Intraocular lens (IOL) power calculation plays a vital role in recovering a patient's visual acuity by cataract surgery. Different formulas can be used for IOL power calculation, all depending strongly on the axial length and the corneal power. Hence, a device that accurately and reliably measures these variables will generate a superior outcome for the patient.
One optical biometer capable of measuring the discussed variables is the Pentacam (Oculus), using Scheimpflug technology. It measures the total corneal refractive power (TCRP) using the ray-tracing method, applying Snell's law of refraction to calculate the refractive power at any given point on the cornea; the manufacturer states that TCRP should realistically reflect the actual refractive power. The Scheimpflug device also measures the true net power (TNP), incorporating the curvature of the anterior and posterior corneal surface, and the refractive indices of the different media.1,2 Simulated keratometry (K) is an estimate of the total corneal power based on the assumption that the cornea is a single refracting surface. The anterior corneal curvature and refractive index are used for its calculation.3
Although older optical biometers by Zeiss, such as the IOLMaster 500, only measured the anterior corneal curvature, the Pentacam also takes the posterior corneal curvature into consideration. Studies have shown that the measurement of the posterior corneal curvature reduces systematic measurement error and improves the refractive accuracy, suggesting that patients could benefit from taking the posterior corneal curvature into consideration when measuring the total corneal astigmatism.4,5 The newer optical biometer IOLMaster 700 includes the posterior corneal curvature, and can incorporate the influence of the posterior corneal curvature with the measurement of the total corneal power (total K [TK]).6
The recently introduced IOLMaster 700 measures axial length and corneal power using swept-source optical coherence tomography (SS-OCT) technology. SS-OCT scans the eye with a rapid-cycle, tunable wavelength laser, which is known to give a better signal-to-noise ratio and has improved tissue penetration and image quality.7 Several published studies have shown good results for agreement between the IOLMaster 700 and other devices, as well as high repeatability; however, comparison so far has only looked at the anterior corneal surface and not the total corneal power. In addition, the IOLMaster 700 showed a higher success rate in obtaining measurements in eyes with posterior subscapular and dense nuclear cataracts.8–10
The purpose of this study was to evaluate the repeatability and validity of the corneal power measurements compared with the corresponding variables of both optical biometers. The main focus was on the comparison of TK (IOLMaster 700) with TCRP and TNP (Pentacam).
This prospective controlled study included 93 eyes of 93 volunteers, acquiring proper informed consent for participation of every patient enrolled. All patients were examined in the Department of Ophthalmology, Ludwig-Maximilians-University, Munich, Germany. The study was performed in conduct with the principles of the Declaration of Helsinki and was approved by the local institutional review board.
Exclusion criteria included previous ocular surgery, known corneal irregularities, previous corneal pathology, dry eyes, and a corneal astigmatism above 3.00 diopters (D). In addition, only measurements were included, which achieved the image quality recommended by the manufacturer (for the IOLMaster 700, measurements marked with an exclamation mark were excluded; for the Pentacam, it needed to be marked “OK” by the automated quality check).
One of 3 trained optometrists performed 3 complete automated measurements with each of the 2 devices. The order of measurement was chosen randomly. In between each finished measurement, the patient was asked to stand up, and the position of the device's joystick and chin rest was changed to ensure that the measurements were independent of one another.
For the Pentacam, the patients were asked to fixate on the light, and Scheimpflug pictures were taken as soon as the correct alignment was achieved. For the IOLMaster 700, the patients positioned their chin on the headrest and looked at the fixation light.
The repeatability of the devices used to collect 3 measurements of each eye was determined by calculating the within-subject SD (Sw) according to Bland and Altman.11 In addition, the repeatability was calculated by multiplying the Sw with × 1.96 (=2.77). This value indicates that the difference between 2 measurements for the same patient is expected to be less than 2.77 × Sw for 95% of pairs of measurements.11
The total refractive power TK (IOLMaster 700) was compared with standard K (IOLMaster), simulated K (Scheimpflug), TCRP (Scheimpflug), and TNP (Scheimpflug). In addition, total corneal curvature TR (IOLMaster) was compared with standard R (IOLMaster), simulated R (Scheimpflug), and TNP (Scheimpflug). For each comparison, the validity was evaluated with the Bland–Altman analysis.12 The difference between 2 measurements is plotted over the mean of those 2 measurements to obtain a Bland–Altman diagram. The graph indicated any systematic error, showing how far 2 readings are separated from each other. The coefficient of repeatability (CoR) and the relative CoR (CoR/average measurement) were calculated. Furthermore, limits of agreement (LOA = mean ± CoR) were recorded, which define the variability of the values measured. The analysis of the refractive power also included an independent 2-sample t-test. A P value of less than 0.05 was considered statistically significant.
