August 2009

How cadavers could change your practice

What do studies of fresh-frozen limbs have to do with you and your patients? Work being done at Iowa State University has significant implications for the management of plantar fasciitis, metatarsal and tibial stress fractures, and PTTD.


Photostogo 18183334

Photostogo 18183334

by Erin D. Ward, DPM

It has been difficult for biomechanists to determine the complex interactions between the bones, ligaments and tendons within the foot due to the invasiveness required to ascertain the information. However, utilizing current motion analysis systems and software, complex three dimensional analyses are within reach.

Several new technologies are allowing researchers to analyze dynamic foot function. However, each of these technologies has its pros and cons. First is the advent of complex dynamic cadaver gait simulators.1,2 Second is the use of 3-D CT/MRI scans to recreate an in vivo foot within a computer model.3-5 Third is the dynamic MRI for developing a finite element model of the in vivo foot.6

The foot is generally assessed in the gait laboratory as a single-segment or two-segment (rearfoot and forefoot) system.7-11 While a single- or two-segmented foot model may be adequate for whole body gait analysis, it is not a realistic representation of the kinematics or kinetics of the in vivo foot. However, there has been some hesitance in creating a multi-segment foot model12-14 for several reasons.

First, it is difficult to find a shoe that allows for multiple marker sets to be placed directly on the foot rather than on the shoe. Second, there exists the potential for significant skin motion artifacts. Third, determining what is a segment and what is not a segment is a challenge. Fourth, labs would need to come to a consensus about landmarks for marker placement so that data from different studies could be compared. And, lastly, the vastness of the data that would be collected from a five- or six-segment model presents its own challenges.

Foot and ankle research in our gait lab is centered around acquiring as much data as possible on fresh-frozen cadaveric specimens using a dynamic cadaver gait simulator developed at Iowa State University (figure 1). The goal of our research is to begin developing a complete understanding of foot function by beginning with intact fresh frozen below the knee amputated cadaveric specimens and end the process with only the bones of each specimen. We begin the process by placing the intact specimen in a Microscribe digitizer, which allows for a three dimensional digital image of the skin, muscles/tendons, ligaments and bones to be created within a computer simulation. From there, we are open to perform the remainder of whatever project for which we have a hypothesis.

Segmental foot motion

Figure 1. Dynamic cadaver gait simulator.

Figure 1. Dynamic cadaver gait simulator.

The initial cadaver study was performed by two separate research groups, one based at Iowa State University and the other at the University of Salford in the U.K., to derive the precise motions of 13 bones of the foot and ankle from just prior to heel strike to just before toe off.15 In order to validate these motions, the University of Salford team transferred the data to a gait lab to conduct an in vivo bone pin study in six subjects.16 The information gained from the project indicated that the bones of no two feet have the same motion, but that there are groups of bones that move correspondingly and can be considered collectively as a segment for in vivo motion analysis.

The cadaveric specimens remained at Iowa State University and two of the specimens were chosen to be dissected, digitized and measured layer by layer to gain a more complete understanding of the structures producing the observed motions. The dissection continued until all that remained were osseous structures, which were scanned in a three-dimensional scanner. The remainder of the data will be disseminated either via the SimTK open access software and/or the I-Fab (International Foot and Ankle Biomechanics) website so that future studies can build on our efforts rather than duplicating them.

The clinical applications of these studies will not be established until skin mounted marker sets are tested in vivo. The interest is to add a more complex foot and ankle model to OpenSim for the entire biomechanics and clinical community to have the opportunity to access the model. Eventually, the goal is to create a model capable of simulating any foot or ankle pathology and predicting the effects of surgical or orthotic correction.

Plantar fascia release
A second study conducted was to determine the functions of the plantar fascia and spring (plantar calcaneonavicular) ligament complex and biomechanical changes within the foot following simulated release or rupture of the plantar fascia.17 In this study, micro force probes from Microstrain were placed within the central and lateral bands of the plantar fascia as well as the superomedial portion of the spring ligament. Each foot was walked with the plantar fascia intact, 1/3 transected and completely released. Kinematics of the calcaneus to tibia were collected for each trial.

