Report of Activities 2002 - Research - Mayo Clinic

Report of Activities 2002 

Biomechanics/Motion Analysis Laboratories 

Division of Orthopedic Research 

Mayo Clinic/Mayo Foundation 

Rochester, MN 55905 USA

April, 2003 

Dear Friends and Colleagues: 

The Orthopedic Biomechanics/Motion Analysis Laboratories are still involved in very diverse research 

activities ranging from human locomotion to cell mechanics. The strong collaboration between 

scientists and clinicians that takes place in our lab make it possible for a broader range of ongoing 

research projects. This report summarizes the activities that have taken place in the calendar year 

2002. We wish to say thank you to our colleagues, collaborators, and supporters for another successful 

year. We continue to welcome visitors from around the world and to train future leaders in biomechanics 

through our fellowship programs. 

On behalf of the staff and fellows of the Orthopedic Biomechanics/Motion Analysis Laboratories, we 

wish you all the very best in your future endeavors. 

Sincerely, 

Kai-Nan An, Ph.D. 

Director, Biomechanics Laboratory 

Kenton R. Kaufman, Ph.D. 

Director, Motion Analysis Laboratory 

KNA/KRK/ge 

Mayo Clinic 

200 First Street SW 

Rochester, Minnesota 55905 

507-284-2511 

Department of Orthopedic Surgery 

Biomechanics Laboratory

Selected Project Summaries 

Table of Contents 

Hand........................................................................................................... 6 

Wrist......................................................................................................... 15 

Elbow........................................................................................................ 18 

Shoulder .................................................................................................. 21 

Hip/Knee .................................................................................................. 25 

Foot/Ankle ............................................................................................... 28 

Gait........................................................................................................... 34 

Rehabilitation .......................................................................................... 37 

Muscle Mechanics ................................................................................... 45 

Tissue Mechanics .................................................................................... 48 

Publications ......................................................................................................... 51 

Staff ......................................................................................................................55 

Acknowledgements ............................................................................................. 58

6 ····························································· HAND ························································ 

Flexor Tendon and Related Research 

Despite numerous advances in suture repair methods and rehabilitative techniques, poor outcomes following 

flexor tendon injury and repair are still common. Our research has focused on the gliding interaction 

between the repaired tendon and the surrounding structures. Our working hypothesis has 

been that by decreasing the gliding resistance at the repair site, tendon excursion will be improved, 

which will further reduce the development of adhesions. In the past we have considered improving 

the gliding surface by using various low friction suture repairs. During 2002 we studied how these 

repairs affected the outcome in an in vivo canine model. We also continued exploring how adhering 

an exogenous substance to the tendon surface may reduce the gliding resistance. 

When is Early Mobilization too Early? (Short 

term validation of the optimal timing for postoperative 

rehabilitation after flexor digitorum 

profundus tendon repair in vivo) 

Project Coordinator: Chunfeng Zhao: 

zhao.chunfeng@mayo.edu 

Ideally, timing considerations for starting 

postoperative rehabilitation after flexor tendon 

repair should include factors such as total digit 

resistance, tendon gliding resistance, and repair 

strength. Based on our previous studies comparing 

digit resistance at 1, 3, 5, and 7 days after 

flexor tendon repair, we found that the digit resistance, 

including tendon gliding resistance, was 

lowest at day 5 after tendon repair without weakening 

the repair site. Is this the appropriate timing 

for starting the mobilization after flexor digitorum 

profundus (FDP) tendon repair? To answer 

this question, the following study in a canine in 

vivo model was performed comparing two starting 

times for beginning mobilization. 

Adult mongrel dogs were divided into two 

groups (n=12 per group); group 10A started postoperative 

therapy (passive flexion/extension digits 

with the wrist in 45º flexion) at day 1 (a clinically 

popular starting date) and group 10B began 

therapy at day 5 (lowest digit resistance from our 

previous study). The surgical digits were fully 

lacerated and repaired with a modified Kessler 

repair. The dogs were sacrificed 10 days after 

surgery. The repaired, sham operated, and contralateral 

control digits were harvested and work of 

flexion and repair strength were measured. 

Eight tendons from four dogs ruptured in 

group 10A (33%) while there were no ruptures in 

group 10B. There was no significant difference in 

total work of flexion (TWOF) or internal work of 

flexion (IWOF) between groups 10A and 10B, 

however TWOF and IWOF in repaired tendons 

were significantly higher than both control and 

sham (p

digits at this stage may exceed the tolerance of 

the repair. Therefore, in this canine model, postoperative 

day 5 may be the best timing to start 

rehabilitation after flexion tendon repair. 

Characteristics of Total and Internal Work of 

Flexion of Human Cadaver Flexor Digitorum 

Profundus Tendons After Different Suture 

Techniques 



The work of flexion fails to partition the work 

between gliding resistance and other sources, 

making it difficult to interpret the source of any 

limitation observed. The purposes of this study 

were to directly compare the total work of flexion 

and the work of flexion within the tendon 

sheath for intact and repaired human FDP tendon. 

Two different tendon repairs, the modified 

Kessler and modified Becker, were studied in 18 

digits from six freshly frozen human cadaver 

hands. The work of flexion and the work of flexion 

within the synovial sheath were tested using 

a digit resistance testing device (Figure 2). 

After tendon repair, the total work of flexion 

increased 11.2% and 26.9% for the modified 

Kessler and modified Becker groups, respectively. 

The difference in increase between the 

two groups was significant (p=0.001) (Figure 3). 

Work of flexion within the synovial sheath in- 

Figure 2: Testing 

device. 1. Actuator, 

2. Proximal transducer, 

3. Specimen, 

4. Distal transducer, 

5. Weight, 6. Frame. 

HAND 7 

creased 126.8% and 308.8% for these two 

groups, respectively (p=0.046). The work of 

flexion within the synovial sheath accounted for 

16.4% of total work of flexion for the intact tendon, 

this percentage was 28.6% and 45.0% for 

modified Kessler and modified Becker groups, 

respectively. 

Figure 3: Change in WOF after laceration and repair. 

Surface Modification of Extrasynovial 

Tendon by Carbodiimide Derivatized (cd-HA) 

gelatin 



Extrasynovial tendons are often used for 

tendon grafts; however, their rough surface leads 

to a higher frictional force which is associated 

with more adhesion than the intrasynovial tendon 

graft. Although friction can be reduced with the 

application of hyaluronic acid (HA) applied to 

the tendon surface, its effect is diminished after 

repetitive motion because of the weak binding 

between HA and the tendon surface. 

Furthermore, the half-life of HA in tissues is 

short. Increasing HA half-life and the binding 

strength on the tendon surface are essential to 

improve the outcomes of the HA application. 

Carbodiimide derivatization, a chemical 

modification of HA, has been recently 

developed. EDC [1-Ethy1-3 (3- 

Dimethylaminopropyl) carbodiimide 

hydrochloride] activates the carboxylic groups 

(COOH) in the HA molecule and forms the 

intermediate O-acylisourea, which can 

chemically bind to exposed amino groups, such 

as those that exist in the collagenous tendon

8 HAND 

matrix. 

In this study 30 canine peroneus longus 

tendons were divided into six different formula 

groups; 1) control, 2) 1% HA + 0.25 ECD (cd- 

HA), 3) 10% gelatin alone, 4) 10% gelatin + 

0.25% EDC (cd-gelatin), and 5) 1% HA + 10% 

gelatin + 0.25% EDC (cd-HA gelatin). The 

frictional force of both treated and non-treated 

tendons were measured. 

After 100 cycles of simulated flexion/ 

extension repetitions, the cd-gelatin group had 

significantly lower gliding resistance than the 

control, cd-HA, and gelatin treated tendons. The 

frictional force of the cd-HA gelatin group was 

significantly lower than all of the other groups. 

(Figure 4). 

After testing, the tendons were examined by 

scanning electron microscopy. The tendon 

surface after being treated with cd-HA gelatin 

was smoother than the non-treated tendon surface 

after 500 cycles of motion. With high 

magnification, the surface of the peroneus longus 

tendon after treatment with cd-HA gelatin was 

similar to the surface of the intrasynovial tendon 

(Figure 5). 

Figure 4: Frictional force with different formulas of modified 

HA at different cycles. C: control, G: 10% gelatin 

alone, HE: 1% HA + 0.25% EDC, GE: 10% gelatin + 0.25 

% EDC, HGE: 1% HA + 10% gelatin + 0.25% EDC. 

Figure 5: A: PL tendon after 500 cycles gliding under the 

pulley (25 X); B: PL tendon treated with cd-HA gelatin 

after 500 cycles motion (25 X). C: PL tendon after 500 

cycles motion (10K X); D: FDP tendon (intrasynovial) 

surface (20K X) 

Tendon Surface Modification by Chemically 

Modified HA Coating After FDP Tendon 

Repair 



Since cd-HA gelatin improved the appearance 

of the extrasynovial tendon surface and decreased 

its gliding resistance, further studies 

were carried out to determine whether this surface 

modification could improve a flexor tendon 

repair. 

FDP tendons (n=36) from canine hindpaws 

were randomly divided into three groups; A) 

Control – no surface modification after tendon 

repair, B) Simple HA – surface of the repaired 

tendon was covered with 1% HA solution, and 

C) cd-HA gel – surface of the repaired tendon 

was covered with a carbodiimide derivatized HA/ 

gelatin. A complete laceration in the FDP tendon 

was made and repaired with a modified Kessler 

technique. The gliding resistance between the 

FDP and the proximal pulley, flexor digitorum 

superficialis (FDS), and bone was measured and 

data were recorded for 500 cycles of simulated 

flexion/extension motion. 

Following treatment, the change in gliding 

resistance over 500 cycles showed similar 

patterns among the three groups. During the first 

10 cycles there were no significant differences in 

gliding resistance among the three testing groups 

(p>0.05). Between 50 and 500 cycles the gliding 

resistance was significantly lower in the cd-HA

gelatin group than in the control group (p0.05). 

Figure 6: Gliding resistance of three groups. 

Expression of IGF-1, TGF-β, and EGF in the 

First Week After FDP Tendon Repair in 

Canine In Vivo 



Although extensive studies have been 

performed looking at the role of growth factors 

during wound healing, little is known about the 

timing or appearance of these factors during 

tendon healing. Early in the wound healing 

process, it has been shown that transforming 

growth factor beta (TGF-β), insulin-like growth 

factor 1 ( IGF-1), and epidermal growth factor 

(EGF) are expressed. In this study we used 

immunofluorescence staining to observe the 

appearance of these three factors in the first week 

after tendon laceration and repair in an in vivo 

canine model. 

Sixteen FDP tendons from eight adult mongrel 

dogs were used in this study. The tendons 

were completely lacerated and repaired with a 

modified Kessler suture. The operated limb was 

immobilized post-operatively. The tendons were 

harvested at 1, 3, 5, or 7 days after the surgery. 

Each group consisted of tendons from two separate 

dogs. Controls from the respective mobile 

digits on the contralateral paws were also studied. 

To minimize the effects of surgery-related 

inflammation, tendons within 7 mm proximal 

and distal to the operation site were discarded. 

The tendons were embedded in paraffin and sec- 

HAND 9 

tioned at 4 micrometers. Confocal microscopy 

was used to note the location of TGF- beta, EGF 

and IGF-1 staining for each tendon. 

Each of the four time points demonstrated 

expression of IGF-1, with no expression of either 

EGF or TGF-β at any time point. Hoechst 

counterstaining demonstrated cellular 

proliferation in the epitenon, especially after day 

5. Staining intensity of IGF-1 was already seen 

on day 1 (Figure 7A) until day 7. The IGF-1 was 

first lightly seen in endotenon and diffuse in the 

epitenon. On day 7 the expression had a 

particular staining distribution and was located in 

the epitenon where it formed a line underneath 

the thicker layer of nuclei. (Figure 7B). Our data 

provide evidence that IGF-1, a potent growth 

factor known to increase the healing potential of 

tenocytes in culture and stimulator for collagen 

synthesis, is the first to be expressed in the 

healing process of canine flexor tendons. The 

characteristic localization suggests that IGF 

expression, which is noted as early as day 1, 

might be one of the signals stimulating epitenon 

cell proliferation, which is noted on day 5. 

Figure 7: A: IGF-1 at day 1. Note nuclei stained with 

Hoechst counterstain (blue) and IGF (green). B: IGF-1 at 

day 7. 

Expression of Growth Factors in Canine 

Tendon at 10 Days After Flexor Tendon 

Laceration and Repair in Vivo. 



In this study, we evaluated the appearances of 

various growth factors at 10 days after flexor tendon 

repair with different timing to start postop-

10 HAND 

Figure 8: Immunolocalization of TGF-β. (a) Positive 

staining of TGF-β was observed around the tendon surface 

and proximal tendon. (b) Around the repair site, 

positive staining of TGF-β was observed in both epitenon 

and endotenon layers. (c) Negative control had no positive 

staining at the same site of b. (d) At the proximal 

site, positive staining of TGF-β was observed around 

newly formed vessels in endotenon cell layer. (e) Negative 

control staining at the same site with d. 

Figure 9: Immunolocalization of PDGF isoforms 

around the repair site. (a) Positive staining of PDGF-AA 

was detected in both epitenon and endotenon cell layers. 

(b) Positive staining of PDGF-BB was detected only in 

endotenon cell layer. (c) Negative control had no positive 

staining at the same sites. 

Figure 10: Immunolocalization of VEGF. (a) Positive 

staining of VEGF was detected in whole area of tendon, 

equivalently. (b) At the proximal site, positive staining 

of VEGF was observed around the newly formed vessels 

in endotenon and epitenon cell layers. (c) Negative control 

had no positive staining at the same site with b. 

erative therapy in dog model in vivo. The transforming 

growth factor beta (TGF-β), epidermal 

growth factor (EGF), platelet-derived growth factor 

(PDGF), insulin like growth factor (IGF), fibroblast 

growth factor (FGF), and vascular endothelial 

growth factor (VEGF) were studied using 

immunohistochemical staining analysis. 

Eight repaired FDP tendons from eight adult 

mongrel dogs were used. The tendons were completely 

lacerated and repaired with a modified 

Kessler suture. Four tendons were obtained from 

each of the two therapy groups, i.e. 10A 

(commence therapy at day 1) and 10B (commence 

therapy at day 5). The tendons were sectioned into 

10 µm cryosections. 

Immunohistochemical staining was performed 

for TGF-β, EGF, PDGF-AA, PDGF-BB, IGF, 

VEGF, and bFGF. Light microscopy was used to 

note the location of every growth factor staining 

for each tendon. Routine hematoxylin and eosin 

staining was also conducted on the same specimens 

and observed in the same way. 

Our analysis demonstrated the positive staining 

of all growth factors. Especially, the appearance of 

TGF-β, PDGF-AA, PDGF-BB and VEGF were 

well detected (Figures 8-10). These findings provide 

evidence that TGF-β, EGF, PDGF-AA, 

PDGF-BB, IGF, bFGF, and VEGF are all present 

at 10 days after surgery. Considering healing process, 

synthesis of extracellular matrix and angiogenesis 

were the center part of flexor tendon healing 

at postoperative day 10 after inflammation 

phase. These data suggested that the stimulation of 

well described growth factors, TGF-β, PDGF, and 

VEGF are important factors for tendon healing. 

A Histological and Immunohistochemical Study 

of the Subsynovial Connective Tissue in 

Idiopathic Carpal Tunnel Syndrome 



The etiology of idiopathic carpal tunnel syndrome 

(CTS) is poorly understood. The most common 

histological finding in CTS is noninflammatory 

synovial fibrosis. While repetitive, 

forceful hand or wrist motion is believed to be a 

significant etiologic factor, the relationship be-

tween the repetitive movement and the tenosynovial 

thickening is unknown. Although the anatomical 

structure of the carpal tunnel has been 

well studied, the microscopic organization of the 

flexor tenosynovium and subsynovial connective 

tissue has not been examined in detail. 

We performed a retrospective chart review of 

30 patients with idiopathic carpal tunnel 

syndrome who had synovial specimens sent to 

our Laboratory Medicine and Pathology 

Department during carpal tunnel release between 

January 1999 and December 2001. We compared 

these specimens to a control group of 10 fresh 

frozen cadavers who did not have an antemortem 

diagnosis of CTS and who met the same 

exclusion criteria. 

Analysis included hematoxylin and eosin 

(HE) sections for histology analysis and 

immunohistochemistry for the distribution of 

types I, II, III, and VI collagen and TGF-β RI, 

RII, and RIII. 

Histological examination showed a marked 

increase in fibroblast density, collagen fiber size, 

and vascular proliferation in the patients 

compared to the control specimens (Figure 11). 

Collagen types I and II were not found in the 

synovium of either patients or controls, but type 

VI collagen was a major component of both. 

Collagen type III fibers were more abundant in 

the patients than in the controls. 

Figure 11: Mean fibroblast density in SSCT, patients 

(n=30) vs control (n=10). 

Expression of TGF-β RI was found in the 

endothelial cells and fibroblasts in the patient and 

control specimens, with a marked increase in 

expression in the fibroblasts of the patients 

compared to the control tissue. TGF-β RII was 

HAND 11 

found in endothelial cells and endovascular 

smooth muscle in the patient and control 

specimens. There was minimal staining for TGFbeta 

RIII in endothelial cells in both the patient 

and control specimens. 

These findings are similar to those found after 

soft tissue injury and suggest that patients with 

idiopathic (CTS) may have sustained an injury to 

the subsynovial connective tissue (SSCT). 

Effect of Pronosupination Position of the 

Forearm on Activities and TendonEexcursion 

of the APL and EPB 

Study Coordinator: Denny Padgett: 

padgett.denny@mayo.edu 

In 1895, de Quervain described pain along the 

radial styloid region related to the tendons 

coursing in the first dorsal compartment. The 

first dorsal compartment is occupied by the abductor 

pollicis longus (APL) and extensor pollicis 

brevis (EPB) tendons (Figure 12), either or 

both which may be responsible for de Quervain’s 

tenosynovitis. Etiology of the tendinitis known as 

de Quervain disease is thought to be an overuse 

disorder resulting from such things as postpartum 

childcare, trauma, pregnancy associated with 

hormonal changes, metabolic abnormality, and 

congenital synostosis between the scaphoid and 

trapezium. The most frequently cited cause of 

Figure 12: The first dorsal compartment is occupied by 

the abductor pollicis longus (APL) and extensor pollicis 

brevis (EPB).

