Repeatability of Weightbearing Computed Tomography Measurement of First Metatarsal Alignment and Rotation.

Abstract

Weightbearing computed tomography (WBCT) can be used to assess alignment and rotation of the first metatarsal. It is unknown whether these measures remain consistent on sequential WBCTs in the same patient when a patient's standing position may be different. The aim of this study was to establish the repeatability (test-retest) of measurements of first metatarsal alignment and rotation in patients without forefoot pathology on WBCT. We retrospectively identified 42 feet in 26 patients with sequential WBCT studies less than 12 months apart. Patients with surgery between scans, previous forefoot surgery or hallux rigidus were excluded. Hallux valgus angle (HVA) and intermetatarsal angle (IMA) were measured using digitally reconstructed radiographs. Two methods of calculating metatarsal rotation (metatarsal pronation angle [MPA] and alpha angle) were measured on standardized coronal CT slices. Interobserver agreement and test-retest repeatability were assessed using intraclass correlation coefficients (ICCs). Standard error of measurement (SEM) and minimally detectable change (MDC95) were calculated. Interobserver agreement was excellent for HVA and IMA (ICC 0.96 and 0.90, respectively) and was good for MPA and alpha angle (ICC 0.81 and 0.80, respectively). There was excellent test-retest repeatability for HVA (ICC=0.90) and good test-retest repeatability for IMA (ICC=0.77). There was excellent test-retest repeatability for MPA (ICC=0.91) and good test-retest repeatability for alpha angle (ICC=0.87). The MDC95 was 4.6 degrees for MPA and 6.1 degrees for alpha angle. Five percent of patients had a difference outside of the MDC95 for the alpha angle, compared with 2% for the MPA. Measurements of first metatarsal alignment and rotation are reliable between assessors and repeatable between sequential WBCTs in patients without forefoot pathology. Subtle differences in patient positioning during image acquisition do not significantly affect measurements, supporting the validity of this method of assessment in longitudinal patient care. Level IV, retrospective case series.