Validity of the Pedar Mobile system for vertical force measurement during a seven-hour period

Abstract

Objective measurement of weight bearing during a long-term period can give insight into the postoperative loading of the lower extremity of orthopedic patients to avoid complications. This study investigated the validity of vertical ground reaction force measurements during a long-term period using the Pedar Mobile insole pressure system, by comparing it with a Kistler force platform. In addition, the validity of a new sensor drift correction algorithm to correct for offset drift in the Pedar signal was evaluated. Ground reaction force data were collected during dynamic and static conditions from five healthy subjects every hour for 7 h. A mean offset drift of 14.6% was found after 7 h. After applying the drift correction algorithm the Pedar system showed a high accuracy for the second peak in the ground reaction force-time curve (1.1 to 3.4% difference, p>0.05) and step duration (−2.0 to 4.4% difference, p>0.05). Less accuracy was found for the first peak in the ground reaction force–time curve (5.2 to 12.0% difference; p<0.05 for the first 3 h, p>0.05 for the last 4 h) and, consequently, in the vertical force impulse (5.5 to 11.0% difference, p>0.05). The Pedar Mobile system appeared to be a valid instrument to measure the vertical force during a long-term period when using the drift correction program described in this study.

Introduction

To avoid complications, instruction on partial weight bearing is often given during the rehabilitation of orthopedic patients with various pathologies of the lower extremity (Chow and Cheng, 2000; Endicott et al., 1974; Gapsis et al., 1982; Huiskes, 1998; Perren and Matter, 1996; Phillips et al., 1991; Siebert, 1994; Tveit and Kärrholm, 2001; Weaver, 1975; Wirtz et al., 1998). It is evident that the ground reaction forces under the foot during weight bearing (i.e. when walking and standing) generate forces and moments in other structures in the lower extremity, such as the hip (Bergmann et al., 1993; Davy et al., 1988). In daily clinical practice, because the forces in the hip cannot be directly measured, the ground reaction force under the foot is used as a load measure, often expressed in percentage body weight. Patients are generally instructed to perform partial weight bearing during a period of 6–8 weeks. To evaluate the effectiveness of this instruction and to quantify the loading of the lower limb during the day, objective measurement is needed of the actual amount of loading (vertical ground reaction force) and other aspects of loading (i.e. step duration, vertical force impulse) during weight bearing, both in and outside the clinic, and during a long-term period.

Portable insole pressure devices can measure the actual amount of load bearing during daily activities and over a long-term period (hours) (Hurkmans et al., 2003). However, the validity of vertical force measurements performed by insole systems (especially during long-term periods) may be influenced by temperature or humidity in the shoe, and by loading of the sensors during an entire day (Cavanagh et al., 1992). Moreover, insole sensors measure the “normal” force, which is not necessarily similar to the vertical ground reaction force (Kalpen and Seitz, 1994; Kernozek et al., 1996; McPoil et al., 1995). Only a few studies have used an insole pressure system for long-term measurements (Abu-Faraj et al., 1997; Perren and Matter, 1996; Siebert, 1994; Tveit and Kärrholm, 2001). Perren and Matter concluded that their insole system (based on a hydraulic principle) was not technically reliable enough for routine use in the clinic. Discrete insole pressure systems, developed by Tveit and Kärrholm (2001) and Abu-Faraj et al. (1997), have some disadvantages compared to matrix insole devices: the transducer may act as a foreign body in the shoe, and inaccuracies may occur due to imprecise positioning of the sensors (Abu-Faraj et al., 1997; Cavanagh et al., 1992; Lord, 1981). No reports were found on the validity of these insole systems to measure the vertical ground reaction force during long-term measurements.

Arndt (2003) performed long-term measurements using Pedar pressure matrix insoles (Novel GmbH, Munich, Germany) and found a 17% sensor creep after 3 h. To correct for this creep, Arndt presented a correction method in which short standing trials were used to reset the signal based on the assumption that the measured body weight does not change during the trial. In that study, no data were presented regarding the validity of the Pedar system after the correction method was used. However, for long-term partial weight bearing measurements, we believe that Arndt's correction method is not optimal because the patient uses a walker or crutches meaning that the total body weight cannot be measured. In the present study, we introduce a drift correction algorithm to correct for the possible offset drift during walking in the Pedar mobile system.

The aim of this study was to investigate the validity of the Pedar Mobile system to measure vertical force during a long-term period. The main research questions were: What is the amount and type of drift when using the Pedar system for 7 h? How accurate is the Pedar system in measuring vertical force over a long-term period when corrected for possible offset drift?