

ORIGINAL ARTICLE 

Year : 2022  Volume
: 2
 Issue : 1  Page : 3539 

Evaluation of prediction accuracy of Barrett Total Keratometry Universal II formula using sweptsource optical biometry
Surekha Mannem, C V Gopal Raju, M Padmini, Ramya Seetam Raju
Department of Ophthalmology, Visakha Eye Hospital, Visakhapatnam, Andhra Pradesh, India
Date of Submission  07Jun2022 
Date of Decision  14Jul2022 
Date of Acceptance  23Jul2022 
Date of Web Publication  05Oct2022 
Correspondence Address: Dr. Surekha Mannem MBBS, DNB Ophthalmology. H.No. 45A38, Aadishankara Nilayam, Palakol, West Godavari District, Andhra Pradesh India
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/jocr.jocr_8_22
Purpose: The purpose of this study was to evaluate the prediction accuracy of Barrett Total Keratometry (TK) Universal II formula using sweptsource optical coherence tomography (SSOCT)based optical biometer. Materials and Methods: One hundred and thirtyfive eyes of 135 patients from Visakha Eye Hospital, Visakhapatnam, India, were prospectively enrolled in this study. Ocular parameters were measured using IOLMaster 700 (Carl Zeiss Meditec, Jena, Germany). Emmetropic spherical equivalent intraocular lens (IOL) power was calculated with Barrett TK Universal II formula using TK and Sanders–Retzlaff–Kraff (SRK)/T formula using standard keratometry. Selected IOL power and predicted refractive errors were recorded. Postoperative manifest refraction was measured at 1 month. Absolute prediction errors (APEs), mean absolute error (MAE), median absolute error (MedAE), and percentage of eyes within ± 0.25, ±0.50, and ±1.00 D of predicted refraction were calculated for each formula. Results: Barrett TK Universal II formula using TK values showed low APEs, MAE, and MedAE. There were a higher percentage of eyes with APE within ±0.25, ±0.50, ±0.75, and ±1.00 D. This formula gave better results when compared to SRK/T formula using K value. However, it was not statistically significant. Conclusion: IOL power calculation using SSOCTbased optical biometer and Barrett TK Universal II gives better results with the least APE and using TK provides superior refractive outcomes which would be beneficial for the patients undergoing phacoemulsification with toric or multifocal IOLs.
Keywords: Barrett Total Keratometry Universal II, intraocular lens power calculation, IOLMaster 700, total keratometry
How to cite this article: Mannem S, Raju C V, Padmini M, Raju RS. Evaluation of prediction accuracy of Barrett Total Keratometry Universal II formula using sweptsource optical biometry. J Ophthalmol Clin Res 2022;2:359 
How to cite this URL: Mannem S, Raju C V, Padmini M, Raju RS. Evaluation of prediction accuracy of Barrett Total Keratometry Universal II formula using sweptsource optical biometry. J Ophthalmol Clin Res [serial online] 2022 [cited 2022 Nov 29];2:359. Available from: http://www.jocr.in/text.asp?2022/2/1/35/357897 
Introduction   
Cataract surgery has evolved to become a refractive surgery in the modern era. Achieving emmetropia is one of the greatest challenges, and this quest for perfection is to restore normal or nearnormal vision after cataract surgery. Accurate biometry and formulas with higher prediction accuracy are the most critical aspects in achieving the lowest prediction error.
For calculating intraocular lens (IOL) power, different formulas have evolved through several generations, from theoretical formulae such as Binkhorst to regression formulae such as Sanders–Retzlaff–Kraff (SRK) I, SRK II, and more advanced mathematical models incorporating elements of both theoretical and regression calculations such as thirdgeneration formulae including SRK/T, Hoffer Q, and Holladay 1, fourthgeneration formulae including Holladay 2, Haigis, and Olsen, and the new fifthgeneration formulae such as Barrett Universal II and HillRBF.^{[1],[2]}>
Graham Barrett developed three new formulae: Barrett Total Keratometry (TK) Universal II (for nontoric IOLs), Barrett TK Toric (for toric IOLs), Barrett True K with TK (for postLVC eyes) Laser vision correction that are based on TK measurements.^{[3]}
Compared to standard keratometry that relies mostly on the anterior corneal curvature, TK additionally considers corneal thickness and posterior corneal curvature.^{[4]} Firstgeneration optical biometers used to rely on different equipment such as Scheimpflugbased corneal topography to measure the total corneal power. With the introduction of the latest generation optical biometer from Zeiss, i.e., IOLMaster 700, it is now possible to measure the posterior corneal curvature directly. TK is a new technology, integrated in IOLMaster 700, that combines telecentric threezone keratometry and sweptsource optical coherence tomography (SSOCT) technology to determine the anterior and posterior corneal surface. This can be advantageous for certain patients who may benefit from more accurate information about their total corneal power (e.g., in astigmatic or postLASIK eyes).^{[5],[6],[7],[8],[9],[10],[11]}
There are many studies that have compared the accuracy of different IOL power calculation formulas.^{[1],[12],[13],[14],[15],[6]} However, there are only limited studies that have evaluated the prediction accuracy of Barrett TK Universal II formula designed to be used with TK value.^{[4],[17],[18]}
The purpose of this study is to evaluate the prediction accuracy of Barrett TK Universal II formula based on TK using IOLMaster 700. We also compared our results with SRK/T formula based on standard keratometry (K).