12 HAND 

the tendinitis is overuse of the hand and the wrist 

joint. 

The first purpose of this study was to examine 

differences in EMG signals detected from the 

APL and EPB muscles due to differences of the 

forearm in order to find the kinesiological 

changes that might produce de Quervain disease. 

We hope to find a difference between APL and 

EPB muscle activities since de Quervain disease 

may be secondary to EPB entrapment as distinct 

from APL tendinitis. The second purpose of this 

study was to determine whether the EPB and 

APL tendons have different degrees of tendon 

excursions due to rotational movement of the 

forearm, wrist, and thumb. 

Seven fresh frozen cadavers were dissected 

to investigate the APL and EPB tendon 

excursions as a function of various hand and 

wrist joint rotations. The APL tendon excursion 

ranged from 0 mm with respect to thumb MP 

flexion/extension to 12.6 mm with respect to 

forearm rotation. The EPB tendon excursion had 

a more limited range of 2.2 mm with respect to 

forearm rotation to 6 mm with respect to wrist 

radial/ulnar deviation. As an adjunct to this investigation, 

the effects of wrist position on the 

gliding resistance of the EPB tendon at the first 

dorsal compartment was initiated. Here 14 fresh 

frozen cadavers were first used to establish the 

range of angles between the EPB tendon and the 

retinaculum ligament. These specimens were 

then mounted in the testing device allowing complete 

characterization of the gliding resistance as 

a function of seven different EPB tendon angles. 

These tests were performed with and without the 

retinacular septum. Results showed the EPB tendon 

angle was smallest, as was the gliding resistance 

with the wrist at neutral position. Removing 

the retinacular septum tended to further reduce 

gliding resistance in all wrist positions. 

Finally the effect of forearm and wrist position 

on the EMG activities of the APL and EPB 

muscles were investigated in six healthy adults. 

Data were collected during isolated maximum 

voluntary contraction of the APL and EPB then 

during a functional pinch task in several wrist 

and forearm positions. Results showed that the 

APL and EPB activities were affected by forearm 

rotation significantly and that the APL muscle 

activity with the wrist palmar-flexed at 60º was 

higher than that at any other position. 

Development, Validation and Clinical 

Application of a Quantitative Method for 

Thumb Trapeziometacarpal Joint 

Project Coordinator: Christine Hughes: 

hughes.christine@mayo.edu 

The purpose of this study was to develop and 

validate a method for assessing the maximal 

workspace of the thumb’s trapeziometacarpal 

(TMC) joint motion in vivo using an electromagnetic 

tracking device. 

Due to the intricate anatomical structures at 

the base of the thumb, it is difficult to clinically 

measure the motion of the TMC joint. Movement 

patterns of the thumb make the current methods 

of measuring thumb joint motion difficult and are 

not for objective documentation of thumb motion 

or thumb impairment. Currently, measurement 

of the range of thumb motion is limited to flexion-extension 

and abduction-adduction planes at 

the metacarpal (MCP) and interphalangeal (IP) 

joints. However, if such motions take place simultaneously 

or are combined with a more complex 

joint such as occurs at the TMC joint, it becomes 

difficult, if not impossible, to evaluate the 

precise kinematics of the thumb. 

Three experiments have been designed for 

this study (2 implemented and 1 in progress). In 

the first study, a model for obtaining and calculating 

the maximal workspace of the TMC joint 

has been developed. The path of the head of the 

first metacarpal bone was measured and the surface 

area of the workspace was determined 

Figure 13: The predictive curve of the extreme circumduction 

of the TMC joint to form the maximal working 

space of the first metacarpal head motion (a), predictive 

paths of abd/adduction and (b) flexion/extension of the 

TMC joint.

(Figure 13). In addition, we investigated the accuracy 

and reliability of the model using eight 

cadaver hands (Figure 14). Flexion/extension, 

abduction/adduction, opposition, and circumduction 

movements of the TMC joint cadaveric 

hands were measured with the electromagnetic 

tracking system. The first sensor was transfixed 

to the first metacarpal bony segment and the second 

was mounted on the skin of the head of the 

first MCP. The third sensor was transfixed into 

the trapezium and scaphoid. The fourth sensor 

was attached on the skin of the dorsal aspect of 

the head of the third metacarpal. In the second 

part of the study, we developed a maximal work- 

Figure 14: Schematic of in-vitro setup. The hand was securely 

transfixed with a plastic rod stabilized to a plastic 

frame. A constant load of 100g was applied to each of the 

four extrinsic tendons to provide a static load across the 

TMC joint. 

Figure 15: In-vivo setup for the quantitative measurement 

of the trapeziometacarpal joint 

HAND 13 

space from 20 normal subjects (Figure 15). This 

database is intended to be the basis for a subsequent 

study in which we will compute the percentage 

of actual use compared to the “ideal” 

value for the given thumb metacarpal length and 

determine the loss of motion associated with arthritis 

of the thumb and surgical corrective procedures. 

The third part of this study will be the 

clinical application. Motion of the TMC joint for 

patients with osteoarthritis will be recorded preoperatively 

and then compared with postarthroplasty 

data. 

The results from the in-vitro study indicated 

that the spherical model, as measured by the 

electromagnetic tracking system, was a reliable 

tool for assessing the movement of the TMC 

joint. The intraclass correlation coefficient (ICC) 

values for the reliability estimation of the repeated 

measurements of the radius and surface 

area of all specimens were 0.91 and 0.98, respectively. 

The mean coefficient of variance (CV) of 

the measured radius and the surface area were 

2.04% and 3.65%, respectively. 

The in-vivo collection of 20 subjects, 10 with 

repeated measurements, indicated the quantitative 

model as measured by the electromagnetic 

tracking device to be a reliable tool for assessing 

the range of motion of the TMC joint. The ICC 

values of the repeated trials of all subjects within 

the first test day, the repeated trials of 10 subjects 

within the second test day and the repeated measurements 

of the 10 subjects between 2 different 

test days was 0.99, 0.99 and 0.97. The 95% confidence 

interval of the reliability coefficient was 

0.92-0.99. The mean CV of all of the measured 

maximal workspace of the TMC joint was 

3.42%. The collection of data from patients with 

osteoarthritis of the TMC joint will be measured 

pre- and post- operatively within the next year.

14 HAND 

Publications: 

An, KN; Zobitz, ME; Zhao, CF; Amadio, PC: 

Gliding characteristics of tendon. Fourth World 

Congress of Biomehcanics 2002. 

Erhard, L; Schultz, FM; Zobitz, ME; Zhao, CF; 

Amadio, PC; An, KN: Reproducible volar partial 

lacerations in flexor tendons: A new device 

for biomechanical studies. J Biomech 35:999- 

1002, 2002. 

Erhard, L; Zobitz, ME; Zhao, CF; Amadio, PC; 

An, KN: Treatment of partial lacerations in 

flexor tendons by trimming. J Bone Joint Surg 

84-A(6):1006-1012, 2002. 

Momose, T; Amadio, PC; Zobitz, ME; Zhao, CF; 

An, KN: Effect of paratenon and repetitive motion 

on the gliding resistance of tendon of extrasynovial 

origin. Clin Anat. 15:199-205, 2002. 

Paillard, PJ; Amadio, PC; Zhao, CF; Zobitz, ME; 

An, KN: Gliding resistance after FDP and FDS 

tendon repair in zone II. An in vitro study. Acta 

Orthop Scand 73(4):465-470, 2002. 


An, KN: Pulley plasty versus resection of one 

slip of the flexor digitorum superficialis after repair 

of both flexor tendons in Zone II. A biomechanical 

study. J Bone Joint Surg . 84-A 

(11):2039-45, 2002. 


An, KN: Comparison of strategies to improve 

results after laceration of both flexor tendons in 

zone II: A biomechanical study. Annual Meeting 

of the Orthopaedic Research Society 2002. 

Zhao, CF; Amadio, PC; Berglund, LJ; Zobitz, 

ME; An, KN: A new testing device for measuring 

gliding resistance and work of flexion in a 

digit. Annual Meeting of the Orthopaedic Research 

Society 2002. 

Zhao, CF; Amadio, PC; Momose, T; Couvreur, 

P; Zobitz, ME; An, KN: Effect of synergistic 

wrist motion on adhesion formation after repair 

of partial flexor digitorum profundus tendon lacerations 

in a canine model in vivo. J Bone Joint 

Surg 84-A(1):78-84, 2002. 

Zhao, CF; Amadio, PC; Momose, T; Zobitz, ME; 

Couvreur, P; An, KN: Remodeling of the gliding 

surface after flexor tendon repair in a canine 

model in vivo. J Orthop Res 20:857-862, 2002. 

Zhao, CF; Amadio, PC; Paillard, PJ; Tanaka, T; 

Zobitz, ME; An, KN: Digit resistance within 

short term after FDP tendon repair in canine in 

vivo. Annual Conference of the American Society 

of Biomechanics 2002. 

Zhao, CF; Amadio, PC; Zobitz, ME; An, KN: 

Resection of the flexor digitorum superficialis 

reduces gliding resistance after zone II flexor 

digitorum profundus repair in vitro. J Hand Surg 

27(2):316-321, 2002. 

Zhao, CF; Amadio, PC; Zobitz, ME; An, KN: 

Gliding characteristics after flexor tendon repair 

in canine in vivo. Annual Meeting of the Orthopaedic 

Research Society 2002. 

Zhao, CF; Amadio, PC; Zobitz, ME; Momose, T; 

Couvreur, P; An, KN: Effect of synergistic motion 

on flexor digitorum profundus tendon excursion. 

Clin Orthop 396:223-230, 2002.

································································ WRIST ······················································ 15 

Analysis of Symmetry of Displacement of the 

Distal Radioulnar Joint in Normal Subjects 

Project Coordinator: Mark Zobitz: 

zobitz.mark@mayo.edu 

Evaluating a patient with suspected instability 

of the painful distal radioulnar joint (DRUJ) 

remains a difficult problem. Historically, the 

evaluation of unstable DRUJ has been done by 

physical examination and radiographic means. 

However, there has been no standard quantitative 

method of measuring instability. Cross sectional 

CT imaging has been used to evaluate instability 

of bilateral DRUJs providing a comparison of the 

normal side and injured side. Since instability of 

the DRUJ is a function of displacement of the 

radius relative to the ulna, we developed a new 

method which allows a quantitative in-vivo 

measurement of forearm torque strength and dynamic 

CT imaging comparing the behavior of the 

right and left sides. This study looks at the use of 

this method on the normal population to validate 

its use in patients. 

Ten bilateral wrists (6 right-hand dominant 

and 4 left-hand dominant) with no history of disease 

or trauma to either upper extremity were 

subjected for preliminary studies. The data from 

both right and left hand dominants were analyzed 

as 2 different groups. A device was developed to 

allow standard CT imaging simultaneously of 

bilateral DRUJs in preset positions with isometric 

muscle loading. The protocol produces an image 

of both DRUJs under conditions of no load, 

resisted pronation and resisted supination in neutral 

forearm rotation, 60° pronation, and 60° 

supination. A quantitative value called the Instability 

Index was used to define displacement of 

the distal radius relative to the ulnar head. Additionally, 

a special testing apparatus which incorporated 

a torque cell was used to actively measure 

generated peak torque in the same conditions 

as in the CT protocol. 

The weakest torque strength was found in resisted 

pronation in the pronated position, and resisted 

supination in the supinated position. The 

results of calculating the Instability Index revealed 

differences that showed trends of displacement 

related to testing condition, whereby 

the values were least variable in 60 o of forearm 

supination. 

These data are being compiled as a normative 

data base and will be used to compare with values 

derived from similarly tested patients with 

suspected unilateral instability of the distal radioulnar 

joint. Further analysis will be carried out 

to define the underlying cause of asymmetrical 

values, such as hand dominance and anatomic 

asymmetry. 

Biomechanical Effects of Kienbock’s Disease 

Project Coordinator: Mark Zobitz 


Joint leveling procedures such as the radial 

wedge and radius shortening osteotomy are commonly 

applied techniques in the treatment of 

Kienböck’s disease. In patients with Kienböck’s 

disease, the lunate has to endure excessive pressure, 

typically by abnormal joint contact across 

its proximal surface. By performing a radial osteotomy, 

the geometry of the radiolunate contact 

is altered, thereby theoretically reducing the load 

transmitted across the joint. There have been 

three radial wedge osteotomies proposed for this 

purpose: lateral opening wedge osteotomy, lateral 

closing wedge osteotomy, and medial closing 

osteotomy. These procedures all have theoretical 

advantages for load redistribution. However, 

it must be remembered that the distal radius 

not only forms the foundation for the wrist, but is 

also a part of the distal radioulnar joint. Any alteration 

in the alignment of the distal radius has a 

predilection for altering the pronation-supination 

characteristics of the forearm. The purpose of 

this study was to evaluate the mechanical effects 

on the forearm after performing a distal radius 

osteotomy. 

Fourteen human cadaveric forearms were dissected. 

Static loading and dynamic resistive 

loading conditions were simulated throughout a 

complete pronation-supination motion on a forearm 

simulator. Torque required to produce full 

rotation of the forearm was measured. Surgical 

models simulated radius shortening (STG), lateral 

opening (LOP), lateral closing (LCS), and 

medial closing wedge (MCS) osteotomies. We 

compared torque values at 80% and 100% of the 

full range of motion, in both supination, and pro-

16 WRIST 

nation. 

With static loading, LCS required more torque 

to achieve 80% and 100% pronation compared to 

control. Eighty percent and 100% of supination 

also required more torque, but were not statistically 

significant. LOP, STG, and MCS did not 

have a significant effect on torque (Figure 16). 

The resistive loading conditions showed the same 

tendencies as the unloaded conditions. 

The results suggest that if all osteotomies are 

equal in their efficacy to decompress the lunate, 

one would choose the osteotomy that has the 

least effect on otherwise normal forearm mechanics. 

We found the lateral closing to be the 

least favorable with respect to the functioning of 

the forearm. This appears to be independent of 

the anatomical configuration of the joint. 

Figure 16: Torque required to achieve a full range of 

motion with resistive muscle loading for each surgical 

condition. All specimen groups are combined (n=14). 

The Clinical Role of Wrist Joint 

Mechanoreceptors 

Project Coordinator: Diana K. Hansen: 

hansen.diana@mayo.edu 

Wrist pain is the most common joint-related 

disability in the upper extremity. As with other 

joints, the precise mechanism for pain reception 

and transmission remains poorly understood. 

Recently, branches of the anterior (AIN) and posterior 

interosseous nerves (PIN), found in close 

anatomical proximity to each other, were found 

to be pure wrist joint capsular afferent nerves. 

Sectioning of the AIN and PIN at this location 

has been 80% successful clinically in resolving 

intractable wrist pain. Anesthetic injection at this 

common location of the AIN and PIN can be 

used to predict the effects of surgical intervention. 

The normal function of these afferent 

branches of the AIN and PIN, and the effect of 

their resection on joint mechanics, remains unknown. 

The specific purpose of the current study is to 

determine the effect of diminished or absent joint 

capsular afferent input on wrist proprioception. 

Many investigators have studied the influence of 

capsuloligamentous mechanoreceptors on joint 

proprioception. However, none of these investigators 

were able to isolate the effect of joint afferent 

input from other systems that contribute to 

proprioception (e.g., muscle spindle, Golgi tendon 

apparatus or cutaneous mechanoreceptors). 

Therefore, the relative contribution of the capsuloligamentous 

structure of the joint could not effectively 

be delineated. In the present study, a 

local anesthetic in the region of the AIN and PIN 

is used to block joint mechanoreceptor afferent 

signals without affecting other proprioceptive 

systems. The hypothesis is that there will be a 

quantifiable change in acuity of static wrist position 

sense by blocking the anterior and posterior 

interosseous nerves. 

The testing is performed by seating the subject 

in a chair with both forearms (proximal to 

the wrist) supported on padded troughs. The subject 

grasps a vertical post mounted on an air 

bearing to minimize resistance to wrist flexionextension 

(Figure 17). An opaque shield is 

placed between the subject’s face and hands to 

obstruct visual observation of wrist position. 

Headphones convey white noise to reduce auditory 

stimuli from the testing devices. Displacement 

data from the grasp apparatus is detected 

with a three-dimensional electromagnetic tracking 

system. 

Ongoing tests involve both passive and active 

wrist motion within a randomized double-blind 

prospective research design. Subjects are tested

at seven target locations within the typical wrist 

range of motion. The active-active condition 

simulates typical functional use of the wrist, 

while the passive-passive condition effectively 

isolates joint sensory input by minimizing input 

associated with muscle activation. 

Eighty volunteers will complete both 

active-active and passive-passive conditions prior 

to, and after, receiving an injection of placebo 

(normal saline) or local anesthetic (1% 

Lidocaine) in the region of the AIN and PIN 

proximal to the wrist. Comparisons will be made 

between proprioception accuracy associated with 

each testing condition and between pre- and postinjection 

data. 

Figure 17: Placement of hand in the testing apparatus. 

The forearm is stabilized in a neutral position. The electromagnetic 

sensor (circled) detects wrist motion, looking 

specifically at flexion/extension. 

Publications: 

Park, MJ; Cooney, WP III; Hahn, ME; Looi, KP; 

An, KN: The effects of dorsally angulated distal 

radius fractures on carpal kinematics. J Hand 

Surg 27(2):223-232, 2002. 

Sauerbier, M; Hahn, ME; Fujita, M; Neale, PG; 

Berglund, LJ; Berger, RA: Analysis of dynamic 

distal radioulnar convergence after ulnar head 

resection and endoprosthesis implantation. J 

Hand Surg 27A(3):425-434, 2002. 

Adams BD, Berger RA: An Anatomic Reconstruction 

of the Distal Radioulnar Ligaments for 

Posttraumatic Distal Radioulnar Joint Instability. 

J Hand Surg 27A:243-251, 2002. 

