ORIGINAL ARTICLE Year : 2014  Volume : 62  Issue : 9  Page : 931934 Comparison of two optical biometers in intraocular lens power calculation Sheng Hui, Lu Yi Department of Ophthalmology, Eye and ENT Hospital, Fudan University, Shanghai, China Correspondence Address: Aims: To compare the consistency and accuracy in ocular biometric measurements and intraocular lens (IOL) power calculations using the new optical lowcoherence reflectometry and partial coherence interferometry. Subjects and Methods: The clinical data of 122 eyes of 72 cataract patients were analyzed retrospectively. All patients were measured with a new optical lowcoherence reflectometry system, using the LENSTAR LS 900 (Haag Streit AG)/ALLEGRO BioGraph biometer (Wavelight., AG), and partial coherence interferometry (IOLMaster V.5.4 [Carl Zeiss., Meditec, AG]) before phacoemulsification and IOL implantation. Repeated measurements, as recommended by the manufacturers, were performed by the same examiner with both devices. Using the parameters of axial length (AL), corneal refractive power (K1 and K2), and anterior chamber depth (ACD), power calculations for AcrySof SA60AT IOL were compared between the two devices using five formulas. The target was emmetropia. Statistical analysis was performed using Statistical Package for the Social Sciences software (SPSS 13.0) with ttest as well as linear regression. A P value < 0.05 was considered to be statistically significant. Results: The mean age of 72 cataract patients was 64.6 years ± 13.4 [standard deviation]. Of the biometry parameters, K1, K2 and [K1 + K2]/2 values were significantly different between the two devices (mean difference, K1: −0.05 ± 0.21 D; K2: −0.12 ± 0.20 D; [K1 + K2]/2: −0.08 ± 0.14 D. P <0.05). There was no statistically significant difference in AL and ACD between the two devices. The correlations of AL, K1, K2, and ACD between the two devices were high. The mean differences in IOL power calculations using the five formulas were not statistically significant between the two devices. Conclusions: New optical lowcoherence reflectometry provides measurements that correlate well to those of partial coherence interferometry, thus it is a precise device that can be used for the preoperative examination of cataract patients.
Subjects and Methods In this retrospective study, seventytwo cataract patients (122 eyes) who received phacoemulsification and IOL implantation were analyzed. Measurement technique During the preoperative examination, biometry was first performed using an IOLMaster V5.4 followed by assessment with the LENSTAR device. Patients were excluded if measurement of any parameter was not possible to obtain with either device. Five AL and ACD measurements, and three keratometry measurements, were taken using the IOLMaster, followed by five consecutive measurements with the LENSTAR. The same examiner, who was trained according to the manufacturer's recommendations, performed all the biometry testing. The AcrySof SA60AT IOL (Alcon, Inc.) was used for the purpose of IOL power calculations and the target in each patient was emmetropia. The IOL powers were calculated with both biometers using five formulae: SRK II, SRK/T, Hoffer Q, Holladay, and Haigis. Statistical analysis Statistical analysis was performed using Statistical Package for the Social Sciences software (SPSS 13.0) with ttest as well as linear regression. A P < 0.05 was considered to be statistically significant. Results We measured 160 cataract eyes. 28 eyes were excluded from analysis due to measurements failures for IOLMaster, and 38 eyes were excluded for LENSTAR. 122 eyes were successfully measured by both IOLMaster and LENSTAR. The mean age of the patients was 64.6 years ± 13.4 (standard deviation). The mean values and standard deviations for AL, K1, K2, and ACD with both biometers, as well as the difference between the devices, are shown in [Table 1]. The two biometers provided generally similar and reliable results. The differences of K1, K2 and [K1 + K2]/2 between the two devices were statistically significant (mean difference, K1= −0.05 ± 0.21 D; K2= −0.12 ± 0.20 D; [K1 + K2]/2= −0.08 ± 0.14 D. P <0.05; ttest). The difference of AL between the two devices was (0.02 ± 0.10 mm) and (−0.97 to 0.33 mm). The difference of ACD between the two devices was (−0.02 ± 0.17 mm) and (−0.68 to 0.58 mm). The differences of AL and ACD between the two devices were not found to be statistically significant (P < 0.05; ttest).{Table 1} [Table 2] shows the SA60AT IOL power calculation results using the five formulas. The differences of IOL power calculation between the two devices were 0.00 ± 0.17 (SRK/formula), 0.01 ± 0.22 (SRK/T formula), 0.03 ± 0.25 (Hoffer Q formula), 0.03 ± 0.22 (Holladay formula) and 0.04 ± 0.25 (Haigis formula), respectively. There were no statistically significant differences between the two devices (P > 0.05; ttest). [Figure 1] demonstrates the distribution of the differences in SA60AT IOL power for each formula. Overall, a mean of 45.6% ±5.8% of eyes were within a ± 0.10 D difference; 71.2% ±8.2% were within ± 0.20 D; 85.9% ±5.4% were within ± 0.30 D; 93.6% ±2.3% were within ± 0.40 D, 96.7% ±1.7% were within ± 0.50 D; 99.5% ±0.4% were within ± 1.00 D, and all were within ±1.50 D. The difference in SA60AT IOL power calculations, between the two devices, was lowest with the SRK II formula, and was highest with the Haigis formula. [Figure 2] shows the correlations of the AL, K1, K2, and ACD measurements between the LENSTAR and IOLMaster. Linear regression showed an excellent correlation. The Pearson r for AL, K1, K2, and ACD were 0.999, 0.991, 0.992, and 0.927, respectively.