Peripheral arterial disease affects kinematics
Rolando Celis, MD,a,bIraklis I. Pipinos, MD,a,bMelissa M. Scott-Pandorf, MS,cSara A. Myers, MS,c
Nicholas Stergiou, PhD,cand Jason M. Johanning, MD,a,bOmaha, Neb
Objective: Claudication is the most common manifestation of peripheral arterial disease (PAD) producing significant
ambulatory compromise. The purpose of this study was to use advanced biomechanical analysis to characterize the
kinematic ambulatory pattern of claudicating patients. We hypothesized that compared with control subjects, claudicat-
ing patients have altered kinematic gait patterns that can be fully characterized utilizing advanced biomechanical analysis.
Methods: The study examined fourteen PAD patients (age: 58 ? 3.4 years; weight: 80.99 ? 15.64 kg) with clinically
diagnosed femoro-popliteal occlusive disease (Ankle Brachial Index (ABI): 0.56 ? 0.03, range 0.45 to 0.65) and five
healthy controls (age: 53 ? 3.4 years; weight: 87.38 ? 12.75 kg; ABI > 1). Kinematic parameters (hip, knee, and ankle
joint angles in the sagittal plane) were evaluated during gait in patients before and after the onset of claudication pain and
compared with healthy controls. Joint angles were calculated during stance time. Dependent variables were assessed
(maximum and minimum flexion and extension angles and ranges of motion) and mean ensemble curves were generated.
Time to occurrence of the discrete variables was also identified.
Results: Significantly greater ankle plantar flexion in early stance and ankle range of motion during stance was observed
in PAD patients (P < .05). Time to maximum ankle plantar flexion was shorter and time to maximum ankle dorsiflexion
was longer in PAD patients (P < .05). These differences were noted when comparing PAD patients prior to and after the
onset of claudication with healthy controls. The analysis of the kinematic parameters of the knee and the hip joints
revealed no significant differences between PAD patients and controls.
Conclusion: PAD patients with claudication demonstrate significant gait alterations in the ankle joint that are present
prior to the onset of claudication pain. In contrast, the joint motion of the hip and knee did not differ in PAD patients
when compared with controls. Further research is needed to verify our findings and assess the impact of more proximal
disease in PAD patients as well as the effect of revascularization on joint kinematics. (J Vasc Surg 2008;??:???.)
Peripheral arterial disease (PAD) of the lower extremi-
ties is a manifestation of atherosclerosis, affecting 20% to
30% of older patients in general medical practices1,2and up
to 12 million people in the United States population.3,4
Intermittent claudication (IC), defined as lower extremity
pain that causes the patient to stop walking and resolves
within few minutes of rest, is considered the classic symp-
tom of PAD. Recently IC has been identified as an ambu-
latory disorder. This is supported by studies demonstrating
PAD patients to have lower daily physical activity,5reduced
strength in lower extremities,6worse self-perceived ambu-
latory function,7lower health related quality of life,8im-
paired balance, and higher prevalence of falling.9
The characterization of the gait of patients with IC
until recently has been limited to the measurement of
simple temporal and spatial parameters of the patients’
walking performance. Such evaluations have documented
that PAD patients have decreased step length, cadence,
walking speed and increased stance time.10,11These mea-
sures suggest the presence of ambulation abnormalities in
claudicating patients; however they provide limited insight
into the specific site and mechanisms producing the abnor-
Biomechanical analysis, in contrast to the previously
used rudimentary measurements mentioned above, repre-
sents an important diagnostic tool with the ability to pro-
vide detailed and accurate quantitative gait analysis. Fur-
thermore, biomechanical evaluation is common practice is
several other medical domains (ie, orthopedics, pediatrics,
neurology, etc.) and has been useful in both research and
clinical settings for directing treatment in varying patholo-
gies as well as in outcome evaluation of the results of such
treatments.12-15In contrast to the progress made in other
fields where advanced biomechanics has been imple-
mented, very little has been done to provide an in depth
analysis of the underlying biomechanical gait abnormalities
produced by PAD.16,17
The purpose of the current study was to determine the
gait of patients with symptomatic PAD before and after the
onset of claudication utilizing advanced biomechanical
analysis. We hypothesized that the lower extremities of
PAD patients have altered joint displacement compared
with control subjects both before and after the onset of
claudication, and that biomechanical kinematic analysis
represents a diagnostic tool with appropriate sensitivity to
detect subtle differences in a subject’s gait. The current
From the University of Nebraska Medical Center,athe Veterans Affairs
Medical Center of Nebraska,band the University of Nebraska at Omaha.c
Grant funding for the current study was provided by the American Geriatrics
Society’s Hartford Foundation Dennis W. Jahnigen Award, the Nebraska
Research Initiative and the University of Community on Research and
Creative Activity at University of Nebraska at Omaha.
