Reproducibility of the radiographic metatarsophalangeal angle in hallux surgery.

Background: Five different methods have been described to define the longitudinal axis of the first metatarsal for radiographic measurements of the first metatarsophalangeal angle. None of these methods has been validated for both preoperative and postoperative assessment of patients undergoing hallux valgus surgery. Previous studies have demonstrated conflicting results regarding the measurement accuracy of these methods.

Methods: To evaluate the measurement accuracy of these five methods, we calculated the intraobserver and interobserver coefficients of repeatability for all five methods with use of twenty preoperative and twenty postoperative standardized plain dorsoplantar weight-bearing radiographs of patients undergoing chevron distal osteotomy.

Results: The preoperative assessment of the metatarsophalangeal angle revealed small differences among the five methods. The intraobserver coefficient of repeatability ranged from 2.10 degrees to 3.34 degrees, and the interobserver coefficient ranged from 2.17 degrees to 3.44 degrees. The postoperative assessment demonstrated substantial differences between methods in which the diaphysis of the first metatarsal is used as a reference (intraobserver coefficient, 5.06 degrees to 7.23 degrees; interobserver coefficient, 5.29 degrees to 8.19 degrees) and methods in which there is one reference point in the metatarsal head and one reference point in the base of the first metatarsal (intraobserver coefficient, 1.88 degrees to 2.67 degrees; interobserver coefficient, 1.86 degrees to 2.34 degrees).

Conclusions: For the assessment of patients undergoing a distal metatarsal osteotomy, we cannot recommend methods in which the metatarsal shaft is used as a reference for the axis of the first metatarsal. Such methods had poor measurement accuracy, especially postoperatively. Methods with reference points distal and proximal to any possible osteotomy site had much better measurement reproducibility. We recommend the method described by Miller in 1974, in which a line is drawn from the center of the first metatarsal head through the center of the base of the first metatarsal, as it was the most precise method and was least biased by postoperative effects.