Comparison of accuracy of a toric calculator with predicted vs measured posterior corneal astigmatism

Prevalence data for total corneal astigmatism in cataract patients

PURPOSE

To report the prevalence data for total corneal astigmatism (TCA) in cataract patients.

METHODS

The authors retrospectively collected and analyzed the preoperative biometric data of the patients who underwent cataract surgery in the Department of Ophthalmology, Peking University Third Hospital, from January 2019 to May 2023.

RESULTS

The mean age of the 10817 patients was 71 ± 10 years; the male/female ratio was 4653/6164. The mean TCA obtained by the IOLMaster 700 (Carl Zeiss Meditec AG, Jena, Germany), the Abulafia-Koch (AK) formula, and the Barrett toric calculator was 1.11 ± 0.81 diopter (D), 1.13 ± 0.75 D, and 1.12 ± 0.74 D respectively, which was significantly greater than the mean standard keratometric (K) astigmatism (0.99 ± 0.75 D) obtained by IOLMaster 700. Against-the-rule (ATR) astigmatism was dominant in all the TCA measurements, and its proportion increased with age. TCA measurements by different methods exhibit high variability, with a total of 1574 (8.9%) data sets from 1016 (9.4%) patients showing a difference larger than 0.5 D in at least one pair of TCA measurements.

CONCLUSION

The use of TCA rather than K astigmatism significantly influenced the choice of intraocular lenses (IOLs) as more patients would be candidates for toric IOLs. It was essential to carefully compare and select TCA obtained with multiple methods for optimal postoperative visual quality.

Accuracy of Toric Intraocular Lens Formulas With Measured Posterior Corneal Astigmatism of Different Orientations

PURPOSE

To assess whether the use of measured posterior corneal astigmatism (PCA) values improves the prediction accuracy of toric intraocular lens power formulas, compared to predicted PCA values, when the orientation of the steep axis of PCA is non-vertical.

DESIGN

Retrospective observational cohort study.

METHODS

Four hundred eighteen eyes of 344 patients were included in the study. Prediction errors (PE) for postoperative refractive astigmatism at 4 weeks postoperatively were determined using vector analysis and compared for the following toric intraocular lens power formulas: Barrett Toric with predicted posterior corneal astigmatism (PPCA); Barrett Toric with measured posterior corneal astigmatism (MPCA); EVO Toric PPCA; EVO Toric MPCA; Holladay I with Abulafia-Koch regression. Subgroup analysis compared PEs for eyes with a vertically orientated steep axis of PCA (60-120°) to eyes with a non-vertically orientated steep axis of PCA.

SETTING

Cathedral Eye Clinic, Belfast, United Kingdom and Tan Tock Seng Hospital, Singapore.

RESULTS

Standard keratometry was with-the-rule in 48% of eyes, while the steep PCA axis was vertically orientated in 91% of eyes. For all eyes, EVO-PPCA had a smaller mean absolute error than Barrett-MPCA, Barrett-PPCA, and Abulafia-Koch (P < .01 for all). EVO-PPCA had the highest percentage of eyes within 0.50D of predicted postoperative astigmatism for eyes with vertical PCA (61%), while EVO-MPCA had the highest percentage for eyes with non-vertical PCA (54%). EVO-MPCA had the smallest centroid error for all eyes, and the subgroups (P < .01 for all). Eyes with non-vertical PCA had a lower percentage within 0.50D than eyes with vertical PCA when using PPCA (43% vs 61%, P = .034), but there was no significant difference between these groups when MPCA is used for eyes with non-vertical PCA (54% vs 61%, P = .40).

CONCLUSIONS

When the steep axis of posterior corneal astigmatism is not vertically orientated, the use of measured posterior keratometry values improves prediction accuracy.