Competition of interest: none.
Reprint requests: Jason M. Johanning, MD, University of Nebraska Medical
Center, Section of Vascular Surgery, Omaha, NE 68198-3280 (e-mail:
Copyright © 2008 Published by Elsevier Inc. on behalf of The Society for
ARTICLE IN PRESS
kinematic study, which focuses on the lower extremity
joints’ angular displacement independently of the generat-
ing forces, complements the kinetic analysis previously
described by our group,18which evaluated the forces ex-
erted by the subjects weight-bearing limb on the ground.
Our work seeks to further enhance our understanding of
the abnormal gait in subjects with PAD, thus providing the
foundation for the development of new rehabilitation strat-
egies and the quantification of treatment outcomes for
patients with symptomatic PAD.
tained prior to initiation of the study and all subjects
provided informed consent. Patients with clinically diag-
nosed PAD presenting with classic symptomatic claudica-
tion were recruited from our vascular surgery clinics. Se-
lected patients were free of any associated co-morbidities
limiting or altering their gait. Specifically, subjects were
excluded if they had recent myocardial infarction or
ambulation-limiting heart failure, angina, or pulmonary
disease. Additionally, subjects were excluded if they had
gait altering neurological or musculoskeletal disease such as
paresis, sciatica, arthritis, diabetic neuropathy, or arthrop-
athy. History and physical examination of the subjects
evaluated was performed by board certified vascular sur-
geons (J.J., I.P.). Lower extremity arterial disease was
verified by classic clinical symptoms confirmed utilizing
noninvasive testing (ankle-brachial indexes ? 0.9) and the
level of disease identified with the aid of noninvasive vascu-
lar examination complemented by computerized tomogra-
phy, magnetic resonance, or invasive angiography. Based
on this assessment, limbs with occlusive disease and typical
Rose claudication symptoms19were established as “claudi-
cating limbs” and selected for biomechanical analysis.
Control subjects were recruited from the community.
Detailed history and physical examination performed by
vascular surgeons documented absence of PAD and co-
morbidities as described for PAD patients. Absence of PAD
was confirmed by noninvasive testing (ankle-brachial in-
dexes) and absence of pain during ambulation. Each leg of
these individuals was used as “control limb.” To eliminate
variability in gait due to shoes, all subjects wore the same
standard laboratory shoes (Cross Trekkers, Payless Shoes,
Lower extremity kinematics. Upon arrival in the lab-
oratory, patients were prepared for data collection. Height,
weight, body mass index, age, and anthropometric mea-
surements were obtained. Reflective markers were placed at
specific anatomical locations of each subject’s lower limb
utilizing the systems used by Vaughan20and Nigg21and as
described in Fig 1. The subjects’ lower extremity three-
dimensional kinematics was acquired with a high speed
analog video Peak Performance system at 60 Hz (Peak
Performance Technologies, Englewood, Colo). Marker
identification was conducted using the Peak Motus (Vicon-
Peak Performance Technologies, Inc., Centennial, Colo)
software. The exported marker data was scaled and smoothed
using a Butterworth low-pass filter with a selective cut-off
algorithm according to Jackson.22The cut-off values used
were seven to 14 Hz. This analysis was performed using
custom software in Matlab (Mathworks Inc., Natick,
Mass), where the exported data was also converted to unit
vectors for each local reference frame. Anthropometric
measurements were combined with three-dimensional
marker data from the anatomical position calibration trial
(see below) to provide positions of the joint centers and
define anatomical axes of joint rotations.20The positions of
the reflective markers during the movement provided the
three-dimensional joint angles and were determined
through triangulation of the position of the markers.
Prior to the walking trials, patients stood in the calibra-
tion device for five seconds while kinematic video was
calibration trial provided an anatomical reference position.
