Comparison of Anterior Segment Measurements with Optical Low-coherence Reflectometry and Partial-coherence Interferometry Optical Biometers.

AIMS: To evaluate a new noncontact optical biometer using partial-coherence interferometry and to compare the clinical measurements with those obtained from the device using optical low-coherence reflectometry (OLCR). SETTING AND
DESIGN: Ondokuz Mayis University, Samsun, Turkey. Nonrandomized, prospective clinical trial. SUBJECTS AND METHODS: The study was performed on the healthy phakic eyes of volunteers in the year 2014. Measurements of axial length (AL), anterior chamber depth (ACD), central corneal thickness (CCT), mean keratometry (K), and white-to-white (WTW) measurements obtained with the low-time coherence interferometry (LTCI) were compared with those obtained with the OLCR. STATISTICAL ANALYSIS USED: The results were evaluated using Bland-Altman analyses. The differences between both methods were assessed using the paired t-test, and its correlation was evaluated by Pearson's coefficient.
RESULTS: We examined seventy participants with a mean age of 33.06 (±9.7) (range: 19-53) years. AL measurements with LTCI and OLCR were 23.7 (±1.08) mm and 23.7 (±1.1) mm, respectively. ACD was 3.6 (±0.4) mm and 3.5 (±0.4) mm for LTCI and OLCR, respectively. The mean CCT measurements for both devices were 533 (±28) mm and 522 (±28) mm, respectively. The mean K readings measurements for LTCI and OLCR were 43.3 (±1.5) D and 43.3 (±1.5) D, respectively. The mean WTW distance measurements for both devices were 12.0 (±0.5) mm and 12.1 (±0.5) mm, respectively.
CONCLUSIONS: Measurements with LTCI correlated well with those with the OLCR. These two devices showed good agreement for the measurement of all parameters.

Introduction

Accurate biometry measurements and the intraocular lens (IOL) calculation are of great importance in modern cataract and refractive surgery.[1] For many years, ultrasound measurements have been the gold standard for axial length (AL) and anterior chamber depth (ACD) measurement. The IOL MASTER (Carl Zeiss Meditec AG, Jena, Germany) was first introduced in 1999, and several studies have shown the accuracy and advantages of partial-coherence interferometry.[2] The instrument has been improved several times and considered as the gold standard optical biometer. In the year 2008, the LENSTAR LS 900 biometer (Haag-Streit AG, Koeniz, Switzerland) that uses optical low-coherence reflectometry (OLCR) was introduced. The instrument can measure AL, central corneal thickness (CCT), ACD (from the corneal endothelium to the anterior lens surface), lens thickness (LT), retinal thickness (RT), white-to-white (WTW) distance, and keratometric (K) readings. It also measures the pupil size (PS) and centricity. All parameters are measured in a single step with a single alignment. Previous studies show that the IOL MASTER 500 and the LENSTAR LS 900 provide reliable intraobserver and interobserver measurements and can be used interchangeably.[3456] Hence, LENSTAR LS 900 has also been accepted as a gold standard optical biometer for the last few years. Recently, a new optical biometer that uses low time-coherence interferometry (LTCI) became available. The AL-SCAN (NIDEK Co., Ltd., Gamagori, Japan) is one of the most recent diagnostic equipment added to the cataract surgeon's diagnostic equipment world. The purpose of this study was to evaluate and compare the AL-SCAN with the LENSTAR LS 900 in a cohort of volunteers with healthy eyes.

Subjects and Methods

The study was conducted between February and April 2014 on the healthy phakic eyes of volunteers. All examinations were performed by a single practitioner who was trained according to the manufacturer's recommendations. The Ethics Committee of Ondokuz Mayis University, Samsun, Turkey, approved the study and written informed consent was obtained from each participant after the study had been fully explained. The study was conducted in accordance with the Declaration of Helsinki. Participants having corneal pathology, previous refractive surgery, or any abnormality in the anterior segment were excluded from the study. To compare the two devices, AL, ACD, CCT, WTW distance, and (K) readings were analyzed. Three consecutive measurements were taken per eye. As the AL-SCAN measures ACD from the front surface of the cornea and LENSTAR LS 900 measures ACD from the endothelium, the corneal thickness calculated by the LENSTAR LS 900 was added to its anterior chamber measurement from the back surface of the cornea.