Materials and Methods   
This prospective, observational study was conducted in adherence to the tenets of the Declaration of Helsinki and approved by the local Institutional Clinical Research Ethics Committee. Informed consent was taken from the study subjects who presented to the Department of Ophthalmology at Visakha Eye Hospital for cataract surgery between December 2019 and June 2021.
Our selection criteria for the study subjects and methods followed the recommendations of recent studies regarding the protocols for studies of IOL formula accuracy.^{[19]}
All the patients with senile cataract over 35 years of age undergoing uneventful phacoemulsification with aspheric monofocal nontoric IOL implantation were included in this study. A single eye from each study subject was taken as the “study eye.”
Patients with cylindrical power more than 1 D, previous keratorefractive procedures or other ocular surgeries, silicone oil implantation, coexisting ocular comorbidities, active ocular infection or inflammation, intraoperative complications, the postoperative manifest refraction of <6/9, the patients who lost to followup, and the patients in whom IOLMaster 700 could not give the quality scans were excluded from the study.
All the patients underwent routine preoperative evaluation. Ocular parameters were measured using IOLMaster 700 (Carl Zeiss Meditec AG, Jena, Germany, software version1.88.1.64861). Emmetropic spherical equivalent IOL power was calculated using Barrett TK Universal II and SRK/T formula. Selected IOL power and predicted refractive errors were recorded. The patients underwent scheduled standard phacoemulsification and implantation of aspheric monofocal IOL (AcrySof IQ SN60WF, Alcon Laboratories, Inc.) performed by a single, experienced surgeon using a 2.2 temporal corneal incision with surgically induced astigmatism (SIA) of 0.12 D to avoid the variability of outcomes. Postoperative manifest refraction was measured at 1month followup. Absolute prediction errors (APE, i.e., the difference between the postoperative manifest refraction and the predicted target refraction), mean absolute error (MAE), median absolute error (MedAE), and percentage of eyes within ±0.25, ±0.50, ±0.75, and ±1.00 D of APE were calculated for each formula.
Statistical analysis
The data required for sample size calculation were collected from the reference article by Fabian and Wehner: ^{[15]} Prediction Accuracy of Total Keratometry Compared to Standard Keratometry Using Different Intraocular Lens Power Formulas. The standard deviation (SD) of Barrett TK Universal II was 0.113; an alpha error was 5%; the estimation error was 0.02. These values were substituted in the following formula to calculate the required sample size:
Minimum sample size − 123 and estimation error − 0.02.
Data were entered in MS Excel and analyzed using SPSS V24. Descriptive statistics were represented with percentages, mean with SD Chisquare test, paired ttest were calculated. P < 0.05 was considered statistically significant.
Results   
One hundred and seventysix eyes of 176 patients were enrolled in this study. Out of which, 11 patients were excluded due to ocular comorbidities. Seventeen patients were excluded due to scan quality issues. Nine patients were lost to followup. Four patients had vision <6/9. Only 135 eyes of 135 patients met the inclusion criteria and were analyzed. [Figure 1] shows the overview of the selection criteria of the patients. [Table 1] summarizes the patient demographics and the baseline characteristics of the ocular biometric measurements.
Absolute prediction error in spherical equivalent calculated with Barrett Total Keratometry Universal II formula
Barrett TK Universal II showed an MAE of 0.266 D (SD – 0.311 D) and MedAE of 0.17 D. As shown in [Table 2] and [Figure 2], APE with Barrett TK Universal II formula using TK data tended toward higher frequencies in the lower error ranges. Overall, a percentage of eyes with APE of 0 were 3% (n = 4), within ± 0.25 D were 62.3% (n = 84), within ± 0.50 D were 85.3% (n = 115), within ± 0.75 D were 95.7% (n = 129), and within ± 1.00 D were 97.9% (n = 132). Only 2.2% (n = 3) showed an APE of > ±1.00 D.  Figure 2: Cumulative percentage of eyes within the specified range of APE in SE for Barrett TK Universal II and SRK/T formula. APE: Absolute prediction error, TK: Total keratometry, SRK: Sanders–Retzlaff–Kraff, SE: Spherical equivalent
Click here to view 
 Table 2: Absolute prediction error in spherical equivalent with Barrett Total Keratometry Universal II and Sanders RetzlaffKraff/T formula
Click here to view 
Absolute prediction error in spherical equivalent calculated with Sanders–Retzlaff–Kraff/T formula
SRK/T showed MAE of 0.271 D (SD – 0.304 D) and MedAE of 0.18 D. As shown in [Table 2] and [Figure 2], APE with SRK/T using standard keratometry data also tended toward higher frequencies in the lower error ranges. A percentage of eyes with APE within ±0.25 D were 58.5% (n = 84) of the study population, within ±0.50 D were 85.2% (n = 115), within ±0.75 D were 93.3% (n = 129), and within ±1.00 D were 96.3% (n = 132). Only 3.7% (n = 3) showed an APE of > ±1.00 D.