Optimizing IOL Calculators with Deep Learning Prediction of Total Corneal Astigmatism

Background/Objectives: This study aims to identify the most accurate regression model for predicting total corneal astigmatism (TCA) from anterior corneal astigmatism (ACA) and to fine-tune the best model's architecture to further optimize predictive accuracy. Methods: A retrospective review of 19,468 eyes screened for refractive surgery was conducted using electronic medical records. Corneal topography data were acquired using the Pentacam HR. Various types (7) and subtypes (21) of regression learners were tested, with a deep neural network (DNN) emerging as the most suitable. The DNN was further refined, experimenting with 23 different architectures. Model performance was evaluated using root mean squared error (RMSE), R2, average residual error, and circular error. The final model only used age, ACA magnitude, and ACA axis to predict TCA magnitude and axis. Results were compared to predictions from one of the leading TCA prediction formulas. Results: Our model achieved higher performance for TCA magnitude prediction (R2 = 0.9740, RMSE = 0.0963 D, and average residual error = 0.0733 D) compared to the leading formula (R2 = 0.8590, RMSE = 0.2257 D, and average residual error = 0.1928 D). Axis prediction error also improved by an average of 8.1° (average axis prediction error = 4.74° versus 12.8°). The deep learning approach consistently demonstrated smaller errors and tighter clustering around actual values compared to the traditional formula. Conclusion: Deep learning techniques significantly outperformed traditional methods for TCA prediction accuracy using the Pentacam HR. This approach may lead to more precise TCA calculations and better IOL selection, potentially enhancing surgical outcomes.

Comparative Accuracy of Barrett Integrated Keratometry Toric Calculator With Predicted Versus Measured Posterior Corneal Astigmatism

PURPOSE

To compare the prediction accuracy of the Barrett toric calculator using standard or integrated keratometry (IK) mode in combination with predicted or measured posterior corneal astigmatism (PCA) in a group of patients with cataract implanted with non-toric IOLs.

METHODS

In this retrospective clinical cohort study, the medical records of patients with age-related cataract who underwent phacoemulsification with the implantation of an aspheric monofocal IOL were reviewed. Four methods, including standard keratometry with predicted PCA (PPCA), IK combined with predicted PCA (IK-PPCA), and IK combined with measured PCA derived from IOLMaster 700 (Carl Zeiss Meditec AG) or CASIA2 (Tomey) (IK-MMPCA or IK-CMPCA), were applied to the Barrett toric calculator to calculate the predicted residual astigmatism. The mean absolute prediction error (MAPE), centroid of the prediction error, and proportion of eyes within the prediction error of ±0.50, ±0.75, and ±1.00 diopters (D) were all ciphered out from the four methods, respectively.

RESULTS

Data from 129 eyes of 129 patients were included in this study. The MAPE of the IK-PPCA method (0.57 ± 0.36 D) was significantly smaller than that of the PPCA (0.62 ± 0.38 D) and IK-CMPCA (0.63 ± 0.46 D) methods (P = .048 and .014, respectively). There were no significant differences in the centroid vectors of prediction errors and predictability rates among the four methods (all P > .05).

CONCLUSIONS

In the current version of the Barrett toric calculator, the predictive accuracy of the IK mode incorporating PPCA was slightly superior to using the standard keratometry mode or incorporating MPCA. [J Refract Surg. 2024;40(7):e453-e459.].

Accuracy of Toric Intraocular Lens Calculators with Predicted and Measured Posterior Corneal Astigmatism Across Different Types of Astigmatism

INTRODUCTION

This study is a retrospective case series to compare the accuracy of the Barrett toric calculator using predicted posterior corneal astigmatism (PCA) and PCA measurements using swept-source optical coherence tomography (SS-OCT) and a Scheimpflug camera. This evaluation was conducted across different types of anterior and posterior astigmatism.

METHODS

A total of 146 eyes from 146 patients implanted with toric intraocular lenses were included. Mean absolute prediction error, standard deviation of prediction error, and the percentage of eyes with prediction errors within ±0.50 diopters (D) were calculated using vector analysis. Biometric measurements were conducted using the IOLMaster 700 and Pentacam HR. A subgroup analysis was conducted based on the orientation of both anterior and posterior corneal astigmatism.