WRIST 17 

Haugstvedt JR, Berger RA, Berglund LJ, Sabick 

MB: An Analysis of the Constraint Properties of 

the Distal Radioulnar Ligament Attachments to 

the Ulna. J Hand Surg 27A(1):61-67, 2002 

Sauerbier M, Fujita M, Hahn ME, Neale PG, 

Berglund LJ, An KN, Berger RA: The Dynamic 

Radioulnar Convergence of the Darrach Procedure 

and the Ulnar Head Hemiresection Interposition 

Arthroplasty: A Biomechanical Study. J 

Hand Surg 27B(4):307-316, 2002

18 ·························································· ELBOW ······················································ 

Effect of Length on Kinematics of Radial 

Head 



Comminuted radial head fractures often occur 

in association with medial collateral ligament 

injuries. Positioning of a radial head prosthesis 

relies only on the judgment of the surgeon to reconstruct 

the exact length of the radial column. 

Nothing is known about the effects of slight 

lengthening or shortening of the radius on joint 

kinematics or force transmission through the radiocapitellar 

joint. The goal of this study was to 

evaluate both the ulnohumeral kinematics and the 

force transmission through the radiocapitellar 

articulation after lengthening and shortening the 

radial neck in the medial collateral ligament deficient 

elbow. 

Six fresh frozen cadaveric upper extremities 

were used. The radial neck was cut and a custommade 

fixture was cemented to both the radial 

neck and the radial head in the original orientation, 

permitting lengthening and shortening of 

the radial neck with the insertion or removal of 

acrylic spacers. Three-dimensional spatial orientations 

of the ulna and humerus were measured 

using an electromagnetic tracking device. A motor 

applied to the biceps and brachialis pulled the 

forearm from extension to flexion at a controlled 

rate, with the elbow subjected to gravity valgus 

and varus stresses sequentially and the forearm 

fixed in neutral rotation. Data were collected using 

Motion Monitor software (Innovative Sports 

Training Inc., Chicago, IL) and analyzed in terms 

of ulnar axial rotation and varus/valgus laxity at 

discrete positions through the flexion arc. Force 

at the radiocapitellar joint following lengthening 

was measured with an I-scan sensor (Tekscan 

Inc., Boston, MA) inserted into the joint. The 

force was statically recorded at 5°, 30°, 60°, and 

90° of flexion in a loaded condition with the 

forearm in neutral, pronation and supination, and 

the humerus in a gravity valgus position. 

Compared to the native length condition, total 

varus/valgus laxity increased consistently with 

shortening, and decreased with lengthening of 

the radial neck (Figure 18). Shortening of the 

radial neck caused internal rotation of the ulna, 

whereas lengthening of the radial neck consistently 

caused an external rotation of the ulna 

throughout the flexion arc. The axial force transmission 

across the radiocapitellar joint decreased 

with the flexion angle. In the normal elbow with 

native length, the force is highest in the forearm 

pronated position and lowest in the supinated position. 

With 2.5mm lengthening, the axial contact 

force increased significantly, in all forearm 

positions. 

The results of this study clearly show the importance 

of reproducing the original length of the 

radial neck when implanting a radial head prosthesis. 

Both shortening and lengthening of the 

radial neck, simulating under and overstuffing of 

the joint, caused consistent changes in the kinematics 

of the elbow. Although lengthening increased 

stability of the elbow, this also was coupled 

with increased forces in the radiocapitellar 

joint, which can lead to early degenerative 

changes. The results emphasize the need for intraoperative 

measurement and alignment jigs that 

can be used to ensure the optimal placement of 

the radial head prosthesis. 

Figure 18: Varus valgus laxity with radial neck lengthening 

and shortening.

An in Vitro Biomechanical Study of a Hinged 

External Fixator Applied to an Unstable 

Elbow 

Project Coordinator: Mark Zobitz: 

zobitz.mark@Mayo.edu 

Treatment of an unstable elbow remains difficult 

as long-term immobilization leads to joint 

contracture but early postoperative motion exercise 

frequently results in failure of fracture fixation 

and/or rupture of the repaired soft tissue. Recently, 

through development of the hinged external 

fixator, successful results with immediate 

postoperative exercise without complications 

were reported for unstable elbow. While the 

hinged external fixator could prevent hazardous 

translation of the elbow, it can also maintain normal 

joint motion. The purpose of this study was 

to measure the bending stiffness of the elbow attached 

with a hinged external fixator along with 

determining how the axial force to the joint affects 

mechanical properties of the system. A 

hinged external fixator (Dynamic Joint Distractor-2) 

was attached to the lateral side of seven 

cadaveric elbows. Cantilever lateral bending tests 

were performed at three flexion angles in varus 

and valgus directions. Varied states of joint contact 

and axial loading were studied. Stiffness 

decreased concomitant with increased elbow 

flexion. Introduction of a gap reduced the stiffness 

of the system while increased axial loading 

made the system stiffer, especially in valgus 

loading. Stiffness in varus was approximately 

four times that in valgus. From our study we 

concluded that lateral fixator application with 

half pins is most effective for protecting against 

varus producing forces. When using the external 

fixator for unstable-elbow patients, credence 

should be given to joint condition, and this 

should also be noted for elbow position during 

rehabilitation. 

Snapping Triceps 

Project Coordinator: Paul Kane 

kane.paul@mayo.edu 

Dislocation (snapping) of the medial head of 

the triceps is usually seen in association with ulnar 

nerve dislocation but is often misdiagnosed, 

ELBOW 19 

as only the ulnar nerve dislocation is recognized 

in the majority of cases. Transposition of the ulnar 

nerve alone in symptomatic elbow snapping 

does not result in resolution of symptoms if coexisting 

dislocation of the medial head of the triceps 

is not recognized and treated as well. 

We hypothesize that patients with snapping 

triceps have a pattern of muscle firing during elbow 

motion that differs from that of the normal 

population. It is our impression that this abnormal 

muscle firing pattern leads to hypertrophy of 

the medial head of the triceps. 

The information from this study will be 

utilized to determine if there are abnormal patterns 

of muscle firing that can be recognized and 

used to treat snapping triceps. In turn, muscle 

retraining techniques may then be used to establish 

normal muscle firing patterns, restore muscle 

balance, and prevent future muscle dislocation 

and snapping. Using custom computerized isometric 

strength assessment equipment, the EMG 

activity of the major elbow flexor and extensor 

muscles will be assessed. These seven muscles 

will be tested in resisted isometric elbow flexion 

and resisted isometric extension and flexion at 

five different positions (Figure 19). 

A uniaxial accelerometer placed on the 

medial side of the elbow is used as an indicator 

to aid in determining when snapping occurred 

under test conditions. EMG data will be collected 

simultaneously with accelerometer data to mark 

if, and when, muscle firing activity changes. 

XL 

Figure 19: Experimental set-up A Medial view. Note 

that accelerometer (XL) is fixed over the medial epicondyle. 

B Lateral view: three of seven EMG amplifiers 

are shown fixed to the upper arm of test participant.

20 ELBOW 

Publications: 

Inagaki, K; O’Driscoll, SW; Neale, PG; Uchiyama, 

E; Morrey, BF; An, KN: Importance of a 

radial head component in Sorbie unlinked total 

elbow arthroplasty. Clin Orthop 400:123-131, 

2002. 

Kamineni, S; An, KN; Neale, P; Hirahara, H; Pomianowski, 

S; O'Driscoll, S; Morrey, BF: Partial 

posteromedial olecranon resection - An athlete's 

friend and foe. Annual Meeting of the Orthopaedic 

Research Society 2002.

···························································· SHOULDER ·················································21 

Electromyographic Activity in the Immobilized 

Shoulder Girdle Musculature During 

Contralateral Upper Limb Movements 

Project Coordinator: Denny Padgett, PT: 


Shoulder immobilization is commonly used to 

protect healing tissues after shoulder injury or 

surgery. During this period, individuals typically 

receive specific instructions regarding restricted 

motion of the treated shoulder girdle, but are allowed 

to move the contralateral shoulder girdle 

and upper limb without restriction. Some individuals 

re-injure or aggravate symptoms despite 

immobilization, whereas others suffer the adverse 

effects of immobility. Clinically, describing 

the behavior of the immobilized shoulder girdle 

musculature during well-arm activities can 

facilitate the development of more precise precaution 

guidelines and potentially provide an 

avenue for early muscle activation during periods 

of shoulder immobilization. 

The primary purpose of this study was to 

quantify the EMG activity in the immobilized 

shoulder musculature during quiet resting and 

also during a battery of motions completed with 

the contralateral, non-immobilized upper limb in 

six asymptomatic male volunteers (ages 22-33 

years) recruited from the author's institution. 

Three categories of motions were studied: (1) 

Five basic daily activity motions– straight forward 

reach, cross-body reach, overhead reach, 

downward reach, and backward pull; (2) One 

bimanual daily activity motion- Spaghetti Jar 

Opening; and (3) Four incrementally (5, 15, and 

25 lbs.) resisted daily activity motions- straight 

forward reach, overhead reach, and backward 

pull, and suitcase lift. These motions are commonly 

utilized during daily life and are generally 

unrestricted during periods of immobilization 

when performed with the non-immobilized upper 

limb. It was hypothesized that relatively low levels 

of peak EMG activity (10-20% MVC) such as 

those occurring in normal gait and Phase I of the 

Neer Shoulder Rehabilitation Programs would be 

observed in the immobilized shoulder girdle 

musculature. 

The main outcome measure for this study was 

the mean peak normalized EMG activity. This 

activity was recorded with fine wire 

(supraspinatus, infraspinatus) and surface 

(deltoids, trapezii, biceps, serratus anterior) electrodes 

from 10 immobilized shoulder girdle muscles 

at rest and during the slow, fast, and incrementally 

resisted (5, 15, and 25 lbs.) contralateral 

upper limb motions (Figure 20). The results of 

this study demonstrated that EMG activity in all 

muscles was low during quiet immobilized standing 

(< 1.5% MVC). During slow contralateral 

upper limb motions, EMG activity ranged from 

0.7% to 51.6% MVC (highest in trapezii) and 

was < 15% MVC in the supraspinatus, infraspinatus, 

and anterior deltoid. Bimanual jar opening 

increased biceps activity from 7.8% to 16.1% 

MVC. During fast contralateral upper limb motions, 

peak infraspinatus activity increased from 

6.4% during a slow backward pull to 56.7% during 

a fast straightforward reach (Figure 21). Supraspinatus 

activity was relatively high during all 

resisted backward pulling motions (25.2% to 

32.1% MVC) (Figure 22), whereas resisted forward 

reaching produced relatively little activity 

in the anterior deltoid, supraspinatus, infraspinatus, 

or biceps. Several slow and fast motions produced 

high trapezius activity (> 45% MVC) with 

low rotator cuff, biceps, and anterior deltoid activities 

(< 10% MVC). 

In conclusion, immobilized shoulder girdle 

muscle EMG activity during quiet standing is 

negligible in asymptomatic individuals. Contralateral 

upper limb motions at self-selected speeds 

produce minimal EMG activity in the immobi- 

Figure 20: Subject crouches to lift weighted hand bag. 

EMG data is collected from immobilized shoulder girdle.

22 SHOULDER 

Figure 21: Mean (+ SEM) peak one second normalized 

EMG activity during 5 daily activity motions performed at 

fast speeds. UT = upper trapezius; LT = lower trapezius; 

MT = middle trapezius; AD = anterior deltoid; MD = middle 

deltoid; PD = posterior deltoid; SS = supraspinatus; IS 

= infraspinatus; SA = serratus anterior; BP = biceps. 

Figure 22: Mean (+ SEM) peak one second normalized 

EMG activity during resisted backward pull. UT = upper 

trapezius; LT = lower trapezius; MT = middle trapezius; 

AD = anterior deltoid; MD = middle deltoid; PD = posterior 

deltoid; SS = supraspinatus; IS = infraspinatus; SA = 

serratus anterior; BP = biceps. 

lized shoulder and are therefore not likely to be 

harmful to healing tissues. During early healing 

periods, patients with biceps-labral injury should 

minimize bimanual activities, those with supraspinatus 

injury should avoid backward pulling 

motions, and those with infraspinatus injury 

should avoid fast straightforward reaches. Cross 

body reaches, straightforward reaches, and 

downward reaches at either slow or fast speeds 

preferentially activate some scapular stabilizer 

muscles while minimally activating the rotator 

cuff, biceps, or anterior deltoid muscles. Therefore, 

these motions may be appropriately prescribed 

as rehabilitative exercises that can be initiated 

while the shoulder remains immobilized. 

Effect of Scapular Protraction and Retraction 

on Isometric Shoulder Elevation Strength 

Project Coordinator: Denny Padgett: 


Because of increased emphasis on assessing 

and rehabilitating scapulothoracic dysfunctions 

of patients with shoulder pain, this study 

was performed to initiate investigation into the 

relationship of scapular position to shoulder muscle 

function. The objective of this study was to 

examine the effect of scapular protraction (SP) 

and scapular retraction (SR) on isometric shoulder 

elevation strength measured in the sagittal 

plane. It was hypothesized that strength would be 

significantly reduced when tested in the SP position 

relative to the neutral resting scapular position 

(SN). 

The design of this study was a prospective 

before-after trial of 10 healthy volunteers (5 

male, 5 female) ages 26-43 years. Each subject 

completed three maximal isometric shoulder elevation 

contractions at 90º of sagittal plane elevation 

in the positions of SN, SP, and SR (Figure 

23). The order of scapular positions was varied 

to minimize fatigue effects. The main outcome 

measures were isometric shoulder elevation 

strength for the three scapular positions. 

The results of this study (Figure 24; Table 

1) showed that isometric strength was significantly 

lower for the SP position compared to the 

SN position (8.5 + 3.4 kg vs. 11.1 + 4.0 kg, p 

< .0005), and for the SR position relative to the 

SN position (7.8 + 3.3 kg vs. 11.1 + 4.0 kg, p 

< .00003). Strength values did not differ between 

the SP and SR positions (p = 0.38). 

Movement of the scapula into a protracted 

or retracted position results in a statistically 

significant reduction in isometric shoulder 

elevation strength as measured in this study. 

Further research is warranted to examine the relationship 

between scapular position and shoulder 

muscle function.

Figure 23: Subject is in test position with scapula in 

neutral; load cell is attached to strap, which is proximal 

to wrist. 

Figure 24: Mean and SD force values by scapular position, 

gender, and group. SN values were significantly 

higher than SP and SR values. 

Table 1: Probability Values 

Upper Extremity Kinematics: Normal Three- 

Dimensional Motion 

Project Coordinator: Denny Padgett 


The use of 3-dimensional upper extremity 

(UE) motion analysis has lagged behind lower 

extremity (LE) analysis, primarily due to a lack 

SHOULDER 23 

of standardization in methodology among investigators 

and due in part to the complex kinematics 

of the shoulder joint. Most clinicians want a 

simple objective tool to identify what compensations 

that are being used to place the hand in a 

functional position. Therefore, the overall purpose 

of this project was to determine 

the feasibility and utility in developing a normal 

kinematic database for head, trunk, shoulder, elbow, 

and wrist motions necessary for the performance 

of cardinal plane motions and basic 

motions essential for activities of daily living 

(ADLs). 

A convenience sample of 5 subjects (3 males, 

2 females) was recruited for this study. Mean 

body mass index (BMI) was 25.0 with a range of 

21.7-33.9. Participants had no history of pathology 

of the UE, neck, or trunk. 

Kinematic data were collected while participants 

performed 3 repetitions, at a self-selected 

speed, of cardinal plane shoulder motions: flexion, 

extension, abduction, internal rotation, and 

external rotation. Data were also collected while 

each subject performed 3 additional repetitions of 

motions commonly involved with ADLs: hand to 

mouth, hand to top of head, hand to back of neck, 

hand to back pocket, and hand to opposite shoulder. 

Start and end points of each repetition began 

and ended with the UE resting at the side of the 

participant. 

In general, the algorithm used to describe UE 

motion was accurate to within 2.5% FS. Between 

trials, this method was shown to be repeatable 

within 99%. Preliminary results of this project 

demonstrated uniformity of movement in 

each anatomical plane with each motion studied. 

Kinematic data describing the motion of 

the shoulder as the hand is placed on the top of 

the head is shown in figures 25-27. Similar to 

gait analysis, UE motion is normalized to percent 

of cycle with the average motion curve and standard 

deviation band representing variability 

among subjects. Other joint kinematic data, including 

trunk and neck motion help to describe 

how the UE moves to place the hand on the head. 

According to O'Neill et al, the hand is the 

main effector of the UE. Since hand position is 

dependent on the movement of the UE, most cli-

24 SHOULDER 

nicians are simply interested in how the wrist, 

elbow and shoulder act to place the hand in a 

functional position. Other authors have pointed 

out that the diagnosis and treatment of orthopedic 

and neurological disorders of the UE can benefit 

from 3-dimensional motion analysis and that the 

management of movement disorders depends 

largely on the ability to objectively quantify 

changes in performance. It is believed that this 

study will help develop an objective tool, which 

can provide valuable information about the effectiveness 

of clinical treatment programs and rehabilitation 

planning. Three-dimensional UE kinematic 

motion analysis and the continued formula- 

Figure 25: Sagittal plane motion of the shoulder when 

placing the hand to the top of head. 

Figure 26: Frontal plane motion of the shoulder when 

placing the hand to the top of head. 

Figure 27: Transverse plane motion of the shoulder 

when placing the hand to the top of head. 

tion of a normal database describing UE motions 

for basic cardinal plane and for activities of daily 

living will become a vital tool that clinicians can 

use to compare pathological movements to normal. 

Publications: 

Halder, AM; O’Driscoll, SW; Heers, G; Mura, 

N; Zobitz, ME; An, KN; Kreush-Brinker, R: 

Biomechanical comparison of effects of supraspinatus 

tendon detachments, tendon defects, 

and muscle retractions. J Bone Joint Surg 84(A) 

5:780-785, 2002. 

Itoi, E; Minagawa, H; Wakabayashi, I; Kobayashi, 

M; An, KN: Biomechanics of multidirectional 

instability of the shoulder. MB Orthop 15 

(5):11-16, 2002. 

Lee, SB; An, KN: Dynamic glenohumeral stability 

provided by three heads of the deltoid muscle. 

Clin Orthop 400:40-47, 2002. 

Luo ZP; Hsu HC; An, KN: An in vitro study of 

glenohumeral performance after suprascapular 

nerve entrapment. Med Sci Sports Exerc 34 

(4):581-586, 2002. 