Instruments

The AL-SCAN is based on LTCI. It measures six values (AL, K readings, ACD, CCT, WTW distance, and PS) within 10 s while using the auto-alignment function. It consecutively takes AL measurements up to 20 times at maximum. The eyes were in focus when the instrument head was approximately 4.5 cm away from the patient's eyes. A red led (700 nm) is used as fixation beam. The device uses LTCI to measure AL and lens thickness using the 830 nm superluminescent diode. Corneal thickness and ACD of the patient's eye are measured using the Scheimpflug principle. For keratometry readings, a mire-ring led (970 nm) is projected on the patient's cornea with a photodetector and K readings are measured with 360 data points by calculating the image. The WTW distance and PS of the patient's eye are measured based on a captured anterior eye segment image. For WTW distance measurements, a 525 nm illumination is projected on the eye, illuminated anterior eye image is captured with a camera, and the distance is measured with image processing. For PS measurements, the device uses a 970 nm illumination with the same processing. The AL-SCAN also comes with the option of a built-in ultrasound biometer and/or an ultrasound pachymeter for denser cataracts. The principles of the LENSTAR LS 900 have been described in other papers before.[56789]

Statistical analysis

Statistical analysis was performed using the Statistical Package for Social Studies (SPSS for Windows, version 15.0, SPSS, Chicago, IL, USA). The Pearson's correlation coefficients (r) were calculated to evaluate each correlation. Bland–Altman method that suggests plotting the differences between the measurements (y-axis) against their mean (x-axis) was used to evaluate agreement. The 95% limits of agreements (LoA) were defined as the mean ± 2 standard deviation (SD) of the differences between the two measurement techniques.[10] Intraobserver reliability between three consecutive AL-SCAN measurements was determined by calculating the mean SD between three consecutive measurements (SD within). P < 0.05 was considered significant.

Results

We included seventy adult volunteers with a mean age of 33.1 (±9.7) (range: 19–53) years.

Biometry measurements assessed by the AL-SCAN and the LENSTAR LS900 unit are presented as the mean with 95% confidence interval and range of each of the parameters in Table 1.

Table 1

Intrasession (three repeats; n=70) average standard deviation of repeated measurements with the AL-Scan

A mean AL of 23.7 (±1.1) mm was found with the AL-SCAN compared with 23.7 (±1.1) mm for the LENSTAR LS 900. Bland–Altman plot showed that the agreement was excellent (95% LoA, −0.06 − 0.07 mm). The mean difference in AL measurements was + 0.01 mm (P < 0.05; 95% CI, 0.003–0.013) [Figure 1]. An excellent correlation was observed regarding the AL measurements (r = 0.999 and P < 0.0001).

Figure 1

Bland–Altman plot of axial length measurement of the AL-SCAN compared with LENSTAR LS 900. The bold horizontal line demonstrates the mean difference between the AL-SCAN and the LENSTAR LS900. The dotted lines above and below represent the 95% limits of agreement interval

The mean ACD measurements were 3.6 (±0.4) mm and 3.5 (±0.4) mm for the AL-SCAN and the LENSTAR LS 900, respectively. The mean difference in ACD measurements was + 0.05 mm (P < 0.05; 95% CI, −0.011 − 0.027) [Figure 2]. Measurements by the two devices regarding the ACD measurements showed an excellent correlation (r = 0.966 and P < 0.0001).

Figure 2

Bland–Altman plot of anterior chamber depth measurement of the AL-SCAN compared with LENSTAR LS 900. The bold horizontal line demonstrates the mean difference between the AL-SCAN and the LENSTAR LS900. The dotted lines above and below represent the 95% limits of agreement interval

The mean CCT measurements for both imaging devices were 533 (±28) mm and 522 (±28) mm, respectively. The mean difference in CCT measurements was + 12 μm (P < 0.0001; 95% CI, −5 − 28.9). Although measurements by the two devices were highly correlated (r = 0.96 and P < 0.05), the AL-SCAN measured a slightly thicker CCT than the LENSTAR LS 900 [Figure 3].

Figure 3

Bland–Altman plot of central corneal thickness measurement of the AL-SCAN compared with LENSTAR LS 900. The bold horizontal line demonstrates the mean difference between the AL-SCAN and the LENSTAR LS900. The dotted lines above and below represent the 95% limits of agreement interval

The mean K readings measurements for the AL-SCAN and LENSTAR LS 900 were 43.3 (±1.45) D and 43.3 (±1.5) D respectively. The mean difference in K readings was 0.04D (P < 0.05; 95% CI, −0.34 − 0.27) [Figure 4]. An excellent correlation was observed regarding the K readings (r = 0.994 and P < 0.0001).

Figure 4

Bland–Altman plot of mean keratometry readings measurement of the AL-SCAN compared with LENSTAR LS 900. The bold horizontal line demonstrates the mean difference between the AL-SCAN and the LENSTAR LS900. The dotted lines above and below represent the 95% limits of agreement interval

The mean WTW distance measurements for both devices were 12.0 (±0.5) mm and 12.1 (±0.5) mm, respectively. The mean difference in WTW distance was 0.13 mm (P < 0.05; 95% CI, −0.43 − 0.16). An excellent correlation was observed regarding the WTW distances (r = 0.956 and P < 0.0001).