Comparison of absolute prediction error in spherical equivalent calculated with Sanders–Retzlaff–Kraff/T and Barrett Total Keratometry Universal II formulas
APE calculated with Barrett TK Universal II formula using TK data and SRK/T formula using standard keratometry data tended toward higher frequencies in the lower error ranges, as shown in [Table 2] and [Figure 2]. The mean APE was lower with Barrett TK Universal II formula based on TK values (0.266, SD – 0.311) compared to SRK/T formula based on K value (0.271, SD – 0.304), with a mean APE difference (K– TK) of 0.005. However, the difference was not statistically significant (P = 0.89). The percentage of eyes with APE in the lower ranges was higher with Barrett TK Universal II formula compared to SRK/T formula. However, the difference was not statistically significant (P = 0.371). There was a good agreement between the two formulas, as shown in [Figure 3].  Figure 3: Agreement between APE with SRK/T and Barrett TK Universal II Formula Bland–Altman plot of APE of Barrett TK Universal LL versus SRK/T. APE: Absolute prediction error, TK: Total keratometry, SRK: Sanders–Retzlaff–Kraff
Click here to view 
Discussion   
In our study, the Barrett TK Universal II formula provided an overall higher predictability of IOL power calculation in terms of the APE, the mean APE, the median APE, and the percentages of eyes within ± 0.25, ±0.5, ±0.75, and ±1.0 D of the APE by applying TK values. There was a good agreement between the Barrett TK Universal II formula using TK value and SRK/T formula using K value with a slightly higher prediction accuracy with Barrett TK Universal II formula.
Previous studies have shown the benefit of using TK derived by SS optical biometry with advanced formulas in predicting postoperative refractive outcomes as shown in [Table 3].  Table 3: Comparison of our study with previous studies on the predictability of the Barrett versus SandersRetzlaff Kraff/T formula
Click here to view 
Fabian and Wehner^{[15]} found that APE with TK was lower than the K data and there were a higher percentage of eyes with lower APE when TK data were used instead of standard keratometry data. These findings were congruent with our results in terms of the prediction accuracy of the TK. However, if we look at the percentage of eyes with APE within ± 0.50 D, our study population got significantly better outcomes. This discrepancy could be possibly due to the involvement of multiple surgeons, different ethnic groups, the higher mean age of their study population, and a different software version.
Another similar study by Srivannaboon and Chirapapaisan^{[20]} found that the TK values resulted in lower MAEs and MedAEs than the K group in all formulas, but there were no statistically significant differences. They also found that the Barrett TK Universal II formula demonstrated the lowest MAEs and higher proportion of eyes within ±0.25, ±0.50, and ±1.00 D of predicted refraction. They concluded that conventional K and TK showed a strong agreement with a trend toward better refractive outcomes using TK.^{[20]} However, it was surprising to see that the proportion of eyes with a prediction of ± 0.5 D for the Barrett TK Universal formula in their study was significantly lower than our study despite following the protocol for studies on IOL formula accuracy, proposed by Hoffer et al.^{[19]} It might be due to the small sample size as the large sample size could have led to better results.
Unlike previous studies, Lee et al.^{[21]} compared the refractive outcomes of cataract surgery with diffractive multifocal IOLs using the standard keratometry and the TK. They found that there was no significant difference between MedAEs or the proportion of eyes within ±0.50 D of predicted refraction from K and TK in each formula and in the subgroup of IOLs. However, they found that the differences in APEs from the K and TK data may vary according to the type of multifocal IOLs and the formula used.^{[21]} Although it was a retrospective study, our results were comparable with the results of their study. Yet, there were some differences, like they used multifocal IOLs and the mean age in their study population was lower than our study population.
Overall, our results suggest that the Barrett TK Universal II formula is one of the most accurate formulas integrated into an advanced SS optical biometer. Using it with the TK will definitely improve the refractive outcomes after cataract surgery.
To the best of our knowledge, our study is the first prospective observational study in the Indian population to assess the accuracy of the new fifthgeneration formula, Barrett TK Universal II, based on TK using a SS optical biometer, the IOLMaster 700.