Another analysis of validity for the comparison mentioned above was performed, emphasizing clinical significance rather than statistical significance. The difference between the first measurement of each of the 2 variables compared was calculated, and all values were highlighted, exceeding a defined limit. For the comparison with TK, any difference in the total corneal astigmatism greater than 0.50 diopters (D) was highlighted, and for the comparison of the axis, any difference greater than 20 degrees, respectively. The sum of cases where a value was highlighted would therefore show which variables when compared had a difference greater than the defined clinically significant limit.
Vector analysis was performed to evaluate the corneal astigmatism, according to Abulafia et al.13 Double-angle plots were generated with the Astigmatism Double Angle Plot Tool.14
The study evaluated 93 healthy volunteers (43 men and 50 women). The mean age was 52.35 ± 19.1 (SD) years (range 18 to 85 years).
Table 1 shows the within-subject SD (Sw) and repeatability (2.77 × Sw) for each variable measured, where a smaller value indicates a better repeatability.
Table 2 shows the P value, CoR, relative CoR, and the limits of agreement of the comparison between the true values (TR and TK measured with the IOLMaster 700) and the corresponding variables. Figure 1 and Figure 2 show the corresponding Bland-Altman plots for each comparison mentioned in Table 2.
The 2-sample t-test for the comparison of astigmatism measured using TK and the astigmatism measured with standard K, simulated K, TNP, and TCRP showed no significant difference between any of the mentioned variables (P = .125 for standard K, P = .153 for simulated K, P = .167 for TNP, and P = .557 for TCRP).
The test we conducted to analyze in how many cases the difference between 2 variables exceeded a clinically significant limit we defined beforehand showed a certain trend. For the difference in astigmatism, 0.50 D was exceeded in 0 cases for the comparison with standard K and in 2 cases for TNP out of the 93 eyes examined. Simulated K exceeded this limit in 4 cases, TCRP in 10. For the difference in meridian, 20 degrees was exceeded in 8 and in 9 cases for standard K and TNP, respectively. In contrast, simulated K exceeded this limit 14 times and TCRP 19 times.
Table 3 shows the astigmatism vector mean of each variable measured with SD and its meridian in degrees. Table 4 shows the difference in the vectors obtained when comparing the different variables, similar to the Bland–Altman analysis. Figure 3 shows the vector analysis with double-angle plots, representing astigmatism measurements obtained using the different corneal power parameters evaluated. Looking at the 95% confidence ellipse of the dataset, standard K and TK look very much alike. This is also the case for the Pentacam's simulated K, TNP, and TCRP.
In this study, we tested the repeatability of variables measured by both the Pentacam and the IOLMaster 700. This analysis was performed, calculating the within-subject SD (Sw). Looking at astigmatism measured with both devices, the repeatability (Sw × 2.77) is very similar throughout, with 0.42 for standard K, 0.40 for TK, 0.45 for TCRP, 0.43 for TNP, and 0.39 for simulated K.
Srivannabonn et al. showed that the repeatability of the IOLMaster 700 was excellent, and compared with the IOLMaster 500, agreement was high. Furthermore, the lens penetration of the IOLMaster 700 was superior compared with its older predecessor.7 Other studies have come to a similar conclusion, demonstrating the high repeatability of both the IOLMaster 700 and the Pentacam.7,8,15 The within-subject SD for standard K found by Kunert et al.16 was slightly larger, where the Sw was 0.20 D compared with 0.15 D in this study. Another study from 2017 also reported a higher within-subject SD for standard K with 0.24 D.17 Reasons for this might be a higher number of patients tested, in addition to 3 measurements of the same eye conducted in comparison to 2 measurements per eye in the study from 2017.
A study from 2016 showed that the highest repeatability of corneal astigmatism measurements was achieved by TCRP measurements with the Pentacam in comparison with various topographers, including the IOLMaster 500. The IOLMaster 700 seems to have closed the gap because our results suggested a slightly better repeatability using TK compared with TCRP.18 However, the results shown in Table 1 also suggest that the repeatability is not given when measuring the meridian of the astigmatism for any of the variables evaluated. Therefore, before calculating a patient's toric IOL cylinder power, repeated measurements of the astigmatism meridian should be taken into consideration to ensure an acceptable outcome.