Mura, N; An, KN; O'Driscoll, SW; Heers, G; 

Jenkyn, TR; Siaw-Meng, C: Biomechanical Effect 

of Infraspinatus Disruption and the Patch 

Graft Technique for Large Rotator Cuff Tears. 

Orthopaedic Research Society, 2002. 

Padgett, DJ; Kaufman, KR; Morrow, DA; Fuchs, 

B; Sim, FH: Oncologic Shoulder Arthrodesis: A 

Functional Assessment. Gait and Clinical Motion 

Analysis Society 2002. 

Smith, J; Kotajarvi, BR; Padgett, DJ; Eischen, JJ: 

Effect of Scapular Protraction and Retraction on 

Isometric Shoulder Elevation Strength. Arch 

Phys Med Rehab. 83(3): 2002. 

Zobitz, ME; An, KN: Stress Analysis of the Rotator 

Cuff: Model Development and Validation. 

American Society of Biomechanics 2002.

····························································· HIP/KNEE ··············································· 25 

Wire Fixation Technique for Tibial Tubercle 

Osteotomy. Mechanical Comparison of 3 

Wires vs. 2 Wires. 

Project coordinator: Kenton R. Kaufman 

kaufman.kenton@mayo.edu 

In the difficult primary total knee arthroplasty 

and revision surgery, exposure with standard 

techniques can be problematic especially in the 

stiff knee. Transection of the quadriceps tendon 

above the patella can aid exposure, but can also 

compromise quadriceps function. Tibial tubercle 

osteotomy has many advantages and results in 

excellent exposure. Using this technique, the risk 

of patellar tendon avulsion should be minimized. 

Optimal fixation of the osteotomized fragments 

remains controversial. There are studies showing 

that two screws provide the strongest fixation. 

But when using an intramedullary stem the 

placement around it may be difficult and potentially 

weaken the bone. Many surgeons use cerclage 

wires for fixation of the tibial tubercle osteotomy. 

This study will compare the static fixation 

strength of 3 vs. 2 cerclage wires to stabilize 

tubercle osteotomy. 

The information derived from this study will 

be a guide for surgeons in fixation techniques to 

minimize the chance of fractures, and will allow 

the determination of rehabilitation protocols after 

repaired osteotomies. 

Testing is underway with a total of twelve 

fresh frozen human knees planned. The knees are 

randomly assigned to the experimental groups 

such that one knee will receive 2 cerclage wires 

and the other will receive 3 wires. An 8-cm long, 

2-cm wide and 1-cm depth tibial tubercle osteotomy 

is created in all specimens. All cerclages 

are created with a tension of 20 pounds and then 

twisted and crimped. A custom-made jig has 

been designed to mount the specimens in a material-testing 

machine (Figure 28). Traction of 

the quadriceps tendon is applied with a servohydraulic 

actuator. Superior and anterior motion of 

the tibial fragment is measured with an optoelectronic 

camera system. Load, displacement and 

peak force at which the osteotomy fails measured. 

From the curve of force vs. displacement 

we obtain stiffness, maximum load, and energy 

absorbed by the devices. Failure is to be defined 

as total displacement greater than 1 cm. 

Figure 28: Testing configuration. 

Femoroacetabular Impingement in Different 

Stages of Slipped Capital Femoral Epiphysis. 

Project Coordinator: Kenton R. Kaufman 

kaufman.kenton@mayo.edu 

Slipped capital femoral epiphysis (SCFE) is a 

hip disorder of adolescent age characterized by a 

sudden or gradual anterior displacement of the 

metaphysis relative to the epiphysis. Patients 

with SCFE typically have a decrease in hip internal 

rotation and flexion due to an impingement 

on the anterior rim of the acetabulum. This is 

caused by the discrepancy between the prominent 

femoral metaphysis and the acetabulum and 

could be an important factor in the development 

of early osteoarthrosis. As the femoral neck impinges 

on the acetabular rim, lateral fraying, 

shear or impaction injuries of the anterior 

acetabular cartilage can be seen with eventual hip 

incongruity leading to secondary arthrosis. 

Some studies had shown that mild and moderate 

slips have an excellent long-term prognosis 

whereas severe slips are associated with early 

degenerative disease progression. Our goal is to 

find the effect of the displacement of the capital 

femoral epiphysis on hip range of motion. These

26 HIP/KNEE 

results will tell us about the reduction of motion 

for each degree of slip. 

This work is underway a hip foam bone 

model and a cadaver model with the proximal 

epiphysis displacement in 3 stages: mild, moderate, 

and severe (Figure 29). We will make an osteotomy 

to reproduce the SCFE, on the basis of 

anatomical and radiological measurements. The 

anatomic position of the physis will be at 65º 

from the longitudinal axis of the femur and 12º 

of retroversion. Based on typical radiographs and 

the criteria of Southwick the posterior displacement 

will be 25º, 50º, and 75º. To these displacements 

we will add an internal rotation and 

valgus displacement of the physis. We will take 

x-rays of each condition in order to compare and 

demonstrate a true displacement. Each model 

will be fitted and tested in a motorized protractor 

in order to record the maximal arc of movement 

achieved will then be recorded. 

We will compare the range of motion and areas 

charts between the different epiphyseal displacements 

and the normal femur. 

Figure 29: Models of slipped epiphysis geometries. The 

three stages degrees of displacement correspond to mild, 

moderate and severe slip. 

Logic Controlled Electromechanical Free 

Knee Orthosis 

Project Coordinator: Steven Irby: 

irby.steven@mayo.edu 

The goal of this project is to design, develop, 

and test an electronically controlled knee joint 

that can be installed on a conventional KAFO. 

This dynamic knee brace system is comprised of 

a novel wrap spring clutch and electric motor 

drive, sensors at the foot and knee, electronic 

control circuitry, and a rechargeable battery pack 

(Figure 30). The knee joint will unlock during 

the swing phase of gait and lock during the 

stance phase of gait. The clinical result of this 

project is expected to improve efficiency of gait 

in patients with poliomyelitis, spinal cord injuries, 

myopathic disorders, congenital spinal defects, 

and acquired paralysis due to infections or 

vascular insults. Four centers distributed across 

the continental United States are participating in 

this clinical trial. Research participants are tested 

over a six-month period using laboratory based 

gait analysis and metabolic energy measurements. 

Subjective user data is being collected 

utilizing a standardized questionnaire. Also, mechanical 

fatigue testing is being conducted to 

evaluate the reliability of knee joint components. 

At this time field and laboratory testing are 

underway with six subjects. Recruitment is ongoing. 

Initial kinematic and kinetic data are 

promising. Preliminary data demonstrate a reduction 

in VO2 when the dynamic knee brace 

system is used (Figure 31). In addition to clinical 

tests conducted in 2002, mechanical testing 

and analysis has been expanded. Four knee joint 

components have undergone fatigue testing to a 

maximum of 1 million cycles. As part of this 

mechanical testing we have been able to evaluate 

a proprietary surface coating which favorably 

increased the friction coefficient by 14% between 

the working surfaces. This has the potential to 

increase the load capacity of the current design 

or, more importantly, provide a means to significantly 

decrease the axial dimension of future designs. 

Secondly, three-point bending tests about 

the varus-valgus axis have been conducted to 

evaluate the stiffness and failure mode of the current 

wrap spring clutch design. Also, finite ele-

ment analysis has been performed to guide future 

designs. Based upon this analysis the clutch design 

has been revised distributing stress concentrations 

and reducing assembly weight with no 

reduction in capacity (Figure 32). Prototypes 

based upon this new design are ready for fatigue 

and bending tests. 

Discussions are underway with industrial entities 

that work at both the component and complete 

system levels. We hope to identify a partner 

with whom we can collaborate on further developing 

this dynamic knee brace system into a 

commercially viable product. 

Figure 30: Bilateral 3-D motion analysis is conducted at 

self-selected walking speeds over level ground. 

Publications: 

Babis GC; Trousdale, RT; Jenkyn TR; Kaufman, 

KR: Comparison of two methods of screw fixation 

in periacetabular osteotomy. Clin Orthop. 

403:221-227, 2002. 

Babis, G.C., An, K.N., Chao, E.Y.S., Rand, J.A., 

Sim, F.H.: Double level osteotomy of the knee: 

A method to retain joint-line obliquity. J Bone 

Joint Surg 84-A(8):1380-1388, 2002. 

Fuchs, B; O’Connor, MI; Kaufman, KR; Padgett, 

DJ; Sim, FH: Iliofemoral arthrodesis and pseudarthrosis: 

a long-term functional outcome 

evaluation. Clin Orthop. 397:29-35, 2002. 

Knee 

HIP/KNEE 27 

Figure 31: Representative data of VO2 data measured 

while walking on a level treadmill. Three test conditions 

were 1) knee brace ‘Locked’ in full extension simulating a 

standard KAFO, 2) Dynamic Knee Brace System ‘Active’ 

releasing the knee during swing, and 3) ‘No Brace’. Approximately 

50% of participants are able to walk for some 

distance without a knee brace. Note that the ‘Active’ data 

show reduced VO2 levels with respect to the standard 

KAFO ‘Locked’ but is still greater than ‘No Brace’ condition. 

Figure 32: Finite element analysis was performed to 

guide wrap spring clutch design modifications. Mediallateral 

bending moments result in stresses which exceed 

the endurance limit of the base material in the original 

design. The modified design addressed these areas of 

high stress while reducing assembly weight by approximately 

5%.

28 ···················································· FOOT/ANKLE ················································ 

Subtalar Implant for Posterior Tibial Tendon 

Dysfunction (PTTD) and Flatfoot 

Project coordinator: Lawrence Berglund 

berglund.lawrence@mayo.edu 

Subtalar implants or arthroereisis are advocated 

for flatfeet in children and more recently in 

adults to correct flatfoot deformity due to PTTD 

while maintaining joint mobility. Subtalar arthroereisis 

were performed with appropriately 

sized implants, modeled after the subtalar MBA 

System (KMI Corp.) placed into the lateral aspect 

of sinus tarsi in a cadaver model (Figure 33). 

The procedure was tested using a dynamic foot 

simulator utilizing our flatfoot model. 

Arthroereisis improved alignment of the arch 

visually, however, significant differences were 

observed at the metatarsal-tibial (forefoot relative 

to the leg) and calcaneal-tibial (hindfoot relative 

to the leg) levels. Metatarsal-tibial and calcanealtibial 

dorsiflexion was not different between intact, 

flatfoot, and implant conditions. The operation 

was successful in improving the visual appearance 

of the flatfoot, while preserving mobility 

of the hindfoot. More critical analysis 

showed that normal function was not achieved 

and over-correction of tarsal bone alignment was 

observed. 

Figure 33: Schematic drawing and typical lateral radiograph 

of a patient who underwent arthroeseisis. 

Lateral Column Lengthening (LCL) for 

Flatfoot Associated with PTTD 

Project coordinator: Lawrence Berglund: 


Lateral column lengthening with a calcaneocuboid 

fusion (LCL) (Figure 34) is being 

performed for adult patients with PTTD. LCL is 

believed to effectively correct the deformity, 

while minimizing loss of motion in the hindfoot 

and midfoot. Utilizing the foot simulator and 

the flatfoot model, LCL was tested by performing 

it on the foot specimens by an open wedge 

osteotomy of the anterior process of the calcaneus, 

inserting a 1 cm wedge shaped from a 

synthetic bone. LCL was found to improve foot 

alignment when viewed clinically, and foot and 

ankle movement was maintained. However, 

LCL corrected only forefoot alignment, hindfoot 

deformity remained. 

Figure 34: Lateral column lengthening with a calcaneocuboid 

fusion (From Hansen ST: Functional reconstruction 

of the foot and ankle, Philadelphia, 2000, 

Lippincott Williams & Wilkins, p319.) 

Effect of Flexor Digitorum Longus Tendon 

(FDL) Transfer and Medial Displacement 

Calcaneal Osteotomy (MDO) for PTTD and 

Flatfoot 



A surgical procedure utilized to restore function 

to the flatfoot is the FDL tendon transfer and 

MDO, performed as either an isolated or combined 

operation. FDL transfer has the potential 

advantage of preserving motion of the hindfoot. 

However, it has limitations in consistently correcting 

deformity. MDO is believed to correct 

for the hindfoot valgus deformity, while minimizing 

loss of joint motion. The purpose of this 

study was to evaluate the effect of these procedures 

on PTTD and flatfoot utilizing the footsimulator. 

MDO was performed with a trans-

verse osteotomy made at a right angle to the lateral 

border of the calcaneus at a 45º angle to the 

plane of the foot. FDL transfer was performed 

by the tendon being passed through a bone tunnel 

in the navicular and sutured in place. Results 

indicated that MDO improved foot alignment 

when observed visually, and demonstrated that 

foot and ankle movement is maintained and 

alignment in the coronal plane was corrected. 

However, the amount of correction was insufficient 

in the transverse plane. FDL tendon transfer 

had no effect in improving alignment when 

combined with MDO. 

Effect of Ankle Braces on PTTD and Flatfoot 



In addition to the above three surgical procedures 

describe to correct for PTTD and flatfoot, 

conservative treatment utilizing two types of ankle 

orthotics were evaluated. These were the 

stirrup type- Aircast standard ankle brace 

(Aircast Inc, NJ) and an articulated type at the 

ankle- the Active Ankle (Active Ankle System 

Inc., KY). 

Kinematic measurements were consistent 

among the ten cadaveric specimens tested and in 

multiple trials of the same specimen. Maximum 

metatarsal-tibial dorsiflexion was not significantly 

different for the normal, flatfoot, or either 

brace. Maximum metatarsal-tibial external rotation 

was significantly greater in flatfoot than normal 

and both braces, and greater in both braces 

than normal. Maximum metatarsal-tibial eversion 

was greater in flatfoot and both braces than normal, 

and there was no significant difference between 

flatfoot and both brace conditions. Maximum 

calcaneal-tibial dorsiflexion was not significantly 

different for the normal, flatfoot, or 

either brace. Maximum calcaneal-tibial external 

rotation was significantly greater in flatfoot than 

normal and both braces, and there was no significant 

difference between normal and both brace 

conditions. This same relationship was true for 

metatarsal-tibial eversion. 

FOOT/ANKLE 29 

Foot and Ankle Biomechanics with Orthosis 

Ambulation 

Project Coordinator: Diana Hansen 

hansen.diana@mayo.edu 

This project deals with the issue of restoring 

lower limb function in patients with severe 

arthritis of the ankle or hindfoot who require the 

use of a standard ankle foot orthosis (AFO) or 

hindfoot orthosis (articulated, non-articulated) 

and to lock the ankle or hindfoot for stability 

during gait. These patients have stiff, arthritic 

joints which are painful and impair their ability 

to walk. Currently, these types of patients are 

fitted with a conventional, rigid polypropylene 

AFO to limit movement. These devices restrict 

ankle, hindfoot, and mid foot movement during 

walking. This may produce ineffecient gait, 

unnecessarily immobilize unaffected joint, and 

may not be tolerated by debilitated patients 

because of their weight or bilateral impairments. 

The goal of this project is to test two lightweight 

hindfoot orthoses (articulated and nonarticulated) 

and a standard AFO in arthritic 

subjects in three different walking environments. 

In level, side-slope, and ramp walking, we tested 

the performance of these three orthoses in 

patients with ankle and subtalar arthritis. The 

results of these studies are expected to improve 

the efficacy of gait in patients with arthritis who 

require conventional AFO treatment. In addition 

to patients with arthritis, these orthoses will be 

useful for patients with ankle instability, subtalar 

instability, and neuromuscular disorders. 

Twenty normal subjects were evaluated, ten 

females and ten males. Subjects ambulated in 

level, 10º ramp up and down, 10º side-slope with 

foot on high side (side-slope high) as well as 

with foot on low side (side-slope low). Subjects 

were tested in unbraced condition with shoe, 

AFO, HFO-articulated (HFO-A) and HFO-rigid 

(HFO-R) braces which were custom made for 

each subject. Holes were cut in each orthosis and 

shoe to allow placement of markers on the skin 

(Figure 35). Kinematic measurements were 

obtained with a ten-camera Motion Analysis 

system with eleven reflective markers applied to 

the skin in a specific foot marker set. Three 

dimensional calcaneal-tibial (cal-tib) and

30 FOOT/ANKLE 

Figure 35: Placement of the 11 reflective markers for 

tracking motion of the tibia, hindfoot (calcaneus), and 

forefoot segments. 

Figure 36: Hindfoot motion: significant differences between 

normals and patients with ankle osteoarthritis in the 

A) sagittal, B) coronal, and C) transverse planes. 

metatarsal-calcaneal (met-cal) motion was 

calculated. Kinetic data was collected from 

forceplates (Kistler Instruments Corp., Amherst, 

NY) for each condition and was timesynchronized 

with the motion data. 

Twelve patients with unilateral ankle 

osteoarthritis were evaluated, five females and 

seven males, excluding patients who had 

previous ankle/hindfoot reconstruction, systemic 

rheumatic diseases, hindfoot arthritis, lower 

extremity joint replacement, or other disorders 

affecting gait. Data was collected using the same 

protocol as used with the normal subjects. 

Patients with ankle arthritis had significantly 

reduced hindfoot plantarflexion, total sagittal 

plane motion, total coronal plane motion, and 

total transverse plane motion, as well as 

significantly reduced forefoot internal rotation 

and total transverse plane motion (Figure 36). 

Effect of Foot Orthoses on Posterior Tibialis, 

Tibialis Anterior, and Peroneus Longus 

Muscle Electromyographic Activity During 

Walking and Running 

Project Coordinator: Brian Kotajarvi: 

kotajarvi.brian@mayo.edu 

This prospective study investigated the effect 

of two commonly used foot orthoses on the EMG 

activity of the posterior tibialis (PT), tibialis 

anterior (TA) and peroneus longus (PL) muscles 

during walking and running. This study will be 

the first to examine the effect of foot orthoses on 

posterior tibialis EMG activity and examine both 

orthosis-inducted range of motion changes and 

lower limb EMG activity simultaneously. 

Acquiring these data is essential to determine the 

relationship between foot and ankle motion 

changes and muscle activity. Studying the PT 

muscle is important given its importance in 

medial longitudinal arch support, and the 

popularity of orthosis use to treat the various 

stages of PTTD. If the results of this study 

support the hypothesis that orthoses reduce PT 

and PL EMG activity, these data will provide a 

biomechanical rationale for orthosis prescription 

in the treatment of PTTD, and provide a 

foundation for further investigation of orthoses in 

subjects with symptomatic PTTD.

