McGrae McDermott M, Greenland P, Liu K, Guralnik JM, Criqui MH, Dolan NC, Chan C, Celic L, Pearce WH, Schneider JR, Sharma L, Clark E, Gibson D, Martin GJ. Leg Symptoms in Peripheral Arterial DiseaseAssociated Clinical Characteristics and Functional Impairment. JAMA. 2001;286(13):1599-1606. doi:10.1001/jama.286.13.1599
Author Affiliations: Departments of Medicine (Drs McDermott, Greenland, Dolan, Sharma, and Martin, and Ms Celic and Mr Gibson), Preventive Medicine (Drs McDermott, Greenland, and Liu, and Ms Chan), and Surgery (Drs Pearce and Schneider), Northwestern University Medical School, Chicago, Ill; Division of Vascular Surgery, Department of Surgery, Evanston/Northwestern Hospital, Evanston, Ill (Dr Schneider); Epidemiology, Demography, and Biometry Program, National Institute on Aging, Bethesda, Md (Dr Guralnik); Department of Family and Preventive Medicine, University of California, San Diego (Dr Criqui); and Division of Vascular Surgery, Department of Surgery, Catholic Health Partners, Chicago, Ill (Dr Clark).
Context Persons with lower-extremity peripheral arterial disease (PAD) are often
asymptomatic or have leg symptoms other than intermittent claudication (IC).
Objective To identify clinical characteristics and functional limitations associated
with a broad range of leg symptoms identified among patients with PAD.
Design, Setting, and Participants Cross-sectional study of 460 men and women with PAD and 130 without
PAD, who were identified consecutively, conducted between October 1998 and
January 2000 at 3 Chicago-area medical centers.
Main Outcome Measures Ankle-brachial index score of less than 0.90; scores from 6-minute walk,
accelerometer-measured physical activity over 7 days, repeated chair raises,
standing balance (full tandem stand), 4-m walking velocity, San Diego claudication
questionnaire, Geriatric Depression Score Short-Form, and the Walking Impairment
Results All groups with PAD had poorer functioning than participants without
PAD. The following values are for patients without IC vs those with IC. Participants
in the group with leg pain on exertion and rest (n = 88) had a higher (poorer)
score for neuropathy (5.6 vs 3.5; P<.001), prevalence
of diabetes mellitus (48.9% vs 26.7%; P<.001),
and spinal stenosis (20.8% vs 7.2%; P = .002). The
atypical exertional leg pain/carry on group (exertional leg pain other than
IC associated with walking through leg pain [n = 41]) and the atypical exertional
leg pain/stop group (exertional leg pain other than IC that causes one to
stop walking [n = 90]) had better functioning than the IC group. The group
without exertional leg pain/inactive (no exertional leg pain in individual
who walks ≤6 blocks per week [n = 28]) and the leg pain on exertion and
rest group had poorer functioning than those with IC. Adjusting for age, sex,
race, and comorbidities and compared with IC, participants with atypical exertional
leg pain/carry on achieved a greater distance on the 6-minute walk (404.3
vs 328.5 m; P<.001) and were less likely to stop
during the 6-minute walk (6.8% vs 36%; P = .002).
The group with pain on exertion and rest had a slower time for completing
5 chair raises (13.5 vs 11.9 seconds; P = .009),
completed the tandem stand less frequently (37.5% vs 60.0%; P = .004), and had a slower 4-m walking velocity (0.80 vs 0.90 m/s; P<.001).
Conclusions There is a wide range of leg symptoms in persons with PAD beyond that
of classic IC. Comorbid disease may contribute to these symptoms in PAD. Functional
impairments are found in every PAD symptom group, and the degree of functional
limitation varies depending on the type of leg symptom.
Intermittent claudication (IC) has long been considered the most classic
manifestation of peripheral arterial disease (PAD).1
However, when noninvasive testing with the ankle-brachial index (ABI) is used
to diagnose PAD objectively, many people with PAD have latent disease or are
asymptomatic (ie, have no exertional leg symptoms), while others have symptoms
other than classic IC.2- 4
Noninvasive testing with the ABI is the clinical standard for diagnosing PAD.
The low sensitivity of IC for PAD may contribute to underrecognition of PAD
in general medicine practices.5,6
With its associated functional limitations, IC is appreciated as a serious
clinical symptom of PAD. However, it is unclear whether other forms of PAD,
with and without leg symptoms, cause functional limitations. Comorbidities
such as hip and knee arthritis, back disease, or neuropathy might alter the
nature of leg symptoms experienced in PAD. Characteristics such as depression
might affect PAD patients' awareness or description of leg pain.