This study also evaluated the validity of the IOLMaster corneal power measurements (TK and TR) compared with simulated K, TNP, and TCRP. The Bland–Altman analysis has shown no significant difference between the IOLMaster 700's TR and TK compared with the corresponding Pentacam measurements of simulated K, TNP, and TCRP, the P value exceeding .05 in each case.
The Bland–Altman graphs in Figures 1 to 2 show a mean difference near zero, implying that the measurements are somewhat comparable. The mean difference was lowest for the comparison between the IOLMaster 700's TK and the Pentacam's simulated K with 0.047. The comparison between TK and standard K, TNP, and TCRP showed a higher mean difference with 0.481, 0.430, and 0.226, respectively. This suggests that the IOLMaster 700 measurements are most comparable to the Pentacam's simulated K followed by TCRP.
The test to evaluate the differences in total corneal astigmatism exceeding 0.50 D showed a clear trend. Differences between astigmatism measured with TK compared with standard K (0 cases) and TNP (2 cases) did not exceed the limit as often as it was the case with simulated K and TCRP (4 and 10 cases). Differences in astigmatism measurements exceeding a clinically significant limit are concealed in a calculation for the mean difference, as in the Bland–Altman analysis. This test reveals these cases and suggests a different conclusion that standard K and TNP show a higher validity compared with TK in a clinical setting. The differences in meridian show more cases where the limit of 20 degrees was exceeded, no matter the variable compared with TK. This is probably due to the fact that the repeatability of meridian measurements was rather poor, as suggested by the analysis of the within-subject SD. The bigger range in measurements would therefore lead to larger differences in the test conducted. Once again, this would imply that the meridian of the astigmatism should be measured more than once in a clinical setting to enhance the accuracy of treatment.
Analysis of the vectors, as shown in Tables 3 and 4, showed great agreement between the astigmatism vector mean of TK and TNP, with a mean difference of 0.01 ± 0.32 D. Comparison between TK and TCRP (difference of 0.15 ± 0.36 D) and between TK and standard K (difference of 0.19 ± 0.09 D) showed weaker agreement. Together with the results summarized above, this would suggest that TK and TNP are most comparable.
Visser et al.19 compared various topographers using vector analysis and found a mean difference in corneal astigmatism of 0.03 ± 0.29 between standard K with the IOLMaster 500 and simulated K. When we compared standard K with simulated K, we found the same mean difference of 0.03 ± 0.29, although we used the newer IOLMaster 700 and had different mean astigmatism vectors. This would suggest a similar comparability and agreement of the measurements made with both devices.
As the concept of a measured TK is fairly new, there are a few things to consider. Zeiss claims that TK can be incorporated in classic IOL power calculations, using optimized IOL constants, such as User Group for Laser Interference Biometry (ULIB) constants, which was verified by Haigis et al.20 So far, ULIB constants are based on keratometry measurements of the anterior corneal curvature and fixed ratios, rather than the total corneal power, considering the actually measured posterior corneal curvature. Therefore, the question arises how TK can still be used with standard ULIB lens constants and whether TK has been adjusted to do so. Because standard K and TK show no significant difference in astigmatism measured (P = 0.125), it can be assumed that Zeiss has adjusted TK to some degree. The growing use of optical biometers capable of measuring the posterior corneal curvature could affect the ULIB constants in the future, shifting them toward measurements considering both the anterior and posterior corneal curvature. Although the repeatability of both devices has been discussed in several studies, the comparability of total corneal power measurements, such as TK, needs to be explored further, especially in a clinical setting.
WHAT WAS KNOWN
- Because of variability in the posterior corneal curvature, a significant difference exists between measuring the anterior surface alone and simulating the total corneal power compared with measuring the total corneal refractive power (TCRP) directly.
- Scheimpflug devices can measure with a high repeatability the anterior and posterior corneal curvature and TCRP.
WHAT THIS PAPER ADDS
- A recently introduced software update allows a widely used optical biometer to measure the TCRP with high repeatability.
- There is a significant difference between total corneal measurements of the optical biometer and the Scheimpflug device.