The calculation of three-dimensional lower extremity seg-
ment orientations and relevant joint angles was referred to
this position. Kinematic data was collected during the
stance phase of walking (from heel contact to toe off).
Initially five walking trials were acquired from each PAD
subject without pain present and represented the “pain
free” condition (PAD-PF). During this condition the pa-
before and between trials to ensure pain free measure-
ments. After PAD patients completed the pain free walking
trials, claudication pain was induced. This was accom-
Fig 1. Anatomic location of markers in lower extremities. The
position and trajectories of these markers were captured by our
cameras. The analysis and process of this data originates curves
corresponding to relative joint angles during stance time of the
JOURNAL OF VASCULAR SURGERY
2 Celis et al
ARTICLE IN PRESS
plished by having patients walk on a treadmill at a 10%
grade at 0.67 m/s until claudication was induced (usually
the treadmill) and then for approximately 45 additional
seconds. Patients returned to the walkway immediately
where five more walking trials were performed without any
resting between the trials. Claudication pain was present
throughout these trials and represented the claudication or
“pain” condition of the PAD patients (PAD-P). Data from
the healthy controls was collected following the protocol
used to obtain the pain free data from the PAD patients,
with claudication data not obtained due to lack of PAD in
Data analysis. Joint angles from the hip, knee, and
ankle were analyzed for the two conditions of the PAD
patients and for the controls. Dependent variables calcu-
lated were the range of motion, the maximum, and the
minimum of the joints’ flexion and extension angles. All
kinematic parameter data files were normalized to 100
points for the stance phase using a cubic spline routine to
enable mean ensemble curves to be derived for each condi-
tion of each subject. All normalization occurred after max-
imums and minimums were determined to ensure that the
normalization did not distort these values.23,24
Statistical analysis. Statistical analysis was performed
using SPSS (version 15; SPSS Inc, Chicago, Ill). Subject
and group means were calculated and inferential statistics
were used to compare the different groups. Independent
t-tests were used to compare mean values of PAD patients
(PAD-PF and PAD-P) with healthy controls. Paired t-tests
were used to compare PAD-PF with the PAD-P condition.
Significance was set at 0.05.
Demographics. Fourteen PAD patients (age: 58 ?
3.4 years; weight: 80.99 ? 15.64 kg; height: 172.12 ?
6.78 cm) with clinically diagnosed femoro-popliteal occlu-
sive disease (Ankle Brachial Index (ABI): 0.56 ? 0.03,
range, 0.45 to 0.65) were recruited. All patients had classic
Rose claudication or Rutherford category 2 symptoms.25
Eighty percent of the patients were hypertensive, 70%
were smokers, 60% had dyslipidemia, and 30% were
obese (BMI ? 30 kg/m2). All patients were treatment
naive. From these 14 patients, a total of 20 symptomatic
PAD legs were included for kinematic analysis. Five control
subjects with absence of PAD and absence of any ambula-
tory disability (age: 53 ? 3.4 years; weight: 87.38 ? 12.75
kg; height: 178.78 ? 4.32 cm and ABI ? 1.00 or greater)
hypertension. From these subjects both legs were utilized
providing a total of 10 legs. Body mass index values were
28.5 ? 0.98 for PAD patients and 27.3 ? 1.5 for control
subjects. Subjects were well-matched regarding age and
body mass index with no significant differences noted be-
tween groups (P ? .05).
Kinematic analysis. At the level of the ankle, signifi-
cant differences were noted between PAD patients and
control subjects (Table I). Increased minimum (negative
values) ankle plantar flexion during the initial stance phase
and increased range of motion (ROM) throughout the
stance phase was observed in claudicants when compared
with controls in both PAD-PF and PAD-P conditions
(Table I; Fig 2). Since no differences were found for
maximum (positive values) ankle dorsiflexion (Table I),
these results are reflected as a deeper “valley” on the ankle
mean ensemble curve for PAD patients when compared
with the control subjects (Fig 2). The time to minimum
plantar flexion and maximal dorsiflexion of the ankle joint
during the stance phase was significantly altered when
comparing PAD-P with control subjects. The PAD-P pa-
tient reached minimum ankle plantar flexion faster and
maximum dorsiflexion later than the control subject (Table
II). When analyzing the effect of claudication on joint
motion at the knee and hip, no significant differences were
noted in joint angles.