The purposes of this study were to document the spectrum of leg symptoms
reported by men and women with PAD and to identify clinical characteristics
and functional limitations associated with leg symptom categories identified
among patients with PAD. Our results should sensitize clinicians to the common
occurrence of leg symptoms other than IC in patients with PAD and highlight
that the type of symptoms reported determines the degree of associated lower-extremity
impairment. Our findings underscore that PAD-associated leg pain may be blurred
by the contribution of symptoms produced by comorbid conditions.
The protocol was approved by the Northwestern University Medical School
and Catholic Health Partners Hospital institutional review boards. Participants
gave informed consent. PAD participants were identified consecutively from
patients diagnosed with PAD in 3 Chicago-area noninvasive vascular laboratories.
Non-PAD participants were identified consecutively from patents with appointments
in a large general medicine practice. In addition, individuals with PAD identified
in the general medicine practice (n = 13) were included in the PAD group.
Individuals were contacted and invited to return to the medical center for
a study visit between October 1998 and January 2000.
Presence of PAD was established using objective, noninvasive testing.7- 11
The current standard for diagnosing PAD is defined as an ABI of less than
0.90. Absence of PAD was defined as an ABI of 0.90 or higher and of 1.50 or
less.10 An ABI of less than 0.90 is 95% sensitive
and 99% specific for angiographically diagnosed PAD.11
At present, angiography is reserved for PAD patients under consideration for
revascularization. Individuals with an ABI greater than 1.50 were excluded
(n = 136) because this indicates poorly compressible leg arteries and inability
to gauge arterial perfusion accurately.7,9
Individuals with PAD who were diagnosed in the noninvasive vascular
laboratory were excluded if their study visit ABI result indicated absence
of PAD. This occasionally occurred when PAD participants were revascularized
between vascular laboratory testing and their study visit. It also occurred
in individuals with borderline ABI values of approximately 0.90, due to measurement
variation. Patients with dementia were excluded because of their inability
to answer questions accurately (n = 7). Nursing home residents (n = 59), wheelchair-bound
patients (n = 27), and patients with foot or leg amputations (n = 77) were
also excluded because they have severely impaired functioning. Non–English-speaking
patients were excluded because investigators were not fluent in non-English
languages (n = 47). Patients with recent major surgery were excluded (n =
The ABI was measured using established methods.7- 10,12
After participants rested supine for 5 minutes, a hand-held Doppler probe
(Nicolet Vascular Pocket Dop II, Golden, Colo) was used to measure systolic
blood pressures in the right brachial artery, right dorsalis pedis and posterior
tibial arteries, left dorsalis pedis and posterior tibial arteries, and left
brachial artery. Each pressure was measured twice: in the order listed and
then in reverse order. The ABI was calculated in each leg by dividing average
pressures in each leg by the average of the 4 brachial pressures.12,13 Average brachial pressures in the
arm with highest pressure were used when 1 brachial pressure was higher than
the opposite brachial pressure in both measurement sets, and the 2 brachial
pressures differed by 10 mm Hg or more in at least 1 measurement set, since
in such cases subclavian stenosis was possible.12,13
The lowest leg ABI measurement was used in our analyses.
Leg symptoms were characterized into 1 of 6 mutually exclusive groups
using the San Diego claudication questionnaire.2
We used symptom groups previously developed by Criqui et al2
as a starting point for our group definitions and made 2 changes a priori
for the present study (Figure 1).
The San Diego claudication questionnaire is a standardized questionnaire based
on the Rose claudication questionnaire1- 3,5
and was administered by certified health interviewers.
Four groups have exertional leg symptoms. Individuals in the IC group
experience exertional calf pain that does not begin at rest, causes the participant
to stop walking, and resolves within 10 minutes of rest. Individuals in the
leg pain on exertion and rest group sometimes experience exertional leg pain
at rest. Two groups experience atypical exertional leg pain. Individuals in
the atypical exertional leg pain/carry on group experience exertional leg
symptoms that do not begin at rest and do not stop the individual from walking.