A total of 5 people with significant pes planus 

(compensated forefoot varus) were fitted with 

both prefabricated (also called an “off-the-shelf”) 

and semirigid custom molded orthoses by a 

single, experienced physical therapist. EMG 

signals were collected from both the PT and PL 

muscles using a kinesiological EMG system. 

Activity was sampled by a combination of fine 

wire indwelling (PT and PL) and surface (TA) 

electrodes during walking and running with and 

without the orthotic devices. Ankle kinematic 

data was collected using a Penny and Giles twin 

axis electrogoniometer (Biometrics LTD, 

Ladysmith, VA) in conjunction with an 

instrumented treadmill (Gaitway Treadmill, 

Kistler Instrument Corp. Amherst, NY). This 

treadmill is designed to measure vertical ground 

reaction forces. The associated software 

calculates center of pressure, temporal gait 

parameters, and accommodates the addition of 

EMG and electrogoniometer inputs described 

above. Kinematic and electroymographic data is 

time synchronized. 

Preliminary data for this project has been 

successfully collected and is shown for a 

representative subject in figures 37-41. The 

graphs represent ground reaction force (pounds) 

and EMG activity (microvolts) from selected 

walking strides. The PT EMG activity is shown 

under the conditions of no orthosis, custommolded 

orthosis, and off-shelf orthoses. 

Although formal data analysis has not been 

completed, visual inspection clearly shows later 

onset of PT EMG activity with both orthosis 

conditions. The tibialis anterior and peroneus 

longus EMG activity showed no obvious change. 

Figure 37: No orthosis-Posterior Tibialis EMG activity 

FOOT/ANKLE 31 

Figure 38: Custom orthosis- Posterior Tibialis EMG 

activity 

Figure 39: Off the shelf orthosis-Posterior Tibialis 

Figure 40: Sagittal plane ankle motion 

Figure 41: Frontal plane ankle motion

32 FOOT/ANKLE 

Ankle Stability: Comparison of Normal and 

Implanted Ankle 

Project Coordinator: Lawrence Berglund: 


Clinical results of early total ankle arthroplasty 

designs (TAA) were disappointing. More 

recent designs of ankle prostheses have reports of 

good mid-term results. There is limited information 

regarding effects of TAA on ankle stability. 

Our hypothesis is that joint stability of unconstrained 

TAA designs are significantly less than 

that of the normal intact ankle. 

Eight cadaveric lower extremities were studied 

utilizing a testing device (Figure 42) which 

enabled measurement of multiaxis force and displacement. 

Tests consisted of anterior-posterior 

and medial-lateral translation and externalinternal 

rotation with axial load equivalent to 

body weight (700N) and minimal (5N) axial 

load. Test sequences were performed for neutral, 

dorsi-flexed and plantar-flexed ankle positions. 

Stability was determined for the intact normal 

ankle, and following insertion of an unconstrained 

TAA (S.T.A.R.). Load-displacement 

curves were analyzed for each test. Statistical 

analysis was performed with paired t-test or repeated 

measures ANOVA with significance at 

p

Publications: 

Crevoisier, XM; Kitaoka, HB; Hansen, D; Kaufman, 

KR: Effects of ankle-foot articulated and 

non-articulated hindfoot orthoses on the foot/ 

ankle kinematics during walking in unlevel 

gound conditions. Orthopaedic Research Society 


Crevoisier, XM; Kitaoka, HB; Hansen, D; Morrow, 

DA; Kaufman, KR: Gait abnormalities associated 

with ankle osteoarthritis. Orthopaedic Research 


Jenkyn, TR: Motion of the ankle complex and 

forefoot twist during walking and medial direction 

changes. Gait and Clinical Movement 


Kotajarvi, BR; Crevoisier, XM; Hansen, DK; 

Kitaoka, HB; Kaufman, KR: Effects of anklefoot, 

articulated, and non-articulated hindfoot 

orthoses on foot/ankle kinematics during walking. 

Gait and Clinical Movement Analysis Society 


Watanabe, K; Kitaoka, HB; Crevoisier, XM; 

Fuiji, T; Berglund, LJ; Kaufman, KR; An, KN: 

Lateral column lengthening for posterior tibial 

tendon dysfunction and flatfoot. Orthopaedic Research 


FOOT/ANKLE 33

34 ···························································· GAIT ························································· 

Ground Reaction Forces During Partial 

Weight Bearing Gait Using Conventional 

Assistive Devices: A Feasibility Study 

Project Coordinator: Brian Kotajarvi, PT: 


The purpose of this project is to describe the 

peak vertical ground reaction force of both lower 

extremities as a percentage of body weight when 

walking independently and with assistive devices 

such as a conventional cane, axillary crutches, 

forearm crutches, and a wheeled walker. It was 

hypothesized that after training healthy subjects 

to use a walker and a steppage gait pattern, they 

would consistently load the right lower extremity 

at a target of 50% body weight when measured 

by a force platform. Performance with axillary 

and forearm crutches would be less consistent 

with ground reaction forces greater than 50% 

body weight; ambulation with a conventional 

cane would be the least consistent with ground 

reaction forces larger than those produced when 

walking with crutches. 

Ten subjects (5 men and 5 women) between 

the ages of 20 and 40 years, were recruited from 

the Mayo Program in Physical Therapy. Each 

subject had no prior history of surgery to the 

lower extremities and minimal experience using 

an assisted ambulatory device. Minimal use was 

defined as use of a cane or crutches in the past 

for up to a week to reduce lower extremity load 

bearing as a result of injury to the foot, ankle, 

knee, or hip. Each subject had adequate upper 

extremity muscle performance to permit proper 

use of an ambulatory aid. Muscle performance 

of the shoulder depressors, elbow extensors, 

wrist flexors and extensors, and finger flexors 

were assessed by a standard manual muscle test 

by a physical therapist. 

Vertical ground reaction force data from both 

lower extremities during walking were recorded 

by a series of four force platforms embedded in a 

9-meter walkway. Foot markers were fixed to the 

subjects' shoes to permit measurement of selfselected 

walking speed, using a passive videobased 

motion measurement system 

Each subject traversed the walkway in the 

Motion Analysis Laboratory using each of the 

following devices: (1) a single conventional ad- 

justable aluminum cane (2) a front-wheeled adjustable 

aluminum walker with 5-inch swivel 

wheels, (3) adjustable aluminum axillary 

crutches, and (4) adjustable aluminum Lofstrand 

crutches. In addition each subject will also walk 

along the walkway without an assistive device. 

Each subject used the ambulatory device and attempted 

to place 50% PWB on the right lower 

extremity. The vertical ground reaction forces 

(Figure 44) differed significantly for each of the 

six walking conditions (normal gait, standard 

walker, front-wheeled walker, axillary crutches, 

Lofstrand crutches, and standard cane). The data 

are currently being processed for statistical significance 

and manuscript preparation. 

Figure 44: Mean and SD vertical ground reaction force 

values (AD= assistive device, AC= axillary crutches, LC= 

Lofstrand crutches, WW= wheeled walker) 

A Pilot Study to Assess Dynamic Balance 

Control in People with Moderate Traumatic 

Brain Injury 



Traumatic brain injury (TBI) is one of the 

most challenging problems faced by the medical 

community. The ability to identify any functional 

impairment after a TBI that may affect patient 

safety is critical for prevention of re-injury 

during the recovery period. Thus, the purpose of 

this study was to quantitatively assess dynamic 

stability that did not have an obvious neuromuscular 

origin in individuals following TBI. Ten 

subjects with documented TBI and 10 age, gender, 

and stature-matched healthy individuals participated 

in the study. The subjects with TBI had

complaints of “imbalance” or “unsteadiness” 

while walking despite a normal gait on clinical 

examination. In order to understand the association 

between cognitive function of TBI patients 

and their ability to maintain balance during balance-challenging 

tasks, such as negotiating obstacles, 

all subjects were instructed to perform 

unobstructed level walking and to step over obstacles 

corresponding to 2.5%, 5%, 10%, and 

15% of their height. A 13-link biomechanical 

model of the human body was used to compute 

the kinematics of the whole body center of mass 

(COM). Subjects with TBI walked with a significantly 

slower gait speed and shorter stride 

length than their matched controls. Furthermore, 

subjects with TBI displayed a significantly 

greater and faster medial-lateral (M-L) COM motion 

and maintained a significantly greater M-L 

separation distance between their COM and center 

of pressure than their matched control subjects. 

These measurements indicate that subjects 

with TBI have difficulty maintaining dynamic 

stability in the frontal plane and have a reduced 

ability to successfully arrest their sagittal momentum. 

These findings provide an objective 

measurement that reflects the complaints of instability 

not observable on clinical examination 

for individuals who have suffered a TBI. 

Development of New Outcome Measures for 

Patients with Progressive Multiple Sclerosis: 

Proposal for Extension of a Feasibility Study 



Multiple sclerosis (MS) is a progressive central 

nervous system disease primarily affecting 

young adults. Deterioration in MS may occur 

over many months or years, making detection 

difficult. Current clinical outcome measures, 

such as the Expanded Disease Status Scale 

(EDSS) and Ambulatory Index (AI), are weakly 

responsive to changes in gait disorders and current 

surrogate markers (MRI) have inadequate 

predictive validity. Lack of sensitive assessment 

tools may inhibit early detection and intervention 

efficacy assessment. Gait analysis adds objective, 

reliable outcome measures (center of mass displacements 

and velocity and temporal distance 

GAIT 35 

parameters such as stride length, step width, cadence) 

sensitive to detecting neurological deterioration. 

In 2000 we completed an eighteen-month 

study on eighteen patients with progressive MS. 

The subjects each completed five separate motion 

analysis studies and it was determined that 

1) gait analysis variables have excellent testretest 

reliability; 2) several gait variables predict 

worsening deterioration despite the absence of 

change in clinical measures; 3) different gait 

variables may predict deterioration for different 

gait abnormalities (spastic, ataxic, and spasticataxic). 

Since only 5 of the 18 patients experienced 

clinically definite disease progression as 

indicated both by gait analysis measures and 

EDSS scores, our ability to determine the predictive 

and concurrent validity of the quantitative 

outcome measures was limited. This extension 

requested to further examine the original thirteen 

subjects who did not exhibit clinical disease 

come back for a 4-year and 5-year visit to determine 

if all of the patients will reach clinically 

detectable deterioration thresholds. Ten of the 

thirteen subjects are able to participate and have 

returned for their 4-year visit and thus far two 

subjects returned for their final 5-year visit. 

Gait data were collected from 10 subjects 

(M=6, F=4; 54± 8 y) with confirmed progressive 

MS. Subjects were studied during level walking 

using a ten-camera motion system to collect 3-D 

marker trajectories from 28 retro-reflective markers 

at 60 Hz. Data from at least three complete 

gait cycles were collected 48 months after their 

initial observation. Whole body center of mass 

(COM) was calculated using a 13-link biomechanical 

model. COM displacements and velocity 

were calculated in three orthogonal directions 

while commercial software was used to calculate 

temporal distance parameters. Data were analyzed 

using a repeated measures ANOVA and 

Dunnetts post-hoc test (p

36 GAIT 

in either the subjects EDSS or AI scores. When 

separating the patients into two groups either 

with their diagnosis (primary progressive or secondary 

progressive) there was no significant difference 

between the groups. Also, when separating 

the patients into either spastic or spastic/ 

ataxic gait there was no significant difference. 

These results agree with other studies that 

have found ML velocity and displacement to be 

sensitive markers of subtle gait dysfunction. Gait 

analysis was able to characterize subtle changes 

indicative of disease progression, which were not 

detected by clinical evaluation scales. Gait analysis 

is an objective technique sensitive to specific 

parameters for the detection of functional deterioration, 

warranting further study as potential 

outcome measures in MS intervention studies. 

Publications: 

Brey, RH; Chou, LS;. Basford, JR; Shallop, JK; 

Kaufman, KR; Walker, AE; Malec, JF; Moessner, 

AM; Brown AE: Optokinetic testing of patients 

with traumatic brain injury compared to 

normal subjects. Association for Research in 

Otolaryngology 2002. 

Chou, LS; Walker, AE; Kaufman, KR; Brey, 

RH; Basford, JR: Identifying dynamic instability 

during obstructed gait in post-traumatic brain injury. 

Fourth World Congress of Biomechanics, 


Kaufman, KR; Brey, RH; Chou, LS; Rabatin, 

AE; Basford, JR: Comparison of subjective and 

objective measurements of balance disorders following 

traumatic brain injury. Foruth World 

Congress of Biomechanics 2002. 

Walker, AE; Basford, JR; Chou, LS; Brey, RH; 

Kaufman, KR: Center of mass gait patterns of 

patients with mild to moderate traumatic brain 

injury. Gait and Clinical Movement Analysis Society 

2002.

····················································· REHABILITATION ········································ 37 

Aerobic Exercise Intervention for Knee 

Osteoarthritis 



Arthritis is one of the most common causes of 

functional limitation and dependency in the 

United States. Individuals with osteoarthritis 

(OA) restrict joint motion and limit activity in 

order to decrease their symptoms. Traditional, 

conservative medical treatment of OA has been 

directed at improving functional status through 

reducing joint pain and inflammation and maintaining 

or restoring joint function. Exercise as an 

adjunct therapy in the clinical management of 

patients with OA of the knee, however, is not 

uniformly accepted. 

This is a 4-year project to study the effect of 

exercise on 306 patients with knee OA in a prospective, 

randomized, controlled trial. The objectives 

are to quantitatively determine the effects 

of aerobic exercise on patients with knee 

OA, compared to a control group. A complete 

orthopedic examination and objective biomechanical 

measurements (x-rays, MRI, and gait 

analysis) will be obtained upon enrollment. The 

joint space will be measured from the knee xrays 

and the cartilage thickness will be measured 

from the MRI. In addition, the subjects will 

complete a general health status questionnaire a 

disease/site specific questionnaire, a visualanalog 

scale rating of their pain, and an activity 

index to assess current activity level. Subjects 

will have Grade II or III OA and will be randomized 

into either a control group or one of two exercise 

groups (treadmill walking or stationary 

cycling). The subjects in the exercise group will 

be required to exercise three times a week for 

one year. 

The hypotheses for this study are: 1) clinical 

outcome measures will be better in patients enrolled 

in exercise programs than in control patients, 

2) quantitative measures of lower extremity 

function will not decline over time with an 

effective aerobic exercise program, and 3) an effective 

exercise program for adults with degenerative 

joint disease is dependent on knee compartment 

involvement, OA stage, BMI, and type 

of exercise prescribed. 

This first year has been spent in preparation to 

see such a large volume of subjects. An exercise 

log has been created and a database is in the final 

stages of development to track exercise compliance. 

Most of the forms have been made scannable 

and we have been testing the procedures to 

ensure we can collect reliable and valid data. 

Also, an edge detection program is being developed 

to calculate joint space narrowing from xray 

images. A set of stairs with seven steps has 

been constructed that have built-in instrumented 

force platforms integrated into the stair structure 

(Figure 45). 

Figure 45: Instrumented Stairs 

Repeatability of X-ray, MRI and Gait 

Analysis 



The reproducibility of quantitative gait analysis 

measurements, knee x-rays and MRI are an 

important consideration when analyzing data of 

both normal subjects and patients. This study is 

being done in order to assess the repeatability 

and validity of these measurement techniques at 

a one-week assessment interval. Selected characteristics 

of gait, such as temporal distance parameters, 

and hip, knee and ankle kinetics and 

kinematics will be studied prospectively. 

Weight-bearing anterior/posterior (AP) and lateral 

radiographs of the knee will be taken to ensure 

the reproducibility of the knee flexion positions 

and consistency of joint space measurements 

(Figure 46). Finally, cartilage thickness

38 REHABILITATION 

measurements will also be analyzed using MRI 

(Figure 47). Forty subjects have been enrolled to 

obtain knee x-rays and twenty of these will also 

receive a gait analysis and MRI. 

In order to obtain inter-rater and intra-rater 

reliability, one technician will acquire knee xrays 

on all forty subjects for their initial visit. 

For the return visit the subjects will be divided so 

the same technician will re-test twenty of the 

subjects and a different technician will test the 

remaining twenty subjects. Similarly, for the gait 

analysis one kinesiologist will test twenty subjects 

for their initial visit. For the return visit the 

subjects will be divided so the same kinesiologist 

will re-test ten of the subjects and a different 

kinesiologist will test the remaining ten subjects. 

Forty normal, health subjects have been recruited. 

All forty subjects have completed their 

x-rays and twenty of them have obtained a MRI. 

Data collection and analysis is ongoing. 

Figure 46: Frontal view x-ray. 

Figure 47: Left: Frontal view MRI. Right: Sagittal 

View MRI. 

Effect of Kyphosis on Balance and Gait in 

Elderly Osteoporotic Individuals: A 

Controlled Trial 

Project Coordinator: Christine Hughes 


Osteoporosis is, and will continue to be, a major 

health concern for the population. Falls in 

combination with low bone mass result in over 

200,000 geriatric hip fractures each year in the 

United States. Imbalance and tripping over obstacles 

during gait were reported as two of the most 

common causes of falls in the elderly. Balance 

may be a factor that can respond to intervention 

and result in reduction in risk of falling. 

Epidemiology of falls has shown that about 

50% of the falls occur during some form of locomotion. 

However, only a few studies on dynamic 

balance control were performed during 

locomotion and were limited to unobstructed 

level walking. MacKinnon found an active hip 

abduction moment about the supporting leg 

played a crucial role in maintaining balance of 

the trunk and swing leg. According to Kaya, 

healthy elderly adults limited momentum generation 

of the whole body by decreasing gait velocity, 

however, excessive lateral momentum was 

found in balance-impaired elderly adults. Postural 

stability and balance performance decrease 

with age, and there is lack of knowledge concerning 

how dynamic stability of the whole body 

is maintained during balance-challenged ambulatory 

tasks, such as obstacle crossing. Objectively, 

the differences in balance, gait and 

strength in subjects with osteoporosis and kyphosis 

have not been adequately studied. 

