Agreement between total corneal astigmatism calculated by vector summation and total corneal astigmatism measured by ray tracing using Galilei double Scheimpflug analyzer

PURPOSE: To evaluate the agreement between total corneal astigmatism calculated by vector summation of anterior and posterior corneal astigmatism (TCA(Vec)) and total corneal astigmatism measured by ray tracing (TCA(Ray)). METHODS: This study enrolled a total of 204 right eyes of 204 normal subjects. The eyes were measured using a Galilei double Scheimpflug analyzer. The measured parameters included simulated keratometric astigmatism using the keratometric index, anterior corneal astigmatism using the corneal refractive index, posterior corneal astigmatism, and TCA(Ray). TCA(Vec) was derived by vector summation of the astigmatism on the anterior and posterior corneal surfaces. The magnitudes and axes of TCA(Vec) and TCA(Ray) were compared. The Pearson correlation coefficient and Bland–Altman plots were used to assess the relationship and agreement between TCA(Vec) and TCA(Ray), respectively. RESULTS: The mean TCA(Vec) and TCA(Ray) magnitudes were 0.76 ± 0.57 D and 1.00 ± 0.78 D, respectively (P < 0.001). The mean axis orientations were 85.12 ± 30.26° and 89.67 ± 36.76°, respectively (P = 0.02). Strong correlations were found between the TCA(Vec) and TCA(Ray) magnitudes (r = 0.96, P < 0.001). Moderate associations were observed between the TCA(Vec) and TCA(Ray) axes (r = 0.75, P < 0.001). Bland–Altman plots produced the 95% limits of agreement for the TCA(Vec) and TCA(Ray) magnitudes from −0.33 to 0.82 D. The 95% limits of agreement between the TCA(Vec) and TCA(Ray) axes was −43.0 to 52.1°. CONCLUSION: The magnitudes and axes of astigmatisms measured by the vector summation and ray tracing methods cannot be used interchangeably. There was a systematic error between the TCA(Vec) and TCA(Ray) magnitudes.