The present study has some limitations. Since we used monofocal IOLs, we caution that these results may not be generalizable to multifocal and toric IOLs. APE for cylinders was not calculated separately, which would have given more accurate information about the results in patients with astigmatism. Most of the cases were within the normal range of all parameters. Hence, these results may not be applicable to the eyes with extreme parameters such as a very long or short axial length. Further studies with a larger sample size are needed to evaluate the accuracy of the Barrett TK Universal II formula based on the TK.
The following are the strengths of this study. A novel optical biometer (IOLMaster 700; Carl Zeiss Meditec, Jena, Germany) that integrates SSOCT and telecentric keratometry was used in this study. All the surgeries were performed by a single, experienced surgeon with 2.2 clear corneal incisions with SIA of 0.12 D so that the SIA will not have any influence on the outcomes.
Conclusion   
IOL power calculation using SSOCTbased optical biometer and Barrett TK Universal II gives better results with the least APE, and using TK provides superior refractive outcomes which would be beneficial for the patients undergoing phacoemulsification with toric or multifocal IOLs.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References   
1.  Gökce SE, Montes De Oca I, Cooke DL, Wang L, Koch DD, AlMohtaseb Z. Accuracy of 8 intraocular lens calculation formulas in relation to anterior chamber depth in patients with normal axial lengths. J Cataract Refract Surg 2018;44:3628. 
2.  Jeong J, Song H, Lee JK, Chuck RS, Kwon JW. The effect of ocular biometric factors on the accuracy of various IOL power calculation formulas. BMC Ophthalmol 2017;17:62. 
3.  
4.  Holladay JT, Prager TC, Chandler TY, Musgrove KH, Lewis JW, Ruiz RS. A threepart system for refining intraocular lens power calculations. J Cataract Refract Surg 1988;14:1724. 
5.  Retzlaff JA, Sanders DR, Kraff MC. Development of the SRK/T intraocular lens implant power calculation formula. J Cataract Refract Surg 1990;16:33340. 
6.  Hoffer KJ. The Hoffer Q formula: A comparison of theoretic and regression formulas. J Cataract Refract Surg 1993;19:70012. 
7.  Haigis W. Matrixoptical representation of currently used intraocular lens power formulas. J Refract Surg 2009;25:22934. 
8.  Xia T, Martinez CE, Tsai LM. Update on intraocular lens formulas and calculations. Asia Pac J Ophthalmol (Phila) 2020;9:18693. 
9.  Haigis W, Lege B, Miller N, Schneider B. Comparison of immersion ultrasound biometry and partial coherence interferometry for intraocular lens calculation according to Haigis. Graefes Arch Clin Exp Ophthalmol 2000;238:76573. 
10.  Lee AC, Qazi MA, Pepose JS. Biometry and intraocular lens power calculation. Curr Opin Ophthalmol 2008;19:137. 
11.  Olsen T. Prediction of the effective postoperative (intraocular lens) anterior chamber depth. J Cataract Refract Surg 2006;32:41924. 
12.  Barrett GD. An improved universal theoretical formula for intraocular lens power prediction. J Cataract Refract Surg 1993;19:71320. 
13.  IOL Power Calculator for Cataract Surgery  HillRBF Calculator. Available from: https://rbfcalculator.com/. [Last accessed on 2021Jun 19]. 
14.  Kane JX, Van Heerden A, Atik A, Petsoglou C. Accuracy of 3 new methods for intraocular lens power selection. J Cataract Refract Surg 2017;43:3339. 
15.  Fabian E, Wehner W. Prediction accuracy of total keratometry compared to standard keratometry using different intraocular lens power formulas. J Refract Surg 2019;35:3628. 
16.  LaHood BR, Goggin M. Measurement of posterior corneal astigmatism by the IOLMaster 700. J Refract Surg 2018;34:3316. 
17.  Akman A, Asena L, Güngör SG. Evaluation and comparison of the new swept source OCTbased IOLMaster 700 with the IOLMaster 500. Br J Ophthalmol 2016;100:12015. 
18.  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;38:20807. 
19.  Hoffer KJ, Aramberri J, Haigis W, Olsen T, Savini G, Shammas HJ, et al. Protocols for studies of intraocular lens formula accuracy. Am J Ophthalmol 2015;160:4035.e1. 
20.  Srivannaboon S, Chirapapaisan C. Comparison of refractive outcomes using conventional keratometry or total keratometry for IOL power calculation in cataract surgery. Graefes Arch Clin Exp Ophthalmol 2019;257:267782. 
21.  Lee H, Chung JL, Kim YJ, Kim JY, Tchah H. Prediction accuracy of standard and total keratometry by sweptsource optical biometer for multifocal intraocular lens power calculation. Sci Rep 2021;11:4794. 
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]