The data indicated that as the plantar fascia was sequentially released, medial to lateral, the lateral band force exponentially increased as well as a significant increase in tension on the superomedial component of the spring ligament. Upon complete release of the plantar fascia, the data indicated that the calcaneus was not able to invert in late midstance. Clinically, if the calcaneus does not invert in late midstance, the calcaneus will contact the ground at the subsequent heel strike in an everted position, which further increases the force on the superomedial component of the spring ligament. This increased force on the spring ligament is often observed in patients with posterior tibial tendon dysfunction (PTTD) as plantar navicular pain. The research suggested that no more than 1/3 of the medial component of the plantar fascia should be surgically released. However, it appears that plantar fasciotomies do continue to be performed without any protocol to determine the amount of plantar fascia released.

Orthotic efficacy
A third study performed using the cadaveric dynamic gait simulator was designed to determine the strain placed upon the second metatarsal throughout stance phase of gait with and without a custom and semi-custom Root-type orthotic.18 A rosette strain gauge was fixed to the dorsal aspect of the second metatarsal. Bone pins with three marker sets each were placed in nine bones to determine the kinematics associated with the observed bone strain during walking that is associated with 2nd metatarsal stress fractures. Each foot was walked for several trials without an orthotic, with a semi-custom orthotic and with a custom orthotic. We were blinded as to which orthotic was custom and which orthotic was semi-custom. The study found that the custom orthotic significantly decreased the tension strain, shear strain, compression rate and shear rate. The semi-custom orthotic decreased tension strains and shear strain rates.

Following data collection, a thin cross-section of the second metatarsal at the strain gauge mounting site was removed. A digital photo of the cross-section was taken; cross-sectional area and area moment of inertia about the medial-lateral axis of the second metatarsal were calculated from the scaled thresholded image. The data indicated that second metatarsals with lower cross-sectional area and lower area moment of inertia about the medial-lateral axis demonstrated a more significant response to both the custom and semi-custom orthotics than those with larger cross-sectional area and area moment of inertia about the medial lateral axis.

The study indicated that clinically a runner would significantly benefit from the use of custom Root-type orthoses to aid in decreasing the prevalence of second metatarsal stress fractures. Runners appear to experience 2nd metatarsal stress fracture more frequently than the average population, due to repetitive strain; women are more susceptible to stress fractures due to the smaller cross-sectional area.

Pressure and PTTD
The fourth study used the Tekscan I-scan system to determine real-time stance phase pressures within the subtalar, calcaneocuboid, talonavicular and navicular-medial cuneiform joints; an intact foot was compared to a foot developing simulated PTTD associated with adult acquired flatfoot, without and with a custom Root-type orthotic.19 The adult acquired flatfoot was simulated by transecting the plantar fascia and the entire spring ligament complex while eliminating the force being applied to the posterior tibial tendon. Kinematics of nine bones of the foot were measured along with the pressures within the joints in an attempt to identify correlations as the medial longitudinal arch begins to collapse and the forefoot abducts.

This study is on-going, but the data collected to date indicates that as PTTD progresses, the pressure in the subtalar and calcaneal cuboid joints increases significantly along with subtalar eversion. The results to date have also indicated that custom Root-type orthotics appear to decrease the joint pressures to levels approaching those of the intact state, as well as decreasing the subtalar eversion.

Testing tibial susceptibility
The next study performed was to determine if impact force and tibial acceleration could be used to calculate the strain being produced in the underlying bone at heel strike, as a more practical alternative to a strain gauge. Strain gauges were attached directly to the medial and posterior aspects of the tibia in a cadaveric specimen. An accelerometer was then attached to the adjacent skin. Heel strike was then simulated, varying impact mass and tibial acceleration. Following impact testing, the cross-sectional area was calculated for each tibia. A small cross-section approximating 1.5 cm in thickness that included the strain gage mounting site was removed from the bone, and a digital photo of the cross section was analyzed.

The study determined that the best models for bone strain prediction were peak impact force and tibial cross-sectional area. The impact force and tibial acceleration each accounted for relatively small percentages of variance in bone strain (62% and 38%, respectively). However, when combined with tibial cross-sectional area, the impact force became a strong predictor of internal bone strain (94% of variance). A similar result was found for tibial acceleration after the effects of impact mass had been accounted for (84%).20 Clinically, this suggests that using a CT scan and a skin mounted accelerometer, a practitioner may be able to determining if a runner is will be susceptible to tibial stress fractures.