Similarly, individuals in the atypical exertional leg pain/stop group experience
exertional leg symptoms that do not begin at rest, but that do stop the individual
from walking, and do not involve the calves or do not resolve within 10 minutes
of rest. Two groups do not have exertional leg pain. Individuals in the no
exertional leg pain/active group do not experience exertional leg pain symptoms
and are considered active because they had walked more than 6 blocks during
the previous week. Similarly, individuals in the no exertional leg pain/inactive
group do not experience exertional leg pain, but are considered inactive because
they had only walked 6 or fewer blocks during the previous week.
Our leg symptom groups differ in 2 ways from those previously defined
by Criqui et al.2 First, for exertional symptoms
other than IC or leg pain on exertion and rest, we defined symptoms according
to whether participants walked through exertional pain (atypical leg pain/carry
on vs atypical leg pain/stop). We reasoned that functioning levels might be
substantially different between these 2 groups. Second, we divided the asymptomatic
category of Criqui et al into 2 groups based on physical activity. In some
instances insufficient physical activity may prevent exertional leg pain symptoms.
Alternatively, absence of leg pain symptoms might indicate a more mild form
of PAD. Of PAD participants without exertional leg pain symptoms, 31% walked
6 or fewer blocks during the previous week. We chose this criterion a priori
to define asymptomatic PAD participants who were inactive.
Algorithms developed for the Women's Health and Aging Study and the
Cardiovascular Health Study were used to document comorbidities.14
These algorithms combine data from patient report, physical examination, medical
record review, medications, laboratory values, and a primary care physician
questionnaire. Criteria developed by the American College of Rheumatology
were used to diagnose knee and hip osteoarthritis.15,16
Comorbidities assessed were angina, diabetes mellitus, myocardial infarction,
stroke, heart failure, pulmonary disease, knee and hip arthritis, spinal stenosis,
disk disease, Parkinson disease, and hip fracture.17- 20
Six-Minute Walk. Six-minute walk data correlate with treadmill performance, physical
activity, and the ABI.21- 23
Following a standardized protocol,21,24
participants walked up and down a 100-foot (30-m) hallway for 6 minutes after
being instructed to cover as much distance as possible.
Accelerometer-Measured Physical Activity. Physical activity levels were measured over 7 days using a vertical
accelerometer (Caltrac).25- 29
To directly compare activity among participants, we programmed each accelerometer
identically for height, weight, age, and sex using previously reported methods.25- 29
Because we had a limited number of accelerometers, they were distributed to
participants whenever available in a nonrandom fashion. There were no significant
differences in age, race, sex, or average ABI of participants based on accelerometer
Repeated Chair Raises. This test measured leg strength and balance.30,31
Participants sat in a straight-backed chair with arms folded across their
chest and stood 5 times consecutively as quickly as possible. Time for 5 chair
raises was measured.
Standing Balance. Participants were asked to stand with 1 foot directly in front of the
other for 10 seconds (full tandem stand).30,31
Four-Meter Walking Velocity. Walking velocity was measured with a 4-m walk performed at usual and
fastest pace. Each walk was performed twice. The faster walk in each pair
was used in our analyses.30,31
The Geriatric Depression Score Short-Form measures depressive symptoms.32 Depression was defined as 6 or more depressive symptoms.32 A monofilament assessed sensation on the dorsal and
ventral surface of each foot in specific locations.33,34
We used an ordinal neuropathy score to represent the number of items missed
(range, 0-22, a lower score equaled a better score). Using the Walking Impairment
Questionnaire,35 participants recorded the
degree to which specific symptoms (including joint symptoms and chest pain)
impaired walking during the past week on a Likert scale ranging from 0 to
For continuous variables, differences among PAD participants with IC,
PAD participants without IC, and non-PAD participants were evaluated using
analysis of covariance adjusted for age, sex, and race. Among PAD participants,
differences in continuous variables between IC and the remaining leg symptom
categories were evaluated using analysis of variance. Rates for dichotomous
variables were estimated using linear regression models with dummy variables
for leg symptom categories; however, the test of significance was based on
logistic regression analyses. Differences in continuous variables between
IC and the remaining leg symptom categories were evaluated using analysis
of covariance adjusted for age, race, sex, ABI, and comorbidities likely to
confound the association between leg symptom group and functioning (neuropathy,
depression, disk disease, spinal stenosis, hip or knee osteoarthritis, and
Each symptom group was ranked from 1 to 6 for the 9 functional measures,
in which 1 represented best performance and 6 represented worst performance.
