Surgically Induced Astigmatism In Surgeon Comparison: A New Visualization Of Surgeon-Specific Sia And Its Implications For Iol Calculation In The Setting Of High Sia Variability

Previous studies on surgically induced astigmatism (SIA) have shown that, under standardized incision parameters, comparative analyses between surgeons are possible. The aim of this retrospective observational study was to use patient data from an ambulatory ophthalmic center to optimize toric intraocular lens (IOL) calculation. The objective was the comparison of SIA between two surgeons. We aimed to evaluate whether the astigmatism induced by surgery differs systematically between the two surgeons. As this is a retrospective exploratory study based on a convenience sample, no formal hypothesis testing in the sense of confirmatory statistics was performed. In addition, this study developed a graphical format for everyday clinical practice

Single-center retrospective study in a high-volume ambulatory ophthalmic surgery center. The study protocol was reviewed and approved by the local ethics committee.

A new visualization format was developed to display SIA using a modified double-angle plot (DAP). The preoperative steep corneal meridian (α) was rounded to integer values, creating 181 categories (0°–180°). For each category, mean and standard deviation of KP(90)_SIA and KP(135)_SIA were calculated. Each point represents the mean, with a line indicating standard deviation; a concentric 0 diopters (D) circle marks unchanged refraction. SIA was calculated per Naeser using key polars KP(90) and KP(135). The concept was inspired by polar trend analysis in glaucoma diagnostics. A total of 2555 eyes were analyzed (surgeon 1: 1619; surgeon 2: 936). Data were collected between March 2018 and December 2024. Biometry used the IOLMaster 500 (Zeiss).

The analysis demonstrates considerable variability, particularly for the KP(90)_SIA component. In some degree categories, clearly positive mean values are observed, whereas in other regions KP(90)_SIA becomes distinctly negative. In contrast, the KP(135)_SIA component remains comparatively more stable. The graphical presentation illustrates that although our assumed SIA value of 0.2 D may be appropriate on average, the variability in individual cases is substantial. The newly developed visualization allows an intuitive assessment of surgeon-specific SIA in relation to the preoperative axis and makes variability directly visible instead of relying on mean values alone.

The new visualization makes surgeon-specific SIA variability directly visible beyond mean values alone. Although 0.2 D reflects the average KP(90)_SIA effect in our cohort, the marked dispersion questions the routine use of a fixed correction factor in toric IOL calculation. Adjustment of SIA values should depend on whether a stable and reproducible surgeon-specific vector pattern can be demonstrated; if variability dominates, assuming 0 D may represent the more neutral approach. The method may also serve as a practical tool to compare surgeons within surgical teams or clinical studies — an aspect often underrepresented in today’s device-driven cataract research.