 17. Kielhorn I, Rajan MS, Tesha PM, Subryan VR, Bell JA. Clinical assessment of the Zeiss IOLMaster. J Cataract Refract Surg 2003; 29:518–522 18. Sheng H, Bottjer CA, Bullimore MA. Ocular component measurement using the Zeiss IOLMaster. Optom Vis Sci 2004; 81:27–34. Available at: http://journals.lww.com/optvissci/Fulltext/2004/ 01000/The_Contact_Lens_and_Myopia_Progression__CLAMP_ .7.aspx. Accessed August 12, 2015 19. Carkeet A, Saw S-M, Gazzard G, Tang W, Tan DT. Repeatability of IOLMaster biometry in children. Optom Vis Sci 2004; 81:829– 834. Available at: http://journals.lww.com/optvissci/Fulltext/2004/ 11000/Repeatability_of_IOLMaster_Biometry_in_Children.8.aspx. Accessed August 12, 2015 20. Lam AK, Chan R, Pang PC. The repeatability and accuracy of axial length and anterior chamber depth measurements from the IOLMasterä. Ophthalmic Physiol Opt 2001; 21:477–483 cı R, Gu an BD, Balcı M, Heps‚en _IF. €ler E, Kulak AE, Erdog 21. Yag Repeatability and reproducibility of a new optical biometer in normal and keratoconic eyes. J Cataract Refract Surg 2015; 41:171–177 22. Bjelos Roncevic M, Busic M, Cima I, Kuzmanovic Elabjer B, Bosnar D, Miletic D. Comparison of optical low-coherence reflectometry and applanation ultrasound biometry on intraocular lens power calculation. Graefes Arch Clin Exp Ophthalmol 2011; 249:69–75
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
2241
RELIABILITY OF BIOMETRY USING OLCI
€ rsoy H, Bas‚mak H, Yildirim N, Usalp Z, C‚olak E. 23. S‚ahin A, Gu Reproducibility of ocular biometry with a new noncontact optical low-coherence reflectometer in children. Eur J Ophthalmol 2011; 21:194–198 24. Tang W, Collins MJ, Carney L, Davis B. The accuracy and precision performance of four videokeratoscopes in measuring test surfaces. Optom Vis Sci 2000; 77:483–491. Available at: http:// journals.lww.com/optvissci/Fulltext/2000/09000/The_Accuracy_ and_Precision_Performance_of_Four.9.aspx. Accessed August 12, 2015 25. Mehdizadeh M. Effect of axial length and keratometry measurement error on intraocular lens implant power prediction formulas in pediatric patients [letter]. J AAPOS 2008; 12:425; reply by M Eibschitz-Tsimhoni, O Tsimhoni, SM Archer, MA Del Monte, 425 426
OTHER CITED MATERIAL A. Bland JM. Frequently asked questions on the design and analysis of measurement studies. How can I decide the sample size for a repeatability study? Available at: https://www-users.york. ac.uk/wmb55/meas/sizerep.htm. Accessed August 12, 2015.
J CATARACT REFRACT SURG - VOL 41, OCTOBER 2015
First author: Jinhai Huang, MD School of Ophthalmology and Optometry and Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, China