These rank scores were summed, resulting in a score in which the minimum possible
score was 9 (best) and the maximum possible score was 54 (worst). This summary
score allowed for numerical comparison of composite functional performance
across all 6 groups.
The significance of between-group comparisons in dichotomous variable
rates was based on logistic regression using dummy variables for leg symptom
groups. When comparing differences between IC and other groups, a significance
level of .01 was used for each comparison so that the overall level of significance
was less than .05 (Bonferroni inequality). Analyses were performed using SAS
statistical software (version 8.0, SAS Institute Inc, Cary, NC).
Figure 2 depicts reasons for
nonparticipation among individuals identified for the study. Participants
included 460 men and women with PAD and 130 without PAD. For participants,
the mean (SD) age was 70.9 (8.4) years and 56.2% were male vs 72.6 (9.8) years
for nonparticipants (49.8% were male). In pairwise comparisons, the 310 PAD
participants without IC performed significantly more poorly on each functional
measure than non-PAD participants (data available from authors).
Of the 460 PAD participants, 150 (32%) had IC, 88 (19%) had leg pain
on exertion and rest, 41 (9%) had atypical exertional leg pain/carry on, 90
(20%) had atypical exertional leg pain/stop, and 91 (20%) had no exertional
leg pain (63 active and 28 inactive). Table
1 compares characteristics among the symptom groups. All results
are presented in reference to IC. The atypical exertional leg pain/carry on
group had a higher ABI measurement (0.70 vs 0.61; P
= .004) and lower prevalences of lower-extremity revascularization (22.0%
vs 45.3%; P = .008) and depression (5.1% vs 25.9%; P = .01). Participants in the leg pain on exertion and
rest group had a higher prevalence of comorbidities associated with leg pain
than those with IC (Table 1).
They also had a higher (worse) neuropathy score (5.6 vs 3.5; P<.001) and higher prevalences of diabetes mellitus (48.9% vs 26.7%; P<.001), spinal stenosis (20.8% vs 7.2%; P = .002), and 3 or more comorbidities (68.2% vs 43.3%; P = .003).
A substantial proportion in the 2 no exertional leg pain groups developed
leg pain during the 6-minute walk. However, these groups were less likely
to develop leg pain than the IC group (33.3% vs 88.6% [P<.001] for the no exertional leg pain/active group; and 53.6% vs
88.6% for the no exertional leg pain/inactive group [P<.001]).
Table 2 shows functioning
levels by symptom group among PAD participants adjusted for age, sex, race,
and comorbidities. Overall, participants with atypical leg pain/carry on and
atypical exertional leg pain/stop had better functioning than those with IC.
The no exertional leg pain/inactive group and the pain on exertion and rest
group performed more poorly than those with IC. The no exertional leg pain/active
group performed comparably with those who had IC. In the 6-minute walk, participants
in the atypical exertional leg pain/carry on group were less likely to stop
(6.8% vs 36.0%; P = .002) and achieved a greater
distance (404.3 vs 328.5 m; P<.001) compared with
the IC group. Participants in the atypical exertional leg pain/stop group
were less likely to stop (19.8% vs 36.0%; P<.01)
and achieved a greater distance in the 6-minute walk (378.8 vs 328.5 m; P<.01) than IC.
The no exertional leg pain/inactive group had a slower walking velocity
(0.78 vs 0.90 m/s for usual pace 4-m walk [P<.01];
1.03 vs 1.24 m/s for the fast 4-m walk [P<.01])
compared with the IC group. The pain on exertion and rest group had a slower
time for 5 chair raises (13.5 vs 11.9 seconds; P<.01),
slower walking velocities (0.80 vs 0.90 m/s [P<.01]
for usual pace; and 1.05 vs 1.24 m/s [P<.01] for
fast-pace 4-m walk), and held the full tandem stand less often (37.5% vs 60.0%; P<.01) compared with the IC group.
Table 3 compares the degree
to which walking was impaired by specific symptoms in each group. Compared
with IC, the atypical exertional leg pain/carry on group reported less walking
impairment from the following: pain, aching, or cramps in the calves/buttocks
(P<.001), pain or aching in thighs (P = .004), pain, stiffness, and aching in lower-extremity joints (knee/ankle/hip)
(P = .007), and weakness in 1 or both legs (P<.001). Compared with the IC group, the leg pain on
exertion and rest group reported greater impairment from pain or aching in
the thighs (P<.001), pain, stiffness, or aching
in lower-extremity joints (P<.001), pain or discomfort
in the chest (P = .01), shortness of breath (P = .002), and heart palpitations (P<.001).
