Reliability of Visual Estimation of the First Intermetatarsal Angle

Abstract The size of the radiographic first intermetatarsal (IM) angle is one of the main considerations in selecting a surgical procedure for hallux abducto valgus. Instead of directly measuring the IM angle with a goniometer or a virtual measurement system, physicians will often estimate the size of the IM angle by direct visual estimation. The purpose of the present study was to determine the accuracy of visual estimation of the IM angle size compared with direct measurement. A total of 45 respondents, composed of students, residents, and attending physicians, reviewed a series of radiographs with IM angles varying from 6° to 19°. The attending physician visual estimates differed from the measured values by 3.28° ± 1.56°. The interrater reliability, as determined by the intraclass correlation (ICC), increased with experience by group category: 0.426 (95% confidence interval 0.239 to 0.672) for students, 0.476 (95% confidence interval 0.290 to 0.710) for residents, and 0.656 (95% confidence interval 0.483 to 0.833) for attending physicians. Larger IM angles, defined as >10° were less accurately estimated than were smaller IM angles (3.96° ± 1.60° [n = 24] for larger IM angles and 3.14° ± 0.79° [n = 21] for smaller IM angles; p = .0389). These results suggest caution when visually estimating IM angles, especially larger IM angles. Direct angle measurement might be more appropriate for surgical decision-making.