Intra- and inter-observer reliability of the distal metatarsal articular angle in adult hallux valgus

There is some uncertainty as to whether the distal metatarsal articular angle (DMAA) is a real entity or just radiographic artifact and whether it can be reliably measured. If it is intrinsic to the bone, it should not change with bone position. If it is clinically useful, it should be reproducible. Pre-operative and post-operative radiographs of 32 patients undergoing a proximal bony procedure of the first ray were evaluated independently by three foot and ankle specialists in order to determine the intra and inter-observer reliability of the distal metatarsal articular angle (DMAA). In addition, the hallux valgus angle (HVA), intermetatarsal angle (IMA) and joint congruency/subluxation were determined. We used ANOVA (Scheffe's F-test) to determine reliability of the angular measurements; a p value of less than 0.05 indicates poor reliability and a p value of greater than 0.05 indicates reliability. Intra-observer reliability was good for all angular measurements (HVA, IMA, DMAA pre-op, and DMAA post-op) with p values ranging from 0.33 to 0.95. Inter-observer reliability of the HVA and IMA was good (p=0.63 and p=0.32). Inter-observer reliability of the pre-op DMAA approached statistically poor reliability (p=0.09) and the post-op DMAA reliability was poor (p=0.002). The DMAA reduced after the proximal procedure as measured by all observers, and averaged a reduction of 3.9degrees. Weighted kappa analysis also revealed that there was poor agreement in the determination of congruency and subluxation (Kappa statistic ranged from 0.07 to 0.19). This study suggests that there may be limited value in the DMAA as a clinical measure as it varies with examiner and with the hallux valgus angle.