Accurate keratometry is crucial for planning toric intraocular lenses (IOLs) in cataract surgery. Approximately 20% of patients have corneal astigmatism of 1.5 diopters (D) or more, marking the need for precise assessment. Errors in cylinder power or misalignment can significantly reduce astigmatic correction. A misalignment of 10 degrees may reduce intended correction by one-third, illustrating the importance of minimizing variability among measurement devices.
Several devices, including the Pentacam HR, IOLMaster 500, and KR-800, utilize different measurement principles, leading to potential discrepancies in estimates. Previous studies have provided mixed conclusions regarding the interchangeability of these devices, often conducted under controlled conditions that may not represent routine clinical practice. This study aims to evaluate the agreement of keratometry measurements from these devices in real-world settings and their implications for toric IOL planning.
A retrospective study was carried out at Inje University Busan Paik Hospital. Records from patients who underwent preoperative evaluations for cataract surgery between August 2021 and December 2023 were analyzed. Inclusion criteria comprised adults aged 19 or older who had full keratometry measurements from all three devices within the same day. Patients with previous corneal surgeries or pathologies, as well as those with poor measurement quality, were excluded. Ethical approval was obtained from the institutional review board, and informed consent was waived due to the retrospective nature of the research.
Measurements were conducted using three devices:
All measurements were performed by trained technicians without a fixed sequence, maintaining the integrity of routine clinical workflow.
Astigmatism data from each device was collated and converted into power vector components (J0 and J45) using standard formulas. This approach allowed a fair statistical comparison of astigmatism data across devices. Measurements were taken sequentially within a single session to mitigate variability associated with timing and conditions.
The study analyzed 227 eyes, revealing no significant differences at the population level among the devices (J0: p = 0.138; J45: p = 0.233). The corresponding intraclass correlation coefficients (ICCs) were 0.894 for J0 and 0.748 for J45. However, the 95% limits of agreement for the devices were relatively wide (0.80–1.21 D). A notable finding was that 64.8% of patients showed predicted residual astigmatism exceeding 0.50 D when using various device pairs. Moreover, the Pentacam HR designated fewer eyes as candidates for toric IOLs compared to the reflection-based devices.
Among those classified as toric IOL candidates (1.0 D astigmatism or greater), 21–33% exhibited axis differences greater than 10°, a threshold associated with a significant reduction in correction. These axis discrepancies were inversely correlated with astigmatism magnitude (Spearman’s rho = -0.220 to -0.272; all p-values < 0.001), indicating the importance of device selection in surgical planning.
The findings highlight acceptable agreement at the population level but illustrate clinically significant variations at the individual level. The average magnitude of disagreement among measurements suggests that reliance on a consistent measurement device might be prudent for surgical planning. While the clinical implications of the observed differences need further validation, these results emphasize the variability inherent in different keratometry devices.
Using a single keratometry device for toric IOL planning may better align with optimal clinical outcomes due to individual-level discrepancies observed among devices. Further prospective studies are warranted to evaluate the clinical impact of these differences and solidify recommendations for device usage in surgical practices.