This study is designed to investigate the influence 

of osteoporosis and kyphosis on gait unsteadiness 

and falls in elderly individuals and to 

correlate the unsteadiness of gait to the related 

muscle weakness. The proposal was for fourteen 

healthy individuals without kyphosis and fourteen 

osteoporotic or osteopenic individuals with 

kyphosis to have a gait analysis, strength test, 

and posturography. The individuals with kyphosis 

will repeat all these tests after a four-week 

trail of use of the Posture Training Support 

(weighted kypho-orthosis). Thus far we have 

completed testing on 13 subjects with kyphosis

(both visits) and 13 subjects without kyphosis. 

Gait and strength data were collected from 13 

female elderly subjects with osteoporosis and 

kyphosis (O/K) and 13 female elderly healthy 

subjects (controls). The O/K subjects had a 

mean age of 76 (±5) and the normal subjects had 

a mean age of 71 (±5). Computerized isometric 

strength evaluations were conducted on all participants. 

3-D motion analysis data were collected 

while the subjects were studied during unobstructed 

level walking and while stepping over 

an obstacle of four randomly assigned different 

heights (2.5%, 5%, 10%, and 15% of the subject’s 

height). The center of mass (COM) displacements 

and velocity was calculated in three 

orthogonal directions. 

The strength data demonstrated that overall, 

the controls are stronger on all muscle groups 

tested (Figure 48). There is a significant difference 

in the A/P and M/L displacements and velocities. 

The results show that the O/K subjects 

had less A/P displacement, greater M/L displacement, 

reduced A/P velocity and increased M/L 

velocity when compared to the controls. This 

was true for all conditions of unobstructed and 

obstructed level walking. Also, there is a significant 

effect of obstacle height on all COM parameters. 

Finally, there is no significant interaction 

for any of the COM parameters between the 

groups and the obstacle heights. When analyzing 

the temporal-distance parameters between the 

two groups there was a significant difference in 

the right and left step length, stride length, velocity, 

and cadence (Table 2). Data collection and 

further analysis are underway. 

Figure 48: Lower extremity isometric strength results 

for patient and control groups. 

REHABILITATION 39 

Table 2: Temporal distance results for patient and control 

groups. 

Modeling of the Upper Extremity in 

Wheelchair Propulsion 

Project Coordinators: Kristin Zhao 

zhao.kristin@mayo.edu 

Wheelchair design affects the performance of 

propelling a wheelchair. Variability in the propulsion 

technique in manual wheelchair users 

due to differences in level of injury and wheelchair 

fit makes detection of subtle changes in 

technique nearly impossible, especially when 

collecting experimental measures dynamically. 

Biomechanical models can add to our understanding 

of how upper extremity segments and 

muscles interact to execute the motor task. The 

purpose of this study was to examine the potential 

propulsion moments at various points during 

the propulsion cycle by simplifying the wheelchair 

movement to a static task. We have collected 

experimental data from subjects exerting 

maximal effort to propel an instrumented wheelchair 

with its wheel in a locked position. Then, 

we used the subject's anthropometry and strength 

data to develop a subject-specific static model of 

the upper arm, forearm, and wheelchair wheel to 

add to our understanding of the wheelchair-user 

interface. 

Five healthy male subjects (mean age, 35.2 

years) who were inexperienced wheelchair users 

participated in this study. Anthropometric measurements 

were collected from all subjects, including 

the length of the upper arm and forearm, 

as well as the shoulder position relative to the 

wheel axle. 

An instrumented wheel system was used to 

measure three-dimensional forces and moments 

on the handrim at different phases during wheelchair 

propulsion (Figure 49) The wheelchair had 

a handrim radius of 25.4 cm and was locked to


Figure 49: Adjustable seat wheelchair 

Figure 50: Four segment model used for static optimization 

of wheelchair propulsion. The shoulder (S), elbow 

(E), and hand (H) positions are indicated. Vector displacements 

from the shoulder, elbow, and wheel axle to 

the hand position are PS, PE, and Pr respectively. θS, θE, 

and θW are the shoulder joint, elbow joint, and wheel angles 

respectively. Finally, the resultant hand force on the 

handrim (Fh) as well as its Cartesian (FX, FY) and polar 

coordinates (Fr, Ft) are indicated. 

prevent forward movement as the subjects propelled 

the handrim with maximum effort. Five 

hand positions corresponding to wheel angles (θ) 

of 120º, 105º, 90º, 75º, and 60º (as defined in 

Figure 50) were assigned to the subject in a random 

order. To estimate the joint strength in an 

isolated loading condition, the isometric shoulder 

Figure 51: Mean and standard deviation of handrim and 

progression moment at five different hand positions. 

Figure 52: Upper extremity and handrim forces at 60° 

wheel angle. 

flexion and extension muscle strengths were 

measured at positions throughout the elbow and 

shoulder range of motion using a Kincom 125 

AP ® dynamometer. 

The rationale for the model is, given a subject-specific 

profile of the strengths of each of 

the upper extremity joints as a function of joint 

angle, there is an optimal direction of force application 

to the handrim to maximize the propulsion 

moment about the wheel axle at each instant 

throughout the propulsion cycle. Kinematics of 

the upper extremity were determined using a 

four-bar linkage model including the upper arm, 

lower arm, hand to wheel axle, and shoulder to 

wheel axle segments (Figure 50). Each subject's 

anthropometry and joint strength profiles were 

input into the model to obtain subject-specific 

model predictions. 

Results of experiments revealed the progression 

moment was greater at both initial and terminal 

propulsion positions (i.e. wheel angles of

120º and 60º respectively) and was smaller in the 

mid-propulsion position (i.e. wheel angle of 90º) 

(Figure 51). The difference in progression moments 

between model and experiment was small 

in initial and mid-propulsion hand positions but 

greater in the terminal propulsion positions. The 

directions of applied force applied to the handrim 

by both experiment and model in the different 

hand positions were similar (Figure 52). 

Results from the current study show that 

the hand force vector is roughly tangential to the 

handrim during the wheelchair propulsion cycle. 

The force vector is directed away from the wheel 

axle when the hand position is posterior to top 

dead center and toward the wheel axle when the 

hand position is anterior to top dead center. To 

generate a push force directed away from the 

wheel axle, the elbow flexor must be activated. 

And this would indeed be beneficial for propulsion 

as the elbow must flex during this phase of 

the cycle (behind top dead center). However, 

halfway through the propulsion phase the applied 

force must change to progress the wheel so the 

elbow extensor needs to be immediately activated 

at that point in the cycle. During static propulsion, 

switching from elbow flexion to extension 

is not difficult, however, the change in muscle 

activation from elbow flexor to elbow extensor 

dynamically may result in a more complex 

and inefficient movement. A downward directed 

handrim force would facilitate elbow extension 

so it may be that users choose to apply force 

downward throughout the whole cycle rather 

than switch from flexion to extension. However, 

it could then be hypothesized that users could be 

trained through biofeedback to activate their 

muscles more optimally and achieve a pattern of 

activation more like that seen during static analysis. 

Kinematics of the Upper Extremity in 

Wheelchair Propulsion 

Project Coordinator: Brian Kotajarvi: 


Upper extremity pain and dysfunction are 

common among people who use manual wheelchairs 

for mobility. For example, surveys involving 

as many as 450 wheelchair-based indi- 


viduals find that as many as 73% report some 

degree of chronic upper extremity pain, which 

they attribute primarily to wheelchair propulsion 

and transfers. Individuals with paraplegia are 

particularly active wheelchair users and, when 

questioned, report prevalence of shoulder, elbow, 

and wrist/hand pain between 30% and 65%, with 

the shoulder the most common site of involvement. 

These issues have not gone unnoticed and 

the biomechanical analysis of manual wheelchair 

propulsion has become increasingly common 

over the last decade. The aim of this research is 

twofold. First, to improve propulsion mechanical 

efficiency and second, to gain better understanding 

of the wheelchair-user interface to address 

the musculoskeletal problems associated with 

wheelchair use. 

Towards these ends, we are studying the 

wheelchair-user interface by (1) refining our 

understanding of the effect of seat position on 

handrim biomechanics and (2) extending our 

research to include a comparison of experienced 

and inexperienced users. We hypothesized that 

lower seat heights and posterior seat positions 

would be associated with a more efficient stroke 

as defined by decreased stroke frequency, 

increased push angle, prolongation of the push 

phase, higher fraction effective force and a larger 

propulsion moment. Furthermore, we suspected 

that experienced users would demonstrate more 

efficient propulsion than their less experienced 

counterparts. 

Twenty inexperienced and thirteen 

experienced users propelled a wheelchair over a 

smooth level floor. Kinetic and kinematic data 

were collected using an instrumented rim and a 

motion analysis system. A ten-camera 

Expertvision System was used to collect the 3- 

D trajectory data of markers placed on the left 

upper extremity and wheel. A wheelchair wheel 

and handrim instrumented with a six-component 

load cell developed and validated in this 

laboratory was used to collect kinetic data. The 

handrim was mounted to one side of the load 

cell, and the other side of the load cell was 

mounted directly to the wheel. Load cell output 

voltages were recorded at a sampling frequency 

of 100 Hz and stored in a miniature data logger 

mounted on the wheel. Data from the motion


analysis system and the data logger were 

synchronized with an external trigger. The side 

frames of the chair containing the axles were 

modified with a custom designed aluminum 

mechanism fixed to the frame of chair, which 

permitted 8-cm changes in the axle’s horizontal 

position and 10-cm changes in its vertical 

position with the turn of a knob (Figure 49). Nine 

different axle horizontal and vertical positions 

were studied. These positions were studied for 

all subjects and included axle positions 

commonly used in the clinical setting. 

Results showed that the experienced users 

demonstrated prolonged stroke distance, 

decreased stroke frequency, longer push time, 

and a longer push angle than the inexperienced 

group (Fig 53-54). A shorter distance between 

the axle and shoulder (low seat height) resulted 

in the highest push times and push angle for both 

groups. Propulsion efficiency as measured by 

the fraction of effective force did not 

Figure 53: Temporal distance variable 

Figure 54: Wheel angle variables 

significantly change with seat position. It was 

concluded that propulsion experience may play a 

major role in minimizing peak handrim loads. 

A Longitudinal Assessment and Biofeedback 

Analysis of Wheelchair Propulsion to Reduce 

the Risk of Injury and Improve Propulsion 

Efficiency 

Project Coordinators: 

Kristin Zhao, zhao.kristin@mayo.edu 

Brian Kotajarvi, kotajarvi.brian@mayo.edu 

More than a million people rely on wheelchairs 

for mobility in the United States. This 

number is growing by tens of thousands each 

year and will continue to do so as the population 

ages. Wheelchair users are a diverse group that 

includes the disabled elderly, as well as individuals 

with mobility limiting conditions such as spinal 

cord injury, stroke, and multiple sclerosis. 

However, despite a high and growing prevalence 

of wheelchair use, relatively little rigorous scientific 

data is available pertaining to manual wheelchair 

propulsion, design, and prescription. 

This study is designed to investigate two aspects 

of wheelchair propulsion. The first aspect 

is the relationship between longitudinal changes 

in propulsional performance, biomechanics, and 

the risk of musculoskeletal injury. The second is 

to study the potential of a novel biofeedback 

training paradigm in optimizing the learning of 

safe and effective propulsion techniques. More 

specifically, we will (1) study wheelchair users in 

a longitudinal fashion in order to understand the 

changes in propulsion efficiency (FEF) and risk 

potential (WPSR) that occur with increasing experience, 

and (2) monitor the feasibility, and assess 

the short term impact, of an audiovisual biofeedback 

program on wheelchair propulsion dynamics 

as reflected by the FEF. 

We have made significant progress in development 

and validation of the instrumented 

wheelchair handrim and biofeedback systems. In 

addition, we have built a new and improved 

wheelchair treadmill that not only replicates the 

"feel" and frictional characteristics of propulsion 

on level ground and inclines, but also incorporates 

safety features that make it suited for pa-

tient use (see below). 

The new wheelchair treadmill is constructed 

on a tubular steel frame and measures 

1.5 x 1.5 m (5 x 5 ft) (Figure 55). Each of the 

wheelchair rear wheels contacts against a single 

38.1 cm (15 in) diameter drum. A frictional brake 

provides mechanical resistance to simulate propulsion 

up ramps. A computer monitor is used 

to display visual feedback data. Preliminary kinetic 

data obtained with the treadmill agrees favorably 

with data collected in our laboratory over 

level ground. 

Data collection and biofeedback software 

Kinetic data acquisition 

A Quickie II ultra-lightweight sport wheelchair 

is used for all data collection with the custom-designed 

wheelchair wheel and a sixcomponent 

load cell instrumented handrim. 

A personal computer system is used to execute 

two versions of visualization biofeedback. 

The first features feedback visualization of the 

propulsion force vectors (point of force application, 

direction, magnitude) along the handrim as 

well as displaying each force component (Fx, Fy, 

Fz, Ft, Fn) versus time (Figure 56). The second 

version calculates and displays fraction effective 

force as a flowchart in time. A third version is 

currently under development that will enable an 

individual to control computer video games by 

driving their wheelchair. 

Figure 55: Quickie II ultra-lightweight sport wheelchair 

with instrumented handrim. 


Following this development phase of the 

project, we will begin testing. Twenty-four individuals 

hospitalized at the Mayo Clinic hospital 

for the new onset of a first spinal cord injury. 

The group will be comprised of eight with traumatic 

injury, eight with a traumatic injury and 

stroke. They will be followed for two years to 

assess longitudinal adaptations during wheelchair 

propulsion as well as the affect of feedback on 

propulsion efficiency. 

Figure 56: Treadmill data collection software interface. 

Objective Quantification of Tremor Severity 

Project Coordinator: Duane Morrow: 

morrow.duane@mayo.edu 

Current clinical practice for the assessment of 

tremor severity in patients with varying pathologies 

relies heavily upon subjective rating scales. 

In order to more properly assess the impact of 

both surgical and/or non-surgical interventions, a 

more objective and precise method of measurement 

is needed. This method, termed Quantitative 

Movement Analysis (QMA), was developed 

in our laboratory as just such a tool, and has recently 

been put into clinical practice at Mayo for 

patients with severe cerebellar tremor associated 

with multiple sclerosis. In brief, QMA utilizes 

power-spectrum analysis techniques to distinguish 

the amount of tremor in a subject’s hand as 

they produce motions which simulate activities 

of daily living. 

Other published measures of tremor measurement 

have shown an ability to differentiate between 

disease states in various patients. This is


mainly due to the fact that the tremor that results 

as sequelae for differing pathologies will have a 

different modal frequency. Our algorithm takes 

not only the modal frequency into account, but 

also quantifies the magnitude of the tremor signal 

as well. 

Hand velocity data is derived from position 

data recorded by a 3-DOF electromagnetic tracking 

device. Activities measured include reaching 

from the subject’s left to right as though manipulating 

objects on a table, and reaching from a tabletop 

towards their mouth to simulate feeding. 

The frequency content of the signals is then calculated 

via fast Fourier transformation, and the 

subsequent power spectral densities (PSD) are 

calculated as well. The area under the modal frequency 

peak on the PSD plots is taken as the 

power of the tremor. 

In preliminary studies, tremor severity as 

measured by QMA has shown promise in its ability 

to predict surgical efficacy (Figure 57). There 

appears to be a threshold in tremor severity, as 

measured by QMA, above which the attempt at 

surgical reduction of tremor is warranted. These 

findings are also supported by the qualitative results 

of “Excellent”, “Good”, and “Poor” as assigned 

by the patients’ neurosurgeon. 

In previously published abstracts, QMA has 

been shown to be repeatable and sensitive to 

changes in condition. Ongoing efforts in the 

Figure 57: Correlation between baseline QMA scores and 

QMA improvement. Note the agreement in the results 

with the subjective evaluation by the neurosurgeon. 

tremor project currently center around refinement 

of the peak-defining algorithm and application of 

QMA in discerning between cerebellar tremor 

and other motion disorders, such as ataxia. Other 

efforts by the laboratory will also include expanding 

the database of collected patient scores 

to include those with other pathologies to determine 

if a similar threshold for surgical efficacy 

may exist. 

Publications 

Guo, LY; ET AL. Modeling of manual wheelchair 

propulsion using optimization. American 

Society of Biomechanics 2002. 

Guo, LY; Su, FC; An, KN: Optimum propulsion 

technique in different wheelchair handrim diameter. 

J Med Biol Eng 22(1):1-10, 2002. 

Kotajarvi, BR; Basford, JR; An, KN; Sabick, M: 

Does elbow angle affect wheelchair propulsion 

effectiveness. American Association of Physical 

Medicine and Rehabilitation 2002. 

Kotajarvi BR; Basford, JR; An, KN: Upperextremity 

torque production in men with paraplegia 

who use wheelchairs. Arch Phys Med Rehabil 

83(4):441-446, 2002. 

Morrow, DA; Matsumoto, J; Rabatin, AE; Kaufman, 

KR: Comparison of quantitative measures 

of tremor as predictors of surgical outcome. 

American Society of Biomechanics 2002. 


KR: Comparison of quantitative measures 

of tremor as predictors of surgical outcome. 

Fourth World Congress of Biomechanics 2002. 


KR: Quantitative analysis for predicting 

surgical reduction of ms tremor. Gait and Clinical 

Movement Analysis Society 2002.

·················································· MUSCLE MECHANICS ···································· 45 

Skeletal Muscle Characterazation by 

Magnetic Resonanace Elastography 

Project Coordinator: Qingshan Chen: 

chen.qingshan@mayo.edu 

Magnetic Resonance Elastography (MRE) is a 

novel technique recently developed at Mayo 

Magnetic Resonance Research Laboratory for 

non-invasively measuring the stiffness of biological 

tissues. The technique employs standard 

MRI equipment with a few modifications and an 

electromechanical vibrator that applies vibration 

to the test material. This technique has been 

safely and successfully applied to in vivo tissues 

in humans, specifically to skeletal muscle. In collaboration 

with the magnetic Resonance Research 

Laboratory, MRE is applied to the study 

of skeletal muscle biomechanics. Skeletal muscle 

is of biomechanical interest since its stiffness 

changes reversibly and voluntarily during muscle 

contraction. The stiffness modulus can be 

measured in multiple muscles simultaneously 

during a biomechanical task and these moduli are 

then used to calculate the tension in each muscle. 