RESULTS

The Barrett toric calculator with predicted PCA yielded the best results, with 78.1% having a prediction error ≤ 0.50 D, which was a significantly higher percentage than the Barrett formula with the two versions of measured PCA (P < 0.05). In the subgroup with a horizontally steep meridian PCA using the IOLMaster 700, the Barrett formula with predicted PCA yielded the best results, with 78.3% of cases having a prediction error of less than 0.5 D. This percentage was significantly higher than the other two measured PCA subgroups (P < 0.05).

CONCLUSION

The Barrett toric formula with predicted PCA demonstrated a statistically significantly higher proportion of cases with a prediction error ≤ 0.5 D compared to the two measured PCA formulas (from the IOLMaster 700 or Pentacam). This trend persisted even when the posterior corneal astigmatism was horizontally steep.

Accuracy of Toric Intraocular Lens Calculations Using Estimated Versus Measured Posterior Corneal Astigmatism

PURPOSE

To compare the prediction accuracy of toric intraocular lens calculations using estimated vs measured posterior corneal astigmatism (PCA).

DESIGN

Retrospective case series.

METHODS

A total of 110 eyes of 110 patients with uncomplicated toric intraocular lens implantation were included in this study. Predicted postoperative refractive astigmatism was calculated with the Barrett Toric Calculator using the estimated PCA (E-PCA), the measured IOLMaster 700 PCA (I-PCA), and the measured Pentacam PCA (P-PCA). Refractive astigmatism prediction errors (RA-PEs), including their trimmed (tr-) centroid (mean vector), spread (precision), tr-mean absolute RA-PE (accuracy), and percentage within a certain threshold, were determined using vector analysis and compared between groups.

SETTING

University Eye Clinic, Maastricht University Medical Center+, the Netherlands.

RESULTS

The tr-centroid RA-PEs of the E-PCA (0.02 diopter [D] at 82.2°), the I-PCA (0.08 D at 35.5°), and the P-PCA (0.09 D at 69.1°) were significantly different from each other (P < .01), but not significantly different from zero (P = .75, P = .05, and P = .05, respectively). The E-PCA had the best precision (tr-mean 0.40 D), which was not significantly lower than the I-PCA (0.42 D, P = .53) and P-PCA (0.43 D, P = .06). The E-PCA also had the best accuracy (0.40 D), which was not significantly different from the I-PCA (0.42 D, P = .26) and significantly better than the P-PCA (0.44 D, P < .01). The precision and accuracy of the I-PCA did not significantly differ from those of the P-PCA. There were no statistically significant differences in the percentage of eyes within a certain absolute RA-PE threshold.

CONCLUSIONS

The Barrett Toric Calculator using the E-PCA, I-PCA, or P-PCA showed a comparable prediction of postoperative refractive astigmatism in standard clinical practice.

Total keratometry for toric intraocular lens calculation: comparison from two swept-source optical coherence tomography biometers

Purpose

To compare the astigmatism prediction accuracy of total keratometry (TK) from the IOLMaster 700 and total corneal power (TCP) from Anterion based on swept-source optical coherence tomography (SS-OCT) technology in toric intraocular lens (toric IOL) calculation.

Design

A retrospective observational study.

Methods

Total corneal astigmatism (TCA) were obtained using IOLMaster 700 and Anterion. Z CALC 2.0 was used to calculate the expected postoperative refractive astigmatism in conjunction with TCA. Prediction errors (PE) in refractive outcomes was analyzed 1 month postoperatively using the vector analysis by the Holladay method, including the mean vector PE magnitude, percentage of cases with vector PE in certain intervals, and the centroid PE.

Results

A total of 56 eyes from 56 patients were enrolled in the study with an insertion of an AT TORBI 709 toric IOL. The difference in mean vector PE of postoperative refractive astigmatism between TK and TCP was not statistically significant (0.48D versus 0.46D, P = 0.281). TK and TCP yielded 27.3 and 40.0% of eyes with vector PE ≤ 0.25D, and 58.2 and 63.6% with vector PE ≤ 0.5D (both P > 0.05), respectively. TK and TCP resulted in similar ATR centroid PE of 0.10D@35° ± 0.60D and 0.15D@22° ± 0.57D, respectively, and there were no significant differences between x-PE component and y-PE component.