When analyzing the effect of claudication pain by com-
paring the PAD-PF with PAD-P conditions, there were no
significant differences noted in joint motion at each joint
Table I. Group means of joint angle parameters in controls and peripheral arterial disease patients both before and after
the onset of claudication
Control (degrees)PAD-PF (degrees)
22.824 ? 3.334
?20.225 ? 3.605
43.050 ? 1.915
19.021 ? 4.849
1.247 ? 4.246
17.773 ? 3.649
?3.458 ? 3.427
14.500 ? 2.407
17.962 ? 4.531
23.094 ? 4.751
?18.773 ? 6.427
41.870 ? 6.542
18.065 ? 6.261
1.102 ? 5.504
16.962 ? 5.493
?7.596 ? 2.520
16.539 ? 4.767
24.1371 ? 4.934
22.9388 ? 5.009
?19.1206 ? 5.390
42.060 ? 5.846
17.525 ? 6.741
0.842 ? 5.677
16.684 ? 4.688
?8.368 ? 2.764
16.254 ? 4.470
24.621 ? 4.363
PAD, peripheral arterial disease; PAD-PF, pain free condition PAD patient; PAD-P, pain condition PAD patient; NS, statistically non significant; ROM, range
†Control vs. PAD-PF.
‡Control vs. PAD-P.
?PAD-PF vs. PAD-P.
JOURNAL OF VASCULAR SURGERY
Volume ??, Number ?
Celis et al 3
ARTICLE IN PRESS
level or in the timing of specific points within the gait cycle.
These results are also reflected in the mean ensemble curves
since the lines for PAD-P and PAD-PF are overlapping
throughout stance (Fig 2).
Our data demonstrated that patients with clinically
diagnosed femoro-popliteal PAD have significant ankle
Fig 2. Average curves for hip, knee, and ankle joints representing healthy control patients and PAD-PF and PAD-P
conditions. PAD, peripheral arterial disease; PF, pain-free; P, pain.
Table II. Group means of time to maximal flexion and extension of the joints in controls and peripheral arterial disease
patients both before and after the onset of claudication
Control (degrees) PAD-PF (degrees)
Hip flex time
Hip ext time
Knee flex time
Knee ext time
Ankle plantarflex time
Ankle dorsiflex time
13.919 ? 6.761
86.439 ? 1.349
28.447 ? 3.783
68.87 ? 1.949
17.087 ? 1.800
72.788 ? 11.327
13.523 ? 7.627
86.180 ? 3.292
26.838 ? 6.065
68.268 ? 5.186
16.224 ? 2.526
78.264 ? 7.134
12.209 ? 5.801
87.241 ? 2.244
26.851 ? 6.383
67.296 ? 5.960
15.710 ? 1.233
79.966 ? 5.815
PAD, peripheral arterial disease; PAD-PF, pain free PAD patient; PAD-P, pain condition PAD patient; NS, statistically non significant; ROM, range of motion.
†Control vs. PAD-PF.
‡Control vs. PAD-P.
?PAD-PF vs. PAD-P.
JOURNAL OF VASCULAR SURGERY
4 Celis et al
ARTICLE IN PRESS
motion alterations with abnormal ankle joint kinematics.
During the stance phase of the gait cycle, the PAD patients
demonstrated rapid foot plantar flexion after initial heel
strike coupled with a significant increase in ankle plantar
flexion. The increase in ankle plantar flexion with subse-
quent normal maximal dorsiflexion resulted in PAD pa-
tients having a significantly increased ankle range of mo-
tion. This phenomenon was present both before and after
the onset of claudication pain. Based on our kinematic
analysis, PAD patients have what appears to be “foot drop”
upon heel touchdown. The etiology of this finding is
currently unknown but is likely secondary to nerve damage
and muscle weakness from chronic ischemia.26-28This
could result in poor eccentric motor control from the foot
dorsiflexors (anterior and lateral compartment leg muscles)
in combination with suboptimal plantar flexor function
(posterior compartment muscles). Taken together, these
findings represent either a compensatory mechanism to
maintain stability due to inherent neuromuscular weakness
of the lower limb or alternatively an adaptation to altered
neuromuscular function due to PAD. Consistent with any
dysfunctional gait, PAD patients have a deviation from
requirement and energy cost.29,30Future research should
expand on the analysis of joint moments at the level of the
ankle to confirm the location of motor dysfunction and the
contribution of nerve dysfunction to the gait abnormality.
