PURPOSE: To evaluate the accuracy of ultrasound biomicroscopy (UBM) in measuring the distance (in mm) from limbus to the insertion of vertical rectus muscles (superior rectus and inferior rectus compared with the "gold standard" surgical caliper at the time of surgery. METHODS: Prospective, masked, observational study of 31 vertical rectus muscle insertions in which we compared the measurements from the limbus as measured by 50 MHz UBM, either preoperatively or at the time of anesthesia, with that measured by surgical caliper intraoperatively. Measurements (UBM and surgical) were evaluated by two different observers and analyzed using the Bland-Altman method. All UBM measurements were done by the same author. The intraclass correlation coefficient (ICC) and Pearson coefficient with 95% confidence intervals were used to quantify the degree of agreement between the two methods. RESULTS: Thirty-one vertical muscles were evaluated (13 superior rectus and 18 inferior rectus, of which 7 muscles were reoperations). The average for UBM measurements was 6.63 mm and for surgical caliper was 7.09 mm. The measurements for the two methods were all within +/-2 standard deviations of the mean. Only three measurements showed differences more than 1 mm. The ICC was 0.78 and Pearson coefficient was 0.85, indicating a "very good" correlation between the two methods. The longest distance from the limbus that could be accurately measured with the UBM was 12 mm. In one case a pseudotendon was differentiated from the true insertion of a previously recessed superior rectus muscle. CONCLUSIONS: The UBM and surgical measurements showed "very good" correlation when allowing for a margin of error of +/-1.0 mm between the two modalities, indicating that the UBM is a good predictor of the position of the vertical rectus muscles.
PURPOSE: To evaluate the accuracy of ultrasound biomicroscopy (UBM) in measuring the distance (in mm) from limbus to the insertion of vertical rectus muscles (superior rectus and inferior rectus compared with the "gold standard" surgical caliper at the time of surgery. METHODS: Prospective, masked, observational study of 31 vertical rectus muscle insertions in which we compared the measurements from the limbus as measured by 50 MHz UBM, either preoperatively or at the time of anesthesia, with that measured by surgical caliper intraoperatively. Measurements (UBM and surgical) were evaluated by two different observers and analyzed using the Bland-Altman method. All UBM measurements were done by the same author. The intraclass correlation coefficient (ICC) and Pearson coefficient with 95% confidence intervals were used to quantify the degree of agreement between the two methods. RESULTS: Thirty-one vertical muscles were evaluated (13 superior rectus and 18 inferior rectus, of which 7 muscles were reoperations). The average for UBM measurements was 6.63 mm and for surgical caliper was 7.09 mm. The measurements for the two methods were all within +/-2 standard deviations of the mean. Only three measurements showed differences more than 1 mm. The ICC was 0.78 and Pearson coefficient was 0.85, indicating a "very good" correlation between the two methods. The longest distance from the limbus that could be accurately measured with the UBM was 12 mm. In one case a pseudotendon was differentiated from the true insertion of a previously recessed superior rectus muscle. CONCLUSIONS: The UBM and surgical measurements showed "very good" correlation when allowing for a margin of error of +/-1.0 mm between the two modalities, indicating that the UBM is a good predictor of the position of the vertical rectus muscles.