Conclusion

IOLMaster 700 and Anterion provided comparable astigmatic predictability in toric IOL implantation using total keratometry and Z CALC 2.0.

The predictive accuracy of Barrett toric calculator using measured posterior corneal astigmatism derived from swept source-OCT and Scheimpflug camera

PURPOSE

To compare the performance of Barrett toric calculator incorporated with measured posterior corneal astigmatism (PCA) derived from IOL Master 700 and Pentacam HR versus predicted PCA.

METHODS

The predicted residual astigmatism using Barrett toric IOL calculator with predicted PCA, measured PCA from IOL Master 700 and measured PCA from Pentacam were calculated with the preoperative keratometry and intended IOL axis with modification. The vector analysis was performed to calculate the mean absolute prediction error (MAE), the centroid of the prediction error and the percentage of eyes with a prediction error within ±0.50 D, ±0.75 D, and ±1.00 D.

RESULTS

In 57 eyes of 57 patients with mean age of 70.42 ± 10.75 years, the MAE among the three calculation methods were 0.59 ± 0.38 D (Predicted PCA), 0.60 ± 0.38 D (Measured PCA from IOL Master 700) and 0.60 ± 0.36 D (Measured PCA from Pentacam) with no significant difference, either in the whole sample, the WTR eyes and the ATR eyes (F = 0.078, 0.306 and 0.083, p = 0.925, 0.739 and 0.920, respectively). Measured PCA obtained from IOL Master 700 resulted in one level reduction (from Tn to Tn-1) in 49.12% eyes in cylindrical model selection, while measured PCA obtained from Pentacam resulted in one level reduction of toric model selection in 18.18% eyes.

CONCLUSION

The present study suggested that the incorporation of measured PCA values derived from IOL Master 700 and Pentacam produce comparable clinical outcome with the predicted PCA mode in Barrett toric calculator.

Is Astigmatism Correction Necessary for Patients With Cataract Who Have Corneal Astigmatism of Less Than 0.75 D?

PURPOSE

To investigate the proportion of patients with predicted refractive astigmatism (PRA) of 0.75 diopters (D) or greater and associated risk factors among cataract surgery candidates with low corneal astigmatism.

METHODS

A retrospective cross-sectional study was conducted in Zhongshan Ophthalmic Center, Guangzhou, China. Patients with cataract who had preoperative simulated keratometric astigmatism of less than 0.75 D were recruited. The PRA was calculated by Barrett toric calculator using posterior corneal astigmatism (PCA) measured by the IOLMaster 700 (Carl Zeiss Meditec AG) and corneal surgically induced astigmatism (SIA). Two corneal incision locations (temporal [0°/180°], 135° incision) and varying magnitudes (0.10 to 0.60 D) were considered for SIA. Multiple logistic regression analysis was used to explore risk factors associated with PRA of 0.75 D or greater and build predictive model. Sensitivity analysis was performed using PRA threshold of 0.50 D.

RESULTS

A total of 1,750 eyes from 1,750 patients were included (mean age: 60.14 ± 13.24 years, 42.91% male, 1,010 right eyes and 740 left eyes). The 135° incision (odds ratio [OR]: 17.86) and against-the-rule (ATR) astigmatism (OR: 37.55) are the major risk factors for PRA of 0.75 D or greater. Higher simulated keratometric astigmatism (OR: 2.03), larger PCA (OR: 1.64), and surgically induced astigmatism (OR: 1.29) also significantly increased the risk of PRA of 0.75 D or greater. Nomogram model were constructed with an area under curve of 0.90.

CONCLUSIONS

For patients with corneal astigmatism of less than 0.75 D, temporal incision and measured PCA is preferred. Those patients with ATR astigmatism should be considered for astigmatism correction when using a 135° incision. [J Refract Surg. 2023;39(12):850-855.].