In contrast to the findings at the ankle in our current
knee in both flexion and extension in both conditions for
PAD and control patients. This result could be due to the
fact that the patients in the current study had clinically
diagnosed femoro-popliteal occlusive disease with classic
Rose claudication with absence of thigh and buttock clau-
dication. One could argue that only the lower leg muscu-
lature was involved in the ischemic process and therefore
the proximal muscles were spared. Further studies will be
necessary to delineate the full spectrum of ambulatory
compromise in patients with isolated aorto-iliac occlusive
disease and multi-level disease.
An important finding in our study is that PAD patients
had evidence of significant ambulatory abnormalities even
also a further alteration in gait function with claudication
pain as the patients in our series showed a trend for in-
creased differences of gait parameters compared with con-
trols such as ankle plantar flexion and range of motion after
onset of claudication pain but the differences did not reach
statistical significance. These results confirm unequivocally
the presence of a significantly altered and dysfunctional gait
prior to the onset of claudication pain despite what appears
we believe reflect a baseline lower extremity dysfunction in
PAD patients with origins at the cellular level.31,32The
abnormalities contributing to the baseline gait dysfunction
include axonal nerve loss26,27and mitochondrial dysfunc-
tion,28,33-37both of which could account for the under-
lying gait dysfunction found in PAD patients from the
first step during ambulation. Ischemia superimposed on
underlying neuromuscular dysfunction would then re-
sult in variably worsening gait as seen in our previous
Previous studies have reported kinematic analysis of
elderly individuals, showing deceased ankle plantar flexion
kinematic analysis of our control patients are similar to
those reported on the literature on healthy elderly subjects.
In contrast, little data exists to document the kinematic
analysis of patients with PAD. Previous literature utilized
simple visual observations in PAD patients to analyze dif-
ferences in gait parameters with conflicting results. A recent
study by Crowther et al has documented the effect of PAD
on gait biomechanical parameters, only before the onset of
claudication pain.17Although difficult to compare to our
study due to differing methodologies, this study also found
differences at the ankle. In contrast to our results, they
found differences in the knee ROM and hip extension. Our
methodology only included patients with clinically diag-
nosed femoro-popliteal disease and focused the analysis of
the joints to the stance phase of the gait cycle. Crowther et
al did not specify a level of disease among the patients and
of motion. Another factor contributing to the precision of
our results is the capability of our lab to provide a three-
dimensional analysis given the number of cameras used;
Crowther et al, in contrast, utilized a two-dimensional
kinematic analysis of the sagittal plane, which is vulnerable
to perspective error.39,40When comparing the ankle curves
of both studies, Crowther et al demonstrated an increased
plantar flexion in the control patients in the swing phase
only. Our results, similar to Crowther et al, showed in-
creased plantar flexion of the controls at the end of stance
phase. In contrast to Crowther’s report, we detected an
increased plantar flexion of the PAD patients in the stance
phase. Both studies document similar alterations at the
ankle level with the differences secondary to methodology
and length of gait cycle analyzed. Both studies confirm,
however, the significant ankle dysfunction in the PAD
patient and provide a start point to elucidate the underlying
joint biomechanical abnormalities found in patients with
Kinematic gait analysis demonstrates that patients with
clinically diagnosed femoropopliteal disease have altered
ankle plantar flexion present before and after the onset of
claudication pain compared with control subjects. Our
data, in conjunction with previous biomechanical analysis,
confirm that patients with symptomatic PAD have an un-
derlying ambulatory abnormality present even prior to
onset of claudication pain. Further biomechanical evalua-
tion of PAD patients should focus on the impact of disease
level on gait dysfunction and evaluation of joint moments
and powers to identify the specific muscular deficits pro-
duced by PAD. Our results provide evidence for the utili-
zation of advanced biomechanical analysis to identify the
JOURNAL OF VASCULAR SURGERY
Volume ??, Number ?
Celis et al 5
ARTICLE IN PRESS
uniquegaitabnormalitiesinPADpatients,thusprovidinga Download full-text
powerful research tool for objective analysis of medical and
surgical PAD treatment and rehabilitation therapy.
