Ocular biometry through fully refocused steady-state magnetic resonance imaging sequence: reliability and agreement...

To evaluate the reliability and agreement between Fully Refocused Steady-State magnetic resonance sequences (FRSS) and the IOLMaster^® 500 optical biometer for measuring anterior chamber depth (ACD) and axial length (AL).

In a sample of 32 healthy volunteers, separate observers measured the ACD and AL of both eyes using both techniques (inter-method) and through repeated FRSS measurements (interobserver) and by the same observer (intraobserver). We employed the Bland-Altman method to determine the agreement between FRSS and partial coherence interferometry (using the IOLMaster^®) and the interobserver and intraobserver variability, providing the limits of agreement (LoA, or mean difference ± 1.96 SD). Correlation coefficients and intraclass correlation coefficients were also provided.

For ACD measurements with FRSS in pseudo-color scale, we obtained an LoA of 0.016 ± 0.266 mm compared with partial coherence interferometry. For AL with FRSS in greyscale, the LoA was 0.019 ± 0.383 mm. Maximum interobserver variability showed a  - 0.036 ± 0.247 mm LoA for ACD with FRSS in pseudo-color scale. Maximum intraobserver variability was 0.000 ± 0.157 mm LoA for AL with FRSS in greyscale.

ACD and AL measurements using FRSS sequencing present high LoA and reliability when compared with partial coherence interferometry using the IOLMaster^® 500. The results were better for FRSS in pseudo-color scale in ACD determination and for FRSS in greyscale in AL determination. FRSS would not be recommended for IOL power calculation due to variability of AL measurement.