{Table 2}{Figure 1}{Figure 2} [Figure 3] shows the BlandAltman plots for the SA60AT IOL power calculations. There was good agreement between the LENSTAR and IOLMaster in terms of power calculation.{Figure 3} Discussion Accurate measurements of AL, K, and ACD are essential in modern cataract surgery. Both IOLMaster and LENSTAR are based on noncontact laser interferometry to assess AL. The IOLMaster uses PCI in a dualbeam configuration to measure AL, and it is powered with a multimode laser diode, [2],[3] whereas the LENSTAR uses OLCR powered with an super luminescent diode (SLD). [10],[11] With the SLD, reflective structures within the cornea, anterior chamber, lens, and retina are measured. Hence, the LENSTAR can simultaneous measure AL, ACD, keratometry, and whitetowhite distance; and also corneal, lens, and retinal thickness parameters, and distance from the endothelium to the anterior surface of the lens. Altogether 160 cataract eyes were measured. Of which 28 eyes were excluded from analysis due to measurements failures for IOLMaster, and 38 eyes were excluded for LENSTAR. 122 eyes were successfully measured by both IOLMaster and LENSTAR. Dense cataracts, poor fixation, and unable to cooperate with examination were causes of measurement failures. Due to the different technologies used, the IOLMaster requires 3 different positions and releases procedures, whereas the LENSTAR acquires all parameters in a single position with one release procedure. Hence, measurements with the LENSTAR need better cooperation of the patients than IOLMaster. In a comparison of the two devices in 200 phakic healthy eyes without cataract, Holzer et al., [10] found a good correlation in AL, keratometry, and ACD measurements. A study by Buckhurst et al., [11] evaluated 112 cataract patients using both devices. The corneal curvature measurements were similar. AL and ACD values were significantly higher with the OLCR device, than with the PCI device, but the differences were not clinically significant. In a subgroup of 32 patients, the OLCR measurements were highly repeatable. Rabsilber et al., [12] evaluated IOL power calculations on 100 cataract eyes using parameters obtained from the LENSTAR and IOLMaster. Of all the biometry parameters, the only statistically significant differences between the two devices were in terms of corneal radii and ACD. The mean difference in IOL power calculations using four formulas were not statistically significant between the two devices. Rohrer K et al., [13] evaluated LENSTAR in 144 eyes of 80 persons and compared the measurements with those obtained from the IOLMaster. Measurements with LENSTAR and IOLMaster for AL, ACD, corneal radius, and axis of the flattest radius in cataractous, pseudophakic, aphakic, silicon oilfilled, and normal eyes correlated well. These previous studies support our findings of reliable measurements using the LENSTAR biometry device. In our study, of the biometry parameters, there was no statistically significant difference in AL and ACD between the two devices for cataract patients who received phacoemulsification and IOL implantation. The correlations of AL, K1, K2, and ACD between the two devices were high. K1, K2 and [K1 + K2]/2 values were significantly different between the two devices, but the differences were not clinically significant. The mean differences in IOL power calculations using the five formulas were not statistically significant between the two devices. The difference between the two devices in ACD measurement might be related to the difference in the measuring technique of the two devices. While the IOLMaster uses lateral slit illumination to determine the distance between the corneal epithelium and the anterior surface of the crystalline lens, the LENSTAR uses OLCR technology to detect the corneal thickness from epithelium to endothelium and the distance from the endothelium to the anterior surface of the crystalline lens which represents the anatomic ACD. Furthermore, the keratometry is required before ACD measurement with IOLMaster. As for corneal refractive power measurement, both IOLMaster and LENSTAR evaluate corneal radius in the flat and steep meridian. The LENSTAR takes readings in two circles, 16 points in each circle, for a total of 32 readings. The inner circle has a diameter of 1.65 mm, and the outer circle has a diameter of 2.3 mm. However, the IOLMaster takes readings from 6 points in one circle which has a diameter of 2.3 mm. In our study, the differences of K1, K2 and [K1 + K2]/2 between the two devices were statistically significant; however, in practice, there was no clinical significance. A limitation of both devices is that they are unable to accurately measure severely compromised eyes, such as those with retinal detachment, severe opacities along the visual axis, and in case of poor patient cooperation. [7] In summary, the new optical lowcoherence reflectometry LENSTAR device provides precise biometry and IOL power calculation in cataract patients, and has good correlation with measurements obtained using the partial coherence interferometry IOLMaster. Therefore, the LENSTAR is a useful new tool for accurate IOL power calculation in cataract patients. References