Intraobserver reproducibility analysis of the measurements was highly reproducible with for each of the parameters with the AL-SCAN [Table 2].

Table 2

Average with 95% confidence interval and range of each of the parameters as assessed by the AL-Scan and LenStar LS 900 optical biometers

Discussion

For better planning of cataract surgery, precise measurements of AL, ACD, and keratometry (R1 and R2) are important parameters for the accurate calculation of the IOL power. Keratometry errors and incorrect AL determination were the most common causes of unsatisfactory refractive outcomes of cataract surgery.[11]

The measurements performed in this study showed an excellent correlation for AL, ACD, CCT, mean K readings, and WTW distance. In general, measurements by the AL-SCAN were slightly larger as compared with the LENSTAR LS 900.

AL as measured by the AL-SCAN was slightly higher than the LENSTAR LS 900 measurements (mean difference + 0.01 mm) and this is an extremely low mean difference. Hill et al.[12] reported that a 0.01 mm difference in AL equals only 0.03 D difference in IOL power calculation.

In spite of a good correlation, there was only moderate agreement between the two instruments in CCT measurements. The AL-SCAN measured a deeper CCT (mean difference 12 μm). The AL-SCAN could be expected to read as much as 28.9 μm above to below 0.5 μm the LENSTAR LS 900 for the CCT. Although it is statistically significant, we think that this reported difference is small and not clinically relevant. Such a difference might be the result of the CCT measurement technique used by each instrument. The AL-SCAN measures CCT with the Scheimpflug principle, while the LENSTAR measures CCT with OLCR using the 820 mm superluminescent diode.

Although the ACD measurement technique used by each instrument is different, there was a good agreement between the two instruments. The LENSTAR LS 900 uses the OLCR method as described previously.[56] The ACD of the patient's eye is measured using the Scheimpflug principle in AL-SCAN. Like the CCT, the AL-SCAN measures the ACD with the Scheimpflug principle. Huang et al. showed that the CCT and ACD measurements with the biometer and Scheimpflug system could be used interchangeably in healthy young participants.[13] Our data are consistent with that study comparing Scheimpflug principle with the OLCR for both CCT and ACD measurements. In a study comparing ACD measurements in phakic and pseudophakic eyes with the IOL MASTER, Kriechbaum et al. found no difference in phakic patients but a greater difference and no correlation in pseudophakic eyes.[14] Since our sample group had healthy eyes with a clear crystalline lens, this situation might have caused similar problems during the measurement of ACD in pseudophakic eyes. Hence, further studies targeting such cases are recommended.

For keratometry measurements with the LENSTAR LS 900, K reading is calculated by 32 projected light reflections arranged on two rings with 16 measuring points each. The inner and the outer rings are 1.65 mm and 2.3 mm in diameter, respectively. In the AL-SCAN, mire-ring led (970 nm) is projected on a cornea and K reading is measured with 360 data points with an inner ring having a diameter of 2.4 mm and an outer ring having a diameter of 3.3 mm. Mean K reading measurements assessed by the AL-SCAN were similar to those determined with the LENSTAR LS 900. In their studies comparing Lenstar LS 900 and IOL Master, Buckhurst et al.[5] and Holzer et al.[6] reported mean K reading differences as 0.04 D and 0.05 D, respectively. Our 0.04 D mean K reading difference was consistent with those results.

The AL-SCAN and LENSTAR LS 900 were also found to measure equivalent values for WTW distance.

Although we did not make an objective comparison between two devices for the duration of the measurement process, all measurements took a longer time with the LENSTAR LS 900 than with the AL-SCAN.

Our study had few limitations. First, our study population did not include eyes having cataract, pseudophakia, or previous refractive surgery. Second, we did not investigate short eyes (<20.5 mm) or long eyes (>27.0 mm) in which there might be different results. Performing tests on healthy individuals would be a good starting point. The next step to further validate the machine must be to compare the two instruments in eyes with different ALs and to use in eyes with cataract to measure IOL power, which is the most important function of biometers.

Conclusions

The AL-SCAN provided clinically interchangeable measurements with those provided by the LENSTAR LS 900, which is currently used as a reliable optical biometer. Except a subjective finding that the AL-SCAN is less time consuming, we did not find any distinctive evidence of differences between the two devices. The repeatability of the device for all reported measurements is excellent. Since our study population comprised only healthy volunteers, studies with large sample sizes should be executed in patients with different conditions.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.