Abstract
BACKGROUND: Currently, the metatarsophalangeal angle (hallux valgus angle) is measured based on radiographic images. However, using X-ray examinations for epidemiological or screening purposes would be unethical, especially in children. For this reason it is discussed to measure the hallux valgus angle of the margo medialis pedis (medial border of the foot) documented on foot outline drawings or foot scans. As a first step on the way to prove the validity of those approaches this study assesses the hallux valgus angle measured on the margo medialis pedis based on the same x-ray pictures as the metatarsophalangeal hallux valgus. METHODS: Radiographic images of the foot were obtained from patients with symptomatic hallux valgus malformation. Twelve sets of contact copies of the 63 originals were made, and were marked and measured according to three different methods, each one performed by two observers and with two repeated measurements. Thus, data sets from 756 individual assessments were entered into the multifactorial statistical analysis.Comparisons were made between the angle of the margo medialis pedis and the metatarsophalangeal angle, which was determined by two different methods. To determine the inter- and intraobserver reliability of the different methods, each assessment was conducted by two independent experts and repeated after a period of several weeks. RESULTS: The correlations between the hallux valgus angles determined by the three different methods were all above r=0.89 (p<0.001) and thus highly significant. The values obtained by measuring the margo medialis pedis angle, however, were on average 4.8 degrees smaller than the metatarsophalangeal angles. No significant differences were found between the observers. No systematic deviations for any observer between repeated measurements were detected. CONCLUSIONS: Measurements of the radiographic hallux angle of the margo medialis pedis are reliable and show high correlation with the metatarsophalangeal angle. Because the hallux valgus angles based on margo medialis pedis measurements were slightly but statistically significantly smaller, these measurements should be considered conservative estimates of the metatarsophalangeal angle. Significant differences between hallux valgus angles based on radiographic and non-radiographic material are unlikely. However this question has to be treated in a second stage in detail.