1. Tonn B, Klaproth OK, Kohnen T. Anterior surface–based keratometry compared with Scheimpflug tomography–based total corneal astigmatism. Invest Ophthalmol Vis Sci 2015;56:291–298
2. Savini G, Barboni P, Carbonelli M, Hoffer KJ. Comparison of methods to measure corneal power for intraocular lens power calculation using a rotating Scheimpflug camera. J Cataract Refract Surg 2013;39:598–604
3. Herrmann C, Ludwig U, Duncker G. [Corneal topography; analysis of the corneal surface]. Opthalmologe 2008;105:193–206
4. Rydström E, Westin O, Koskela T, Behndig A. Posterior corneal astigmatism in refractive lens exchange surgery. Acta Ophthalmol (Copenh) 2016;94:295–300
5. Koch DD, Ali SF, Weikert MP, Shirayama M, Jenkins R, Wang L. Contribution of posterior corneal astigmatism to total corneal astigmatism. J Cataract Refract Surg 2012;38:2080–2087
6. Abulafia A, Hill WE, Koch DD, Wang L, Barrett GD. Accuracy of the Barrett True-K formula for intraocular lens power prediction after laser in situ keratomileusis or photorefractive keratectomy for myopia. J Cataract Refract Surg 2016;42:363–369
7. Srivannaboon S, Chirapapaisan C, Chonpimai P, Loket S. Clinical comparison of a new swept-source optical coherence tomography–based optical biometer and a time-domain optical coherence tomography–based optical biometer. J Cataract Refract Surg 2015;41:2224–2232
8. Kurian M, Negalur N, Das S, Puttaiah NK, Haria D, J TS, Thakkar MM. Biometry with a new swept-source optical coherence tomography biometer: repeatability and agreement with an optical low-coherence reflectometry device. J Cataract Refract Surg 2016;42:577–581
9. Akman A, Asena L, Güngör SG. Evaluation and comparison of the new swept source OCT-based IOLMaster 700 with the IOLMaster 500. Br J Ophthalmol 2016;100:1201–1205
10. Hirnschall N, Varsits R, Doeller B, Findl O. Enhanced penetration for axial length measurement of eyes with dense cataracts using swept source optical coherence tomography: a consecutive observational study. Ophthalmol Ther 2018;7:119–124
11. Bland JM, Altman DG. Measurement error. BMJ 1996;313:744
12. Martin Bland J, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;327:307–310
13. Abulafia A, Koch DD, Holladay JT, Wang L, Hill W. Pursuing perfection in intraocular lens calculations: IV. Rethinking astigmatism analysis for intraocular lens-based surgery: suggested terminology, analysis, and standards for outcome reports. J Cataract Refract Surg 2018;44:1169–1174
14. Astigmatism Double Angle Plot Tool. ASCRS. American Society of Cararact and Refractive Surgery. Available at: https://ascrs.org/astigmatism-double-angle-plot-tool
. Accessed May 15, 2019
15. Sel S, Stange J, Kaiser D, Kiraly L. Repeatability and agreement of Scheimpflug-based and swept-source optical biometry measurements. Contact Lens Anterior Eye 2017;40:318–322
16. Kunert KS, Peter M, Blum M, Haigis W, Sekundo W, Schütze J, Büehren T. Repeatability and agreement in optical biometry of a new swept-source optical coherence tomography–based biometer versus partial coherence interferometry and optical low-coherence reflectometry. J Cataract Refract Surg 2016;42:76–83
17. Shajari M, Cremonese C, Petermann K, Singh P, Müller M, Kohnen T. Comparison of axial length, corneal curvature, and anterior chamber depth measurements of 2 recently introduced devices to a known biometer—ScienceDirect. Am J Ophthalmol 2017;178:58–64
18. Fityo S, Bühren J, Shajari M, Kohnen T. Keratometry versus total corneal refractive power: analysis of measurement repeatability with 5 different devices in normal eyes with low astigmatism - journal of Cataract & Refractive Surgery. J Cataract Refract Surg 2016;42:569–576
19. Visser N, Berndschot TosTJM, Verbakel F, de Brabander J, Nujits RMMA. Comparability and repeatability of corneal astigmatism measurements using different measurement technologies - ScienceDirect. J Cataract Refract Surg 2012;38:1764–1770
20. Haigis W, Sekundo W, Kunert KS, Blum M. Total keratometric power (TKP) derived from corneal front and back surfaces using a full eye-length SS-OCT scan biometer prototype in comparison to automated keratometry. Presented at the annual meeting of the XXXII Congress ESCRS, London, United Kingdom, September 2014