IntroductionIt is customary to measure gait with patients walking barefoot, but certain clinical situations necessitate using shoes. Clinicians would like to objectively measure the effects of heel compression and hind-foot pistoning during gait testing with shoes. The purpose of this study was to provide clinicians with a technique for evaluating the non-rigid behavior of the foot/shoe complex during gait.MethodologyThe single segment foot model used in this study relied on 3D coordinate data from three well separated foot markers, at the heel, MT5 and dorsum (Fig. 1) as well as a lateral maleollus (LA) marker. Two 3D foot coordinate systems (FCS) were defined. The hind-mid foot system (HMFCS) utilized the HF, DF and LF markers, while the ankle-mid foot system (AMFCS) used the LA, DF and LF markers. Also, a 2D ankle-dorsum foot coordinate system (ADFCS) was defined with the LA and DF markers. The shank coordinate system (SCS) was formed by the LA, AS and LK markers. Shank and foot reference planes were defined during a quiet standing test. Additionally, joint centers were defined using three medial markers at the knee (MK), ankle (MA) and forefoot (MF) which were removed prior to the gait tests. The rotations of the FCSs were measured relative to the SCS using Euler angles (Kadaba, et al., 1990) following the sequence of dorsiflexion, internal rotation and inversion. From the quiet standing test, the joint angle offsets for the AMFCS and the ADFCS were calculated relative to the HMFCS and used during the gait tests.Motion data were acquired from nine normal subjects during free cadence walking tests with a six camera Vicon system. The marker coordinate data were low pass filtered with a six Hz digital filter (Winter, et al., 1974). Foot switches were used to define the stride characteristics. Gait tests were performed under two conditions, barefoot and with shoes. Mean ankle motion curves were calculated from two strides of data for each subject. From these individual means, group means were determined for each of the three FCSs.ResultsIn the initial 10% of the barefoot gait cycle the largest difference in ankle plantar flexion angle between the three FCSs was 0.5 degrees (Fig 2). In the corresponding period of foot loading with shoe-clad walking the largest difference in ankle plantar flexion angle between the three FCSs was 1.1 degrees (Fig 3). When the barefoot was going through the dorsiflexion range of motion, the maximum difference in ankle angle between the three FCSs was 1.9 degrees. Walking with shoes demonstrated a maximum difference of 3.9 degrees in ankle dorsiflexion at terminal stance using the three FCSs. The two ankle based FCSs showed reduced dorsiflexion with shoes.DiscussionThe inclusion of the heel marker (HF) makes the hind-mid foot system (HMFCS) more sensitive to heel compression and hind-foot pistoning while walking with shoes (Fig 3). This is evidenced by the greater apparent plantar flexion during loading, when the heel of the shoe is compressed. The HMFCS also measures greater dorsiflexion during single stance when walking with shoes. This is caused by pistoning of the hind-foot within the shoe as the foot lever forces the heel of the shoe away from the hind-foot.The results demonstrate some of the problems encountered when measuring ankle motion of persons wearing shoes. For people wearing prostheses and orthoses which require that shoes be used, the data must therefore be interpreted with care. The use of the FCSs described here provides the clinician with a means to identify what is happening with the foot/shoe complex.