This study shows that patients with PAD, confirmed with noninvasive
ABI testing, have a broad range of symptoms other than classic IC. Additionally,
a substantial proportion of patients with PAD are asymptomatic or have exertional
leg symptoms other than IC. Our findings suggest that comorbid disease and
activity levels contribute to the diversity of leg symptoms in PAD. In particular,
PAD participants with leg pain on exertion and rest have higher prevalences
of comorbid diseases such as diabetes mellitus, neuropathy, and spinal stenosis
that may blur PAD-associated leg symptoms.
Although IC is considered indicative of functional impairment, our data
show functional impairment occurs in PAD patients without IC. Furthermore,
the type of leg symptoms reported is an important correlate of the degree
of functional impairment in PAD.
Participants in the atypical exertional leg pain/carry on group achieved
significantly greater distance in the 6-minute walk than the IC group, which
may indicate greater pain tolerance and/or greater willingness to "walk through"
exertional leg pain. These same atypical exertional leg pain/carry on participants
had a low prevalence of depression, suggesting they may be more motivated
and more likely to "push through" leg symptoms.
Compared with IC, subjects with leg pain on exertion and rest had more
comorbid diseases, greater walking impairment from comorbidity-related symptoms,
and worse functioning. Leg pain on exertion and rest in PAD may result in
part from comorbid diseases, particularly neurogenic and degenerative arthritis
problems affecting the legs. Our data suggest that clinicians should seriously
consider the possibility that comorbid diseases such as diabetes mellitus,
neuropathy, intervertebral disk disease, and spinal stenosis are contributing
to walking impairment in PAD patients with leg pain on exertion and rest.
Since our data show that the leg pain on exertion and rest group often report
walking impairment from symptoms such as shortness of breath and lower-extremity
joint pain, diagnostic assessment might be pursued based on patients' symptoms.
More than half of the no exertional leg pain/inactive group developed
leg pain during the 6-minute walk, suggesting some participants were limiting
their activity to avoid leg pain symptoms. Possibly some had previously experienced
exertional leg symptoms and limited their activity to avoid the pain. Although
we did not ascertain reasons for referral to the noninvasive vascular laboratory,
this could explain why 20% of persons with PAD reported no exertional leg
pain symptoms, despite prior referral to the vascular laboratory. Other potential
explanations follow. First, some may have previously had leg pain symptoms
that resolved after revascularization, despite their lower-extremity arterial
perfusion remaining abnormal. Others may have been referred to the vascular
laboratory for testing prior to coronary artery bypass grafting or because
of a nonhealing leg ulcer. Some patients may have minimized symptoms to the
We did not find clinically meaningful differences in ABI values between
groups. Measurement error for ABI is ± 0.15.8
Thus, while the ABI is highly sensitive and specific for diagnosing PAD, it
cannot be used to classify patients into symptom groups.
This study has several study limitations. First, our findings may not
be generalizable to persons with PAD identified from a community-dwelling
setting. Because participants were required to return for the study, the most
ill or disabled may be underrepresented. We did not systematically collect
data allowing us to compare participants at the 3 vascular laboratories from
where participants were recruited. However, participants in this study are
comparable with individuals with established PAD from other medical centers.2,36,37 We did not perform
power calculations a priori for this study. Data are from a large, observational
study with the primary purpose of identifying predictors of poorer functioning
in PAD. Although our study cohort included 460 participants with PAD, sample
sizes were small for some symptom groups, limiting power to detect significant
differences. We did not have sufficient power to determine whether the relationships
reported here differed significantly between men and women. We did not collect
data on the duration or severity of comorbid disease. However, our data indicate
that associations between leg symptom groups and functional impairment are
independent of comorbid disease. Finally, because this was a cross-sectional
study, we were not able to determine whether lower-extremity revascularization
leads to changes in the leg symptoms experienced by study participants with
In conclusion, our findings show that patients with PAD, as determined
by noninvasive ABI testing, have a range of leg symptoms beyond IC. Our data
suggest that clinicians can use leg symptoms to gauge the degree of functional
impairment associated with PAD. Clinicians should consider the possibility
that comorbid disease and physical activity may alter PAD-associated leg symptoms.
Further study is needed to determine whether treatment of associated comorbid
illnesses and other factors influencing leg symptoms improve functioning in
patients who have PAD associated with symptoms other than IC.