Conception and design: IP, MSP, NS, JJ
Analysis and interpretation: RC, IP, MSP, SM, NS, JJ
Data collection: IP, MSP, SM, JJ
Writing the article: RC, IP, MSP, NS, JJ
Critical revision of the article: IP, SM, NS, JJ
Final approval of the article: RC, IP, MSP, SM, NS, JJ
Statistical analysis: MSP, SM, JJ
Obtained funding: IP, NS, JJ
Overall responsibility: IP, NS, JJ
1. Hirsch AT, Criqui MH, Treat-Jacobson D, Regensteiner JG, Creager
MA, Olin JW, et al. Peripheral arterial disease detection, awareness, and
treatment in primary care. JAMA 2001;286:1317-24.
2. McDermott MM, Kerwin DR, Liu K, Martin GJ, O’Brien E, Kaplan H,
Greenland P. Prevalence and significance of unrecognized lower ex-
tremity peripheral arterial disease in general medicine practice. J Gen
Intern Med 2001;16:384-90.
3. Becker GJ, McClenny TE, Kovacs ME, Raabe RD, Katzen BT. The
importance of increasing public and physiscian awareness of peripheral
arterial disease. J Vasc Interv Radiol 2002;13:7-11.
4. Nehler M, McDermott M, Treat-Jacobson D, Chetter I, Regensteiner
J. Functional outcomes and quality of life in peripheral arterial disease:
Current status. Vasc Med 2003;8:115-26.
5. Sieminski DJ, Gardner AW. The relationship between daily physical
6. Scott-Okafor HR, Silver K, Parker J, Gardner AW. Lower extremity
strength deficits in peripheral arterial occlusive disease patents with
intermittent claudication. Angiology 2001;52:7-14.
7. Regensteiner JG, Steiner JF, Panzer RJ, Hiatt WR. Evaluation of
walking impairment by questionnaire in patients with peripheral artery
disease. J Vasc Med Biol 1990;2:142-52.
8. Feinglass J, McCarthy WJ, Slavensky R, Manheim LM, Martin GJ.
Effect of lower extremity blood pressure on physical functioning in
patients who have intermittent claudication. The Chicago Claudication
Outcomes Research Group. J Vasc Surg 1996;24:503-11.
9. Gardner AW, Montgomery PS. Impaired balance and higher prevalence
of falls in subjects with intermittent claudication. J Gerontol A Biol Sci
Med Sci 2001;56:454-8.
10. Scherer SA, Bainbridge JS, Hiatt WR, Regensteiner JG. Gait character-
istics of patients with claudication. Arch Phys Med Rehabil 1998;79:
11. McCully K, Leiper C, Sanders T, Griffin E. The effects of peripheral
vascular disease on gait. J Gerontol 1999;54A:291-4.
12. Kay RM, Dennis S, Rethlefsen S, Reynolds RA, Skaggs DL, Tolo VT.
The effect of preoperative gait analysis on orthopaedic decision making.
Clin Orthop Relat Res 2000:217-22.
13. Ounpuu S, Gage JR, Davis RB. Three-dimensional lower extremity
joint kinetics in normal pediatric gait. J Pediatr Orthop 1991;11:341-9.
14. McClelland JA, Webster KE, Feller JA. Gait analysis of patients follow-
ing total knee replacement: a systematic review. Knee 2007;14:253-63.
15. Lee EH, Goh JC, Bose K. Value of gait analysis in the assessment of
surgery in cerebral palsy. Arch Phys Med Rehabil 1992;73:642-6.
16. Gardner AW, Montgomery PS. The relationship between history of
falling and physical function in subjects with peripheral arterial disease.
Vasc Med 2001;6:223-7.
17. Crowther RG, Spinks WL, Leicht AS, Quigley F, Golledge J. Relation-
ship between temporal-spatial gait parameters, gait kinematics, walking
performance, exercise capacity, and physical activity level in peripheral
arterial disease. J Vasc Surg 2007;45:1172-8.
18. Scott-Pandorf MM, Stergiou N, Johanning JM, Robinson L, Lynch
TG, Pipinos II. Peripheral arterial disease affects ground reaction forces
during walking. J Vasc Surg 2007;46:491-9.