MRE also has potential as a diagnostic tool inmuscle 

disease such as stroke, hyperthyroidism, 

disuse atrophy, or paralysis. MRE works by creating 

shear waves within tissues, which propagate 

away from the vibration source. The electromechanical 

vibrator, or “tapper”, is the wave 

source. The MRI scanner is sensitized to cyclic 

motion within the tissue so that the shear waves 

can be visualized. Muscle loading devices have 

been developed for the ankle, knee, and the 

neck. Material stiffness is determined from the 

MRE image by measuring the wavelength of the 

shear waves. Regional wavelength estimation 

method is being developed to measure the wavelength 

of the shear waves. 

The MRE technique will be applied in vivo to 

provide elastographic images of abnormal muscle 

with known disorders. The patient groups 

chosen for study are each important in their own 

right, and furnish unique information across the 

spectrum of muscular disease and dysfunction. 

Groups to be studied include individuals with 

new onset of spasticity following an ischemic, 

hemispheric stroke, disuse atrophy as a result of 

immobilization, metabolic (hyperthyroid) 

myopathy and myofascial pain for trigger point 

identification. 

The overall hypothesis of this work is that 

MRE is a novel and exciting technology that will 

bring benefits to both basic research and clinical 

care. 

Figure 58: Wave propagation through gastrocnemius 

Microsensor for Intramuscular Pressure 

Measurement 

Project Coordinator: Duane Morrow: 

morrow.duane@mayo.edu 

Currently, the integrated electromyogram 

(EMG) is the standard used as an indirect indicator 

of the timing and intensity of muscle contraction. 

However, the relationship between EMG 

and muscle tension is unclear. There is a need 

for a reliable measure of muscle tension under 

dynamic conditions. This need may be filled 

through the measurement of intramuscular pressure 

(IMP). The overall objective of this project 

is to provide a useful clinical tool for in-vivo 

quantification of muscle force. The specific aims 

of this proposed project will be to further develop 

a fiber optic microsensor for monitoring intramuscular 

pressure and validate this technology 

by animal testing, theoretical modeling, and invivo 

evaluation of research subjects and patients 

with muscle disorders. 

A prototype optical fiber micropressure sensor 

for in-vivo muscle pressure (IMP) measurement 

(Figure 59) has been developed. The diameter 

of the prototype sensor is 500µm. A patent 

application has been filed regarding this de-

46 

Figure 59: Intramuscular Pressure Microsensor 

sign. 

In the past year, progress has been made to 

examine the performance characteristics and biocompatability 

of the microsensor. We have also 

continued to further develop mathematical modeling 

of muscle performance in order to be able 

to compare experimentally collected results to 

theoretical models of muscle function. Additionally, 

our collaborators at the University of California, 

San Diego have tested the microsensor in 

an animal model to further examine its ability to 

sense both active and passive muscle tension. 

The performance characteristics of the microsensors 

are evaluated in a calibration chamber. 

The microsensor had an accuracy, repeatability, 

and linearity better than 2% full-scale 

output (FSO) and hysteresis slightly higher than 

4.5% FSO over a range of 0-250 mmHg. 

The transducer was evaluated for biocompatibility 

using the International Organization for 

Standardization (ISO) Standard 10993-6: 

“Biological evaluation of medical devices- Part 

6: Tests for local effects after implantation.” 

The test compared the biological response of the 

test sensor to a control material made of inert silica 

glass. The test and control materials were implanted 

in the paraspinal muscles of the adult 

New Zealand white rabbits while the animals 

were anesthetized. This study demonstrated that 

the pressure microsensor is biocompatible. 

A mechanical muscle model for simulating 

muscle mechanics was developed based on 

the finite element method. Nonlinear continuum 

mechanics were used to study the contractile active 

and passive properties of skeletal muscle. 

This model was used to predict intramuscular 

pressure. (Figure 60). 

A relationship between intramuscular 

pressure and active and passive muscle tension 

was determined using the isolated tibialis anterior 

(TA) of a New Zealand white rabbit. The knee 

was fixed in a custom jig, and the distal tendon 

of the TA was attached to a servomotor. A fiber- 

optic pressure transducer was inserted into the 

TA. The peroneal nerve was stimulated to define 

optimal length (L0) and the length-tension curve 

was created. The shape of this curve presumably 

represents a scaled and distorted version of the 

sarcomere length-tension curve previously published. 

Passive muscle tension increased in a 

fairly exponential fashion at length beyond optimal 

length. The length-pressure relationship 

(Figure 61) generally mimicked the shape of the 

length-tension curve (Figure 62). These data indicate 

that IMP measurement provides a fairly 

accurate index of relative muscle tension. 

Figure 60: Comparison of experimental and simulated 

muscle pressure and tension.

Figure 61: Intramuscular pressure-length relationship for 

active and passive tension. 

Figure 62: Muscle length-tension relationship for active 

and passive tension. 

Publications: 

An, KN: Muscle force and its role in joint dynamic 

stability. Clin Orthop 403S:S37-S42, 


Gabriel, DA; Basford, JR; An, KN: Vibratory 

facilitation of strength in fatigued muscle. Arch 

Phys Med Rehabil 83:1202-1205, 2002. 

Gabriel, DA; Proctor, D; Engle, D; Sreekumaran, 

N; Vittone, J; An, KN: Application of the La- 

Grange polynomial in skeletal muscle fatigue 

analysis. Res Q Exerc Sport 73(2):168-174, 


MUSCLE MECHANICS 47 

Jenkyn TR, Koopman B, Huijing P, Lieber RL, 

Kaufman KR: Finite element model of intramuscular 

pressure during isometric contraction of 

skeletal muscle. Physics in Medicine and Biology, 

47:4043-4061, 2002. 

Kaufman, KR; Davis, J; Wavering, T; Morrow, 

D; Lieber, RL: Intramuscular pressure is an indicator 

of muscle force. Fourth World Congress of 

Biomechanics 2002. 

Newcomer, KL; Jacobson, TD; Gabriel, DA; 

Larson, DR; Brey, RH; An, KN: Muscle activation 

patterns in subjects with and without low 

back pain. Arch Phys Med Rehabil 83:816-821, 

2002.

48 ············································· TISSUE MECHANICS ·········································· 

Stress Transfer to Coronary Vasa Vasorum 

After Stenting: A Finite Element Model 



In-stent restenosis is a major health issue in 

the United States. Although it is characterized by 

excessive cell proliferation, the underlying 

mechanisms are not well understood. When vasa 

vasorum in the outer wall of coronary arteries 

become damaged, neointimal proliferation and 

subsequent artery stenosis result. We 

hypothesized that stents damage vasa during 

balloon inflation and through sustained 

compression by the individual stent struts. The 

aim of this study was to develop a finite element 

analysis (FEA) model of stented coronary 

arteries to determine stresses induced in the 

vessel wall in order to predict vasa compression 

by the stent, which could initiate the restenotic 

process. 

A 3D FEA model of the coronary artery wall 

(Figure 63) was developed using ABAQUS finite 

element software. Vasa vasorum were modeled 

in the wall at a location 70% of the wall thickness. 

The vessel wall was modeled as isotropic, 

linear elastic, and incompressible. An internal 

pressure of 95mmHg was applied to simulate the 

mean blood pressure. Two sizes of vasa were 

modeled; 50 and 150 µm diameter. Two degrees 

of stent diameter expansion were considered 

(16% and 34%). The expanded 3D stent geometry 

was simulated from 3D microscopic-CT images. 

Stent expansion was simulated by applying 

radial displacements to the nodes of the vessel 

wall surface where stent contact would occur. 

Figure 64 shows the stress contours through a 

plane in the midsection of the model. On the luminal 

stented side, stresses were greater than 

1,500mmHg with a maximum of 4,000 and 

10,000mmHg for 16% and 34% expansions, respectively. 

The stress decreased in a non-linear 

manner with increased distance from the lumen. 

In the region where vasa are located, stresses exceeded 

500 and 1,000mmHg for the two expansions. 

In comparison, prior to stenting, stresses in 

the wall induced by the blood pressure were 

lower (

Agarose Gel Material Properties 

Project Coordinator: Qingshan Chen: 

chen.qingshan@mayo.edu 

Agarose gel is widely used in various fields of 

biomedical research, particularly in tissue culture 

systems because it permits growing cells and tissues 

in a three-dimensional suspension. This is 

especially important in the application of tissue 

engineering concepts to cartilage repair because 

it supports the cartilage phenotype. Because the 

tissue pieces are embedded in the agarose gel, 

understanding the mechanical properties of agarose 

gels can provide basic background knowledge 

for tissue engineering investigations that 

employee dynamic loading to cells or whole tissues 

in vitro. 

Since agarose gel is a porous solid filled with 

fluid, theoretically the gel and the fluid together 

are expected to behave as a biphasic substance 

with characteristic viscoelastic properties. A fractional 

calculus approach had been proposed as a 

method of describing the viscoelastic materials 

and soft tissues. Such models were in sharp contrast 

to the traditional spring-dashpot models because 

these models had simple expressions in 

their constitutive equation and yet represented a 

fairly complex rheological behavior that was often 

difficult to mimic with spring-dashpot systems. 

In the present study, the dynamic mechanical 

properties of agarose gels as a function of frequency 

were characterized with a simple form of 

fractional derivative model. The relationship between 

the model parameters and the agarose concentration 

was also investigated. Gels with agarose 

concentration (weight/volume, w/v) of 2%, 

3%, 4% and 5% were prepared by dissolving appropriate 

amount of Difco Bacto agar into the 

distilled deionized water. Samples 5 mm in thickness 

were used for testing. 

Mechanical properties of the gels were tested 

in frequency sweep shear mode with a dynamic 

mechanical analyzer (DMA 2980, TA Instruments, 

New Castle, DE) over a frequency range 

of 1-20 Hz at a constant amplitude of 20µm. The 

individual complex modulus, elastic modulus and 

viscous modulus were recorded. A fractional derivative 

model was used to model the stress- 

TISSUE MECHANICS 49 

strain relationship of the gel. 

The model indicated that the elastic modulus 

of the agarose gel significantly prevailed its viscous 

modulus. The elastic modulus of agarose 

gels was usually 1-2 orders of magnitude larger 

than its viscous modulus (Figure 65. Moreover, 

elastic modulus and viscous modulus of the agarose 

gel were only slightly frequency-dependent, 

verifying its usage in dynamic pressure stimulation 

in tissue engineering (Figure 66). 

Figure 65: Complex modulus vs. frequency for different 

concentration agarose gels. 

Figure 66: Elastic and viscous modulus as a function of 

frequency for 3% agarose gel.

50 TISSUE MECHANICS 

Publications 

An, KN: Joint mechanics and clinical applications. 

Fourth World Congress of Biomechanics 


Basford, JR; Jenkyn, TR; An, KN; Ehman, RL; 

Heers, G; Kaufman, KR: Evaluation of healthy 

and diseased muscle with magnetic resonance 

elastography. Arch Phys Med Rehab 83:1530- 

1536, 2002. 

Fujii, T; Sun, YL; An, KN; Luo, ZP: Mechanical 

properties of single hyaluronan molecules. J Biomech 

35:527-531, 2002. 

Miles, KA; Smith, J; Riemer, E; Schaefer, MP; 

Dahm, DL;. Kaufman, KR: Wear characteristics 

of common running shoes. American College of 

Sports Medicine 2002. 

Momose, T; Amadio, PC; Sun, YL; Zhao, CF; 

Zobitz, ME; Harrington, JR; An, KN: Surface 

modification of extrasynovial tendon by chemically 

modified hyaluronic acid coating. J Biomed 

Mater Res 59:219-224, 2002. 

Ko, CC; Swift, JQ; DeLong, R; Douglas, WH; 

Kim, YI; An, KN; Chang, CH; Huang, HL: An 

intra-oral hydraulic system for controlled loading 

of dental implants. J. Biomech 35:863-869, 


Ohtera, K; Zobitz, ME; Luo, ZP; Morrey, BF; 

O’Driscoll, SW; Ramin, KD; An, KN: Effect of 

pregnancy on joint contracture in the rat knee. J 

Appl Physiol 92:1494-1498, 2002. 

Sun, YL; Luo, ZP; Fertala, A; An, KN: Direct 

quantification of the flexibility of type I collagen 

monomer. Biochem Biophys Res Commun 

295:382-386, 2002. 

Sun, YL; An, KN; Luo, ZP: Mechanical properties 

of single type II collagen molecule. Orthopaedic 


Zobitz, ME;. Rosol, M; Goessl, M;. Malyar, N; 

An, KN; Kantor, B: Stress Transfer to Coronary 

Vasa Vasorum after Stenting: A Finite Element 

Model. American Society of Biomechanics 2002.

························································ PUBLICATIONS ·············································51 

• Adams BD, Berger RA: An Anatomic Reconstruction of the Distal Radioulnar Ligaments for Posttraumatic 

Distal Radioulnar Joint Instability. J Hand Surg 27A:243-251, 2002. 

• An, KN: Muscle force and its role in joint dynamic stability. Clin Orthop 403S:S37-S42, 2002. 

• An, KN: Joint mechanics and clinical applications. Fourth World Congress of Biomechanics 2002. 

• An, KN; Zobitz, ME; Zhao, CF; Amadio, PC: Gliding characteristics of tendon. Fourth World Congress of 

Biomechanics 2002. 

• Babis GC; Trousdale, RT; Jenkyn TR; Kaufman, KR: Comparison of two methods of screw fixation in 

periacetabular osteotomy. Clin Orthop. 403:221-227, 2002. 

• Babis, G.C., An, K.N., Chao, E.Y.S., Rand, J.A., Sim, F.H.: Double level osteotomy of the knee: A 

method to retain joint-line obliquity. J Bone Joint Surg 84-A(8):1380-1388, 2002. 

• Basford, JR; Jenkyn, TR; An, KN; Ehman, RL; Heers, G; Kaufman, KR: Evaluation of healthy and diseased 

muscle with magnetic resonance elastography. Arch Phys Med Rehab 83:1530-1536, 2002. 

• Brey, RH; Chou, LS;. Basford, JR; Shallop, JK; Kaufman, KR; Walker, AE; Malec, JF; Moessner, AM; 

Brown AE: Optokinetic testing of patients with traumatic brain injury compared to normal subjects. Association 

for Research in Otolaryngology 2002. 

• Chou, LS; Walker, AE; Kaufman, KR; Brey, RH; Basford, JR: Identifying dynamic instability during obstructed 

gait in post-traumatic brain injury. Fourth World Congress of Biomechanics, 2002. 

• Crevoisier, XM; Kitaoka, HB; Hansen, D; Kaufman, KR: Effects of ankle-foot articulated and nonarticulated 

hindfoot orthoses on the foot/ankle kinematics during walking in unlevel ground conditions. Orthopaedic 


• Crevoisier, XM; Kitaoka, HB; Hansen, D; Morrow, DA; Kaufman, KR: Gait abnormalities associated with 

ankle osteoarthritis. Orthopaedic Research Society 2002. 

• Erhard, L; Schultz, FM; Zobitz, ME; Zhao, CF; Amadio, PC; An, KN: Reproducible volar partial lacerations 

in flexor tendons: A new device for biomechanical studies. J Biomech 35:999-1002, 2002. 

• Erhard, L; Zobitz, ME; Zhao, CF; Amadio, PC; An, KN: Treatment of partial lacerations in flexor tendons 

by trimming. J Bone Joint Surg 84-A(6):1006-1012, 2002. 

• Fuchs, B; O’Connor, MI; Kaufman, KR; Padgett, DJ; Sim, FH: Iliofemoral arthrodesis and pseudarthrosis: 

a long-term functional outcome evaluation. Clin Orthop. 397:29-35, 2002. 

• Fujii, T; Sun, YL; An, KN; Luo, ZP: Mechanical properties of single hyaluronan molecules. J Biomech 

35:527-531, 2002. 

• Gabriel, DA; Basford, JR; An, KN: Vibratory facilitation of strength in fatigued muscle. Arch Phys Med 

Rehabil 83:1202-1205, 2002. 

• Gabriel, DA; Proctor, D; Engle, D; Sreekumaran, N; Vittone, J; An, KN: Application of the LaGrange 

polynomial in skeletal muscle fatigue analysis. Res Q Exerc Sport 73(2):168-174, 2002. 

• Guo, LY; ET AL. Modeling of manual wheelchair propulsion using optimization. American Society of 

Biomechanics 2002.

52 PUBLICATIONS 

• Guo, LY; Su, FC; An, KN: Optimum propulsion technique in different wheelchair handrim diameter. J 

Med Biol Eng 22(1):1-10, 2002. 

• Halder, AM; O’Driscoll, SW; Heers, G; Mura, N; Zobitz, ME; An, KN; Kreush-Brinker, R: Biomechanical 

comparison of effects of supraspinatus tendon detachments, tendon defects, and muscle retractions. J Bone 

Joint Surg 84(A)5:780-785, 2002. 

• Haugstvedt JR, Berger RA, Berglund LJ, Sabick MB: An Analysis of the Constraint Properties of the Distal 

Radioulnar Ligament Attachments to the Ulna. J Hand Surg 27A(1):61-67, 2002 

• Inagaki, K; O’Driscoll, SW; Neale, PG; Uchiyama, E; Morrey, BF; An, KN: Importance of a radial head 

component in Sorbie unlinked total elbow arthroplasty. Clin Orthop 400:123-131, 2002. 

• Itoi, E; Minagawa, H; Wakabayashi, I; Kobayashi, M; An, KN: Biomechanics of multidirectional instability 

of the shoulder. MB Orthop 15(5):11-16, 2002. 

• Jenkyn TR, Koopman B, Huijing P, Lieber RL, Kaufman KR: Finite element model of intramuscular pressure 

during isometric contraction of skeletal muscle. Physics in Medicine and Biology, 47:4043-4061, 2002. 

• Jenkyn, TR: Motion of the ankle complex and forefoot twist during walking and medial direction changes. 

Gait and Clinical Movement Analysis Society 2002. 