Future directions
The dynamic gait simulator in our lab has afforded us the capability of determining bone motions throughout the stance phase of gait using bone-pin-mounted marker sets. These data have enabled us to determine which bones may be grouped as a segment and which bones may not be grouped as a segment. We are now in the process of addressing multi-user definable landmarks and placement of these markers for analysis of shod walking. The simulator has, also, allowing us to determine plantar fascial, bone and ligament strains, and determine the effects of orthotic and surgical interventions on foot and ankle pathologies.17-19

The overall goal of this research is to begin combining what we are learning from dynamic gait simulation into the gait lab and improving surgical procedures by adding biomechanical data to surgery. Eventually, the information gained in these studies will become applicable in every clinical evaluation of a patient with a biomechanical pathology.

Erin D. Ward, DPM, practices and performs research at the Central Iowa Foot Clinic in Perry, IA, and collaborates with researchers in the department of kinesiology at Iowa State University.


  1. Hamel AJ, Sharkey NA, Buczek FL, Michelson J. Relative motions of the tibia, talus, and calcaneus during the stance phase of gait: a cadaver study. Gait Posture 2004;20(2):147-153.
  2. Suckel A, Muller O, Langenstein P, et al. Chopart’s joint loading during gait. In vitro study of 10 cadaver specimens in a dynamic model. Gait Posture 2008;27(2):216-222.
  3. Udupa JK, Hirsch BE, Hillstrom HJ, et al. Analysis of in vivo 3-D internal kinematics of the joints of the foot. IEEE Trans Biomed Eng 1998;45(11): 1387-1396.
  4. Camacho DL, Ledoux WR, Rohr ES, et al. A three-dimensional, anatomically detailed foot model: a foundation for a finite element simulation and means of quantifying foot-bone position. J Rehabil Res Dev 2002;39(3):401-410.
  5. Cheung JT, An KN, Zhang M. Consequences of partial and total plantar fascia release: a finite element study. Foot Ankle Int 2006;27(2):125-132.
  6. Blemker SS, Asakawa DS, Gold GE, Delp SL. Image-based musculoskeletal modeling: applications, advances, and future opportunities. J Magn Reson Imaging 2007;25(2):441-451.
  7. Wright DG, Rennels DC. A study of the elastic properties of plantar fascia. J Bone Joint Surg Am 1964;46:482-492.
  8. Veres G. Graphic analysis of forces acting upon a simplified model of the foot. Prosthet Orthot Int 1977;1(3):161-172.
  9. Onyshko S, Winter DA. A mathematical model for the dynamics of human locomotion. J Biomech 1980;13(4):361-368.
  10. Gilchrist LA, Winter DA. A two-part viscoelastic foot model for use in gait simulations. J Biomech 1996;29(6):795-798.
  11. Morlock M, Nigg BM. Theoretical considerations and practical results on the influence of the representation of the foot for the estimation of internal forces with models. Clin Biomech 1991;6(1):3-13.
  12. Leardini A, Benedetti MG, Catani F, et al. An anatomically based protocol for the description of foot segment kinematics during gait. Clin Biomech 1999;14(8):528-536.
  13. Jenkyn TR, Nicol AC. A multi-segment kinematic model of the foot with a novel definition of forefoot motion for use in clinical gait analysis during walking. J Biomech 2007;40(14):3271-3278.
  14. Wolf P, Stacoff A, Liu A, et al. Functional units of the human foot. Gait Posture 2008;28(3):434-441.
  15. Nester CJ, Liu AM, Ward E, et al. In vitro study of foot kinematics using a dynamic walking cadaver model. J Biomech 2007;40(9):1927-1937.
  16. Lundgren P, Nester C, Liu A, et al. Invasive in vivo measurement of rear-, mid- and forefoot motion during walking. Gait Posture 2008;28(1):93-100.
  17. Ward ED, Smith KM, Cocheba JR, et al. In vivo forces in the plantar fascia during the stance phase of gait: sequential release of the plantar fascia. J Am Podiatr Med Assoc 2003;93(6):429-442.
  18. Meardon S, Edwards W, Ward E, Derrick T. Effects of custom and semi-custom foot orthotics on 2nd metatarsal bone strain during dynamic gait simulation. In press, Foot Ankle Int, 2009.
  19. Edwards WB, Ward ED, Derrick TR. Foot joint pressures during dynamic gait simulation, J Foot Ankle Res 2008;1(Suppl 1):O21.
  20. Edwards WB, Ward ED, Meardon SA, Derrick TR. The use of external transducers for estimating bone strain at the distal tibia during impact activity. J Biomech Eng 2009;131(5):051009.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.