19. Rose G. The diagnosis of ischaemic heart pain and intermittent claudi-
cation if field surveys. Bull World Health Organ 1962;27:645-58.
edition. Cape Town, South Africa: Kiboho; 1999. p. 90-106.
21. Nigg BM, Cole G, Nachbauer W. Effects of arch height of the foot on
angular motion of the lower extremities in running. J Biomech 1993;
22. Jackson KM. Fitting of mathematical functions to biomechanical data.
IEEE Trans. Biomed Eng 1979;26:122-4.
23. Woltring HJ. On optimal smoothing and derivative estimation from
noisy displacement data in biomechanics. Hum Mov Sci 1985;4:
24. Giakas G. Power spectrum analysis and filtering. In: Stergiou N, editor.
Innovative analyses in human movement. Champaign IL: Human Ki-
netics Inc; 2004. p. 223-50.
25. Rutherford RB, Baker JD, Ernst C, Johnston KW, Porter JM, Ahn S,
Jones DN. Recommended standards for reports dealing with lower
extremity ischemia: revised version. J Vasc Surg 1997;26:517-38.
26. Koopman JP, de Vries AC, de Weerd AW. Neuromuscular disorders in
patients with intermittent claudication. Eur J Surg 1996;162:443-6.
27. Laghi Pasini F, Pastorelli M, Beermann U, de Candia S, Gallo S, Blardi
P, Di Perri T. Peripheral neuropathy associated with ischemic vascular
disease of the lower limbs. Angiology 1996;47:569-77.
28. Pipinos II, Judge AR, Selsby JT, Zhu Z, Swanson SA, Nella AA, Dodd
SL. The myopathy of peripheral arterial occlusive disease: part 1.
Functional and histomorphological changes and evidence for mito-
chondrial dysfunction. Vasc Endovascular Surg 2007;41:481-9.
29. Womack CJ, Sieminski DJ, Katzel LI, Yataco A, Gardner AW. Im-
proved walking economy in patients with peripheral arterial occlusive
disease. Med Sci Sports Exerc 1997;29:1286-90.
30. Kuo AD. The six determinants of gait and the inverted pendulum
analogy: A dynamic walking perspective. Hum Mov Sci 2007;26:
31. Brass EP, Hiatt WR. Acquired skeletal muscle metabolic myopathy in
atherosclerotic peripheral arterial disease. Vasc Med 2000;5:55-9.
32. Marbini A, Gemignani F, Scoditti U, Rustichelli P, Bragaglia MM,
Govoni E. Abnormal muscle mitochondria in ischemic claudication.
Acta Neurol Belg 1986;86:304-10.
33. Pipinos II, Judge AR, Zhu Z, Selsby JT, Swanson SA, Johanning JM, et
al. Mitochondrial defects and oxidative damage in patients with periph-
eral arterial disease. Free Radic Biol Med 2006;41:262-9.
34. Pipinos II, Shepard AD, Anagnostopoulos PV, Katsamouris A, Boska
MD. Phosphorus 31 nuclear magnetic resonance spectroscopy suggests
a mitochondrial defect in claudicating skeletal muscle. J Vasc Surg
35. Kemp GJ. Mitochondrial dysfunction in chronic ischemia and periph-
eral vascular disease. Mitochondrion 2004;4:629-40.
36. Makris KI, Nella AA, Zhu Z, Swanson SA, Casale GP, Gutti TL, et al.
Mitochondriopathy of peripheral arterial disease. Vascular 2007;15:
A, Todor A, et al. Abnormal mitochondrial respiration in skeletal
muscle in patients with peripheral arterial disease. J Vasc Surg 2003;38:
38. Judge JO, Ounpuu S, Davis RB. Effects of age on the biomechanics and
physiology of gait. Clin Geriatr Med 1996;12:659-78.
39. Areblad M, Nigg BM, Ekstrand J, Olsson KO, Ekström H. Three-
dimensional measurement of rearfoot motion during running. J Bio-
40. Allard P, Capozzo A, Lundberg A, Vaughan C, editors. Three-
dimensional analysis of human locomotion. West Sussex, United
Kingdom: John Wiley and Sons, Ltd.; 1997.
Submitted Apr 26, 2008; accepted Aug 9, 2008.
JOURNAL OF VASCULAR SURGERY
6 Celis et al
ARTICLE IN PRESS