• Kamineni, S; An, KN; Neale, P; Hirahara, H; Pomianowski, S; O'Driscoll, S; Morrey, BF: Partial posteromedial 

olecranon resection - An athlete's friend and foe. Annual Meeting of the Orthopaedic Research 


• Kaufman, KR; Brey, RH; Chou, LS; Rabatin, AE; Basford, JR: Comparison of subjective and objective 

measurements of balance disorders following traumatic brain injury. Fourth World Congress of Biomechanics 


• Kaufman, KR; Davis, J; Wavering, T; Morrow, D; Lieber, RL: Intramuscular pressure is an indicator of 

muscle force. Fourth World Congress of Biomechanics 2002. 

• Ko, CC; Swift, JQ; DeLong, R; Douglas, WH; Kim, YI; An, KN; Chang, CH; Huang, HL: An intra-oral 

hydraulic system for controlled loading of dental implants. J. Biomech 35:863-869, 2002. 

• Kotajarvi BR; Basford, JR; An, KN: Upper-extremity torque production in men with paraplegia who use 

wheelchairs. Arch Phys Med Rehabil 83(4):441-446, 2002. 

• Kotajarvi, BR; Basford, JR; An, KN; Sabick, M: Does elbow angle affect wheelchair propulsion effectiveness. 

American Association of Physical Medicine and Rehabilitation 2002. 

• Kotajarvi, BR; Crevoisier, XM; Hansen, DK; Kitaoka, HB; Kaufman, KR: Effects of ankle-foot, articulated, 

and non-articulated hindfoot orthoses on foot/ankle kinematics during walking. Gait and Clinical Movement 


• Lee, SB; An, KN: Dynamic glenohumeral stability provided by three heads of the deltoid muscle. Clin Orthop 

400:40-47, 2002. 

• Luo ZP; Hsu HC; An, KN: An in vitro study of glenohumeral performance after suprascapular nerve entrapment. 

Med Sci Sports Exerc 34(4):581-586, 2002. 

• Miles, KA; Smith, J; Riemer, E; Schaefer, MP; Dahm, DL;. Kaufman, KR: Wear characteristics of common

unning shoes. American College of Sports Medicine 2002. 

PUBLICATIONS 53 

• Momose, T; Amadio, PC; Sun, YL; Zhao, CF; Zobitz, ME; Harrington, JR; An, KN: Surface modification 

of extrasynovial tendon by chemically modified hyaluronic acid coating. J Biomed Mater Res 59:219-224, 


• Momose, T; Amadio, PC; Zobitz, ME; Zhao, CF; An, KN: Effect of paratenon and repetitive motion on the 

gliding resistance of tendon of extrasynovial origin. Clin Anat. 15:199-205, 2002. 

• Morrow, DA; Matsumoto, J; Rabatin, AE; Kaufman, KR: Comparison of quantitative measures of tremor as 

predictors of surgical outcome. American Society of Biomechanics 2002. 

• Morrow, DA; Matsumoto, J; Rabatin, AE; Kaufman, KR: Comparison of quantitative measures of tremor as 

predictors of surgical outcome. Fourth World Congress of Biomechanics 2002. 

• Morrow, DA; Matsumoto, J; Rabatin, AE; Kaufman, KR: Quantitative analysis for predicting surgical reduction 

of ms tremor. Gait and Clinical Movement Analysis Society 2002. 

• Mura, N; An, KN; O'Driscoll, SW; Heers, G; Jenkyn, TR; Siaw-Meng, C: Biomechanical Effect of Infraspinatus 

Disruption and the Patch Graft Technique for Large Rotator Cuff Tears. Orthopaedic Research Society, 


• Newcomer, KL; Jacobson, TD; Gabriel, DA; Larson, DR; Brey, RH; An, KN: Muscle activation patterns in 

subjects with and without low back pain. Arch Phys Med Rehabil 83:816-821, 2002. 

• Ohtera, K; Zobitz, ME; Luo, ZP; Morrey, BF; O’Driscoll, SW; Ramin, KD; An, KN: Effect of pregnancy on 

joint contracture in the rat knee. J Appl Physiol 92:1494-1498, 2002. 

• Padgett, DJ; Kaufman, KR; Morrow, DA; Fuchs, B; Sim, FH: Oncologic Shoulder Arthrodesis: A Functional 

Assessment. Gait and Clinical Motion Analysis Society 2002. 

• Paillard, PJ; Amadio, PC; Zhao, CF; Zobitz, ME; An, KN: Gliding resistance after FDP and FDS tendon 

repair in zone II. An in vitro study. Acta Orthop Scand 73(4):465-470, 2002. 

• Paillard, PJ; Amadio, PC; Zhao, CF; Zobitz, ME; An, KN: Pulley plasty versus resection of one slip of the 

flexor digitorum superficialis after repair of both flexor tendons in Zone II. A biomechanical study. J Bone 

Joint Surg . 84-A(11):2039-45, 2002. 

• Paillard, PJ; Amadio, PC; Zhao, CF; Zobitz, ME; An, KN: Comparison of strategies to improve results after 

laceration of both flexor tendons in zone II: A biomechanical study. Annual Meeting of the Orthopaedic Research 


• Park, MJ; Cooney, WP III; Hahn, ME; Looi, KP; An, KN: The effects of dorsally angulated distal radius 

fractures on carpal kinematics. J Hand Surg 27(2):223-232, 2002. 

• Sauerbier M, Fujita M, Hahn ME, Neale PG, Berglund LJ, An KN, Berger RA: The Dynamic Radioulnar 

Convergence of the Darrach Procedure and the Ulnar Head Hemiresection Interposition Arthroplasty: A 

Biomechanical Study. J Hand Surg 27B(4):307-316, 2002 

• Sauerbier, M; Hahn, ME; Fujita, M; Neale, PG; Berglund, LJ; Berger, RA: Analysis of dynamic distal radioulnar 

convergence after ulnar head resection and endoprosthesis implantation. J Hand Surg 27A(3):425- 

434, 2002.

54 PUBLICATIONS 

• Smith, J; Kotajarvi, BR; Padgett, DJ; Eischen, JJ: Effect of Scapular Protraction and Retraction on Isometric 

Shoulder Elevation Strength. Arch Phys Med Rehab. 83(3): 2002. 

• Sun, YL; An, KN; Luo, ZP: Mechanical properties of single type II collagen molecule. Orthopaedic Research 


• Sun, YL; Luo, ZP; Fertala, A; An, KN: Direct quantification of the flexibility of type I collagen monomer. 

Biochem Biophys Res Commun 295:382-386, 2002. 

• Walker, AE; Basford, JR; Chou, LS; Brey, RH; Kaufman, KR: Center of mass gait patterns of patients with 

mild to moderate traumatic brain injury. Gait and Clinical Movement Analysis Society 2002. 

• Watanabe, K; Kitaoka, HB; Crevoisier, XM; Fuiji, T; Berglund, LJ; Kaufman, KR; An, KN: Lateral column 

lengthening for posterior tibial tendon dysfunction and flatfoot. Orthopaedic Research Society 2002. 

• Zhao, CF; Amadio, PC; Berglund, LJ; Zobitz, ME; An, KN: A new testing device for measuring gliding resistance 

and work of flexion in a digit. Annual Meeting of the Orthopaedic Research Society 2002. 

• Zhao, CF; Amadio, PC; Momose, T; Couvreur, P; Zobitz, ME; An, KN: Effect of synergistic wrist motion 

on adhesion formation after repair of partial flexor digitorum profundus tendon lacerations in a canine model 

in vivo. J Bone Joint Surg 84-A(1):78-84, 2002. 

• Zhao, CF; Amadio, PC; Momose, T; Zobitz, ME; Couvreur, P; An, KN: Remodeling of the gliding surface 

after flexor tendon repair in a canine model in vivo. J Orthop Res 20:857-862, 2002. 

• Zhao, CF; Amadio, PC; Paillard, PJ; Tanaka, T; Zobitz, ME; An, KN: Digit resistance within short term after 

FDP tendon repair in canine in vivo. Annual Conference of the American Society of Biomechanics 2002. 

• Zhao, CF; Amadio, PC; Zobitz, ME; An, KN: Resection of the flexor digitorum superficialis reduces gliding 

resistance after zone II flexor digitorum profundus repair in vitro. J Hand Surg 27(2):316-321, 2002. 

• Zhao, CF; Amadio, PC; Zobitz, ME; An, KN: Gliding characteristics after flexor tendon repair in canine in 

vivo. Annual Meeting of the Orthopaedic Research Society 2002. 

• Zhao, CF; Amadio, PC; Zobitz, ME; Momose, T; Couvreur, P; An, KN: Effect of synergistic motion on 

flexor digitorum profundus tendon excursion. Clin Orthop 396:223-230, 2002. 

• Zobitz, ME; An, KN: Stress Analysis of the Rotator Cuff: Model Development and Validation. American 

Society of Biomechanics 2002. 

• Zobitz, ME;. Rosol, M; Goessl, M;. Malyar, N; An, KN; Kantor, B: Stress Transfer to Coronary Vasa 

Vasorum after Stenting: A Finite Element Model. American Society of Biomechanics 2002.

Orthopedic Clinical Staff/Parttime 

Research 

P. C. Amadio, M.D. 

R. A. Berger, M.D., Ph.D. 

W. P. Cooney, III, M.D. 

H. B. Kitaoka, M.D. 

B. F. Morrey, M.D. 

S. W. O’Driscoll, Ph.D., M.D. 

Support Staff 

Lawrence Berglund, B.S. 

Engineer 

Ethel (Lucy) DeSerre 

Secretary 

Virginia Erbe 

Secretary 

Terri Gardner 

Secretary 

STAFF, BIOMECHANICS/MOTION ANALYSIS LABORATORIES 55 

Kai-Nan An, Ph.D., 

Biomechanics Laboratory 

Diana Hansen, B.A. 

Kinesiologist 

Christine Hughes, B.S. 

Kinesiologist 

Steven Irby, M.S. 

Engineer/ Supervisor, 

Motion Analysis Laboratory 

Barb Iverson-Literski 

Secretary 

Paul Kane 

Electronics Technician 

Brian Kotajarvi, M.S., P.T. 

Physical Therapist 

Directors 

Duane Morrow, M.S. 

Analyst/Programmer 

Patricia Neale, M.S. 

Lead Engineer/ Supervisor, 

Biomechanics Laboratory 

Denny Padgett, P.T. 

Physical Therapist 

Fredrick Schultz, M.S. 

Mechanical Design & 

Manufacturing 

Ann Rabatin, M.S. 

Kinesiologist 

Kathie Trede 

Engineer (August 2002) 

Chunfeng Zhao, M.D. 

Research Associate 

Kristin Zhao, M.A. 

Engineer 

Mark Zobitz, M.S. 

Lead Engineer/Supervisor 

(September 2002) 

Collaborators from Orthopedic 

Department 

D. J. Berry, M.D. 

A. T. Bishop, M.D. 

M. E. Bolander, M.D. 

M. E. Cabanela, M.D. 

R. H. Cofield, M.D. 

B. L. Currier, M.D. 

M. B. Dekutoski, M.D. 

A. D. Hanssen, M.D. 

Kenton R. Kaufman, Ph.D., P.E., 

Motion Analysis Laboratory 

J. A. Rand, M.D. 

M. G. Rock, M.D. 

W. J. Shaughnessy, M.D. 

F. H. Sim, M.D. 

A. A. Stans, M.D. 

M. G. Stuart, M.D. 

R. T. Trousdale, M.D. 

R. T. Turner, Ph.D. 

M. Yaszemski, M.D., Ph.D. 

Administrators 

Terry Brandt 

Ken Saling 

Collaborators from Other 

Departments 

K. Amrami, M.D. 

Radiology 

J. R. Basford, M.D., Ph.D. 

Physical Medicine & 

Rehabilitation 

M. D. Brennan, M.D. 

Endocrinology 

R. H. Brey, Ph.D. 

Vestibular/Balance 

S. E. Eckert, D.D.S. 

Prosthodontics 

R. L. Ehman, M.D. 

Diagnostic Radiology 

J. F. Greenleaf, Ph.D. 

Physiology & Biophysics 

B. D. Johnson, M.D. 

Cardiovascular Health Clinic

56 STAFF, BIOMECHANICS/MOTION ANALYSIS LABORATORIES 

Michael Joyner, M.D. 

Anesthesia Research 

B. Kantor, M.D. 

Cardiovascular Diseases 

E. R. Laskowski, M.D. 

Physical Medicine & Rehabilitation 

T. Lauder, M.D. 


J. Y. Matsumoto, M.D. 

Neurology 

K. S. Nair, M.D. 

Endocrine Research 

J. H. Noseworthy, M.D. 

Neurology 

K. D. Ramin, M.D. 

Obstetrics 

R. K Reeves, M.D. 

Physical Medicine and Rehabilitation 

E. L. Ritman, M.D. 


R. A. Robb, Ph.D. 


M. Rodriguez, M.D. 

Neurology 

H. A. Sather, D.D.S. 

Orthodontics 

M. Schaefer, M.D. 


J. Smith, M.D. 

Sports Medicine 

M. S. Stanton, M.D. 

Cardiovascular 

K. Stolp-Smith, M.D. 


J. Vittone, M.D. 

Geriatric Medicine 

B. G. Weinshenker, M.D. 

Neurology 

A. J. Windebank, M.D. 

Neurology 

Mayo Orthopedic Resident 

Joshua Baumfeld, M.D. 

Mayo Sports Medicine Fellow 

Shawn Harrington, M.D. 

Research Trainees 

Anke Ettema, M.D. 

Netherlands 

Jan Hradil 

Otrokovive, Czech Republic 

Aletta Houwink 

Netherlands 

Special Project Associates 

Lan-Yuen Guo 

Yunlin Shien, Taiwan 

Li-Cheih Kuo 

Tainan-city, Taiwan 

Visiting Scientists 

George Babis, M.D. 

Athens, Greece 

Francis Van Glabbeek, M.D. 

Antwerp, Belgium 

Visiting Clinicians 

Toshiki Akasaka, M.D. 

Morioka, Japan 

Hitoshi Sekiya, M.D. 

Tochigi, Japan 

Research Fellows 

Javier Camacho, M.D. 

Mexico, DF, Mexico 

Xavier Crevoisier, M.D. 

Lausanne, Switzerland 

Olivier Guyen, M.D. 

Lyon, France 

Kimberly Harbst, Ph.D. 

LaCrosse, Wisconsin, USA 

Hirotsune Hirahara, M.D. 

Tokyo, Japan 

Thomas Jenkyn, Ph.D. 

Toronto, Canada 

Tsuyoshi Jotoku, M.D. 

Osaka, Japan 

Srinath Kamineni, M.D. 

London, England 

Keiji Kutsumi, M.D. 

Hokkaido, Japan 

Yu-Te Lin, M.D. 

Tao-Yuan Hsien, Taiwan 

Juli Matsuoka, M.D. 

Kanagwa, Japan 

Jinrok Oh, M.D. 

Kangwon-do, Republic of Korea 

Mineo Oyama, M.D. 

Tokyo, Japan 

Murat Sutpideler, M.D. 

Izmir, Turkey 

Tatsuro Tanaka, M.D. 

Kumamoto, Japan 

Tetsu Tsubone, M.D. 

Yamaguchi, Japan 

Roger van Riet, M.D. 

Netherlands 

Hitoshi Watanabe, M.D. 

Yamanashi, Japan 

Kota Watanabe, M.D. 

Sapporo, Japan 

Chao Yang, M.D. 

Tianjin, China

STAFF, BIOMECHANICS/MOTION ANALYSIS LABORATORIES 57 

Medical School 

Students 

Aqiyla Muhammad 

Talya Williams 

Summer Interns 

Matt Urban 

Melissa Hasenbank 

Jeffifer Berumen 

Rose Riemer 

Christine Keenan 

Kalpesh Joshi 

Miranda Shaw 

Kavitha Pundi 

Kelvin Lee 

High School Interns 

Gabriel Marcus 

Richard Shin 

Guthrie Williams 

Christopher Markos 

Michael Kavros 

Andrew Huettner 

Ph.D. Students 

Karen Boyd 

Kee-Won Lee 

Pablo Whaley 

Heather Argadine 

Kara Bliley

58 ACKNOWLEDGEMENTS 

The following financial support of our Laboratories is gratefully acknowledged: 

Title/Principal Investigator Funding Organization 

Repair and Rehabilitation of Flexor Tendon Injury NIAMS 

PI: P. C. Amadio, M.D. 

Characterization of the Skeletal Muscle by MR Elastography NICHD 

PI: K. N. An, Ph.D. 

Biomechanics of the Upper Extremity in Wheelchair Propulsion NICHD 

PI: K. N. An, Ph.D. 

Clinical Role of Wrist Joint Mechanoreceptors NIAMS 

PI: R.A. Berger, M.D. 

Ankle Joint Stability AOFAS 

PI: K.R. Kaufman, Ph.D. 

Microsensor for Intramuscular Pressure Measurement NICHD 

PI: K. R. Kaufman, Ph.D. 

Logic Controlled Electromechanical Free Knee Orthosis NICHD 


Aerobic Exercise Intervention for Knee Osteoarthritis NIH 


Foot and Ankle Biomechanics with Orthosis Ambulation NIAMS 

PI: H. B. Kitaoka, M.D. 

Kinematics and Stability of the Arch of the Foot NIAMS 

PI: H. B. Kitaoka 

The Effect of Foot Orthoses on Posterior Tibialis, Tibialis Anterior, OAFAS #3 

and Peroneus Longus Muscle Electromyographic Activity During 

Walking and Running 

PI: J Smith, M.D. and M. P. Schaefer

ACKNOWLEDGEMENTS 59 

The following institutions, foundations, and corporations are gratefully acknowledged for their 

contribution of research funding, equipment, or medical devices for research projects: 

Anderson’s Wheelchairs 

Ascension Technology 

Avanta Orthopedics 

Ceramconcepts 

CPI/Guidant 

Howmedica 

Inertia Dynamics 

Innovative Sports Training 

Langeloth Foundation 

Mayo Foundation 

Medical Metrics 

Minnesota Arthritis Foundation 

National Arthritis Foundation 

National Institutes of Health 

Orthopaedic Research and Education Foundation 

Prosthetic Laboratories of Rochester 

Prosthetic Orthotic Center 

Smith & Nephew Richards, Inc. 

Whitaker Foundation 

Zimmer

Report of Activities 2002 - Research - Mayo Clinic

Create successful ePaper yourself

Delete template?

Save as template?