The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis of Mortality and Disease Progression in Peripheral Arterial Disease

Abstract

Objective:
Peripheral arterial disease (PAD) afflicts up to 20% of older people and is associated with a high risk of cardiovascular (CV) morbidity, but a rather low risk of progression of leg symptoms. These risk estimations are largely taken from cohort studies performed 20 years ago. To test the validity of this, available data were systematically reviewed and attempts were made to perform meta-analyses of CV risk and disease progression.

Methods:
A database literature search was conducted of the period 1990-2015 using related subject headings. Inclusion criteria were cohort studies for PAD, sample size >100 subjects, follow up time ≥1 year, and studies presenting endpoints covering mortality and/or CV events. Analyses were performed for a reference population, as well as groups with asymptomatic PAD (APAD), symptomatic PAD, and subjects with ankle brachial index <0.9.

Results:
Of 354 identified articles, 35 were eligible for systematic review. Sample size varied between 109 and 16,440 subjects. Mean age in the cohorts ranged from 56 to 81 years (SD 10.8) and mean follow up was 6.3 years (range 1-13). Most included patients with symptomatic PAD had IC (91%). Symptomatic PAD subjects had higher 5 year cumulative CV mortality than the reference population, 13% versus 5%. During follow up, approximately 7% of APAD patients progressed to IC, and 21% of IC patients were diagnosed as having critical limb ischemia, with 4-27% undergoing amputations.

Conclusion:
The risk to the limb is underestimated in PAD patients, whereas the CV related morbidity is more moderate than stated in the guidelines. The latter observation is especially valid for IC patients. These findings should be considered when evaluating patients for treatment.

Full text

REVIEW
The Risk of Disease Progression in Peripheral Arterial Disease is Higher than
Expected: A Meta-Analysis of Mortality and Disease Progression in
Peripheral Arterial Disease
B. Sigvant
a,b,e,*
, F. Lundin
c
, E. Wahlberg
d
a
Institution of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
b
Department of Surgery Central Hospital, Karlstad, Sweden
c
Department of Medical Epidemiology and Statistics Central Hospital, Karlstad, Sweden
d
Department of Cardiothoracic and Vascular Surgery, Linköping University Hospital, Linköping, Sweden
e
Department of Clinical Research and Education, Södersjukhuset, Karolinska Institute, Sweden
WHAT THIS PAPER ADDS
This review of available data of risk to limb and life among subjects with peripheral arterial disease indicates a
higher risk of disease progression than expected. These findings should be considered when evaluating patients
for treatment and interventions.
Objective: Peripheral arterial disease (PAD) afflicts up to 20% of older people and is associated with a high risk of
cardiovascular (CV) morbidity, but a rather low risk of progression of leg symptoms. These risk estimations are
largely taken from cohort studies performed 20 years ago. To test the validity of this, available data were
systematically reviewed and attempts were made to perform meta-analyses of CV risk and disease progression.
Methods: A database literature search was conducted of the period 1990e2015 using related subject headings.
Inclusion criteria were cohort studies for PAD, sample size >100 subjects, follow up time 1 year, and studies
presenting endpoints covering mortality and/or CV events. Analyses were performed for a reference population,
as well as groups with asymptomatic PAD (APAD), symptomatic PAD, and subjects with ankle brachial index <0.9.
Results: Of 354 identified articles, 35 were eligible for systematic review. Sample size varied between 109 and
16,440 subjects. Mean age in the cohorts ranged from 56 to 81 years (SD 10.8) and mean follow up was 6.3 years
(range 1e13). Most included patients with symptomatic PAD had IC (91%). Symptomatic PAD subjects had higher
5 year cumulative CV mortality than the reference population, 13% versus 5%. During follow up, approximately
7% of APAD patients progressed to IC, and 21% of IC patients were diagnosed as having critical limb ischemia,
with 4e27% undergoing amputations.
Conclusion: The risk to the limb is underestimated in PAD patients, whereas the CV related morbidity is more
moderate than stated in the guidelines. The latter observation is especially valid for IC patients. These findings
should be considered when evaluating patients for treatment.
Ó2015 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved.
Article history: Received 19 March 2015, Accepted 26 October 2015, Available online XXX
Keywords: Amputation, Mortality, Natural history, Peripheral arterial disease
INTRODUCTION
Peripheral arterial disease (PAD) may be suspected by
symptoms and is diagnosed by ankle brachial index (ABI)
measurements. PAD is common in older people, with about
20% of those afflicted >60 years of age, increasing to nearly
50% in those aged 85 years. Most PAD is asymptomatic,
but about 35% have symptoms such as intermittent clau-
dication (IC) or critical limb ischemia (CLI).
1
The prevalence is likely to increase even further in the
future with the growing number of older people in society.
PAD is associated with the lowest quality of life of all
symptomatic cardiovascular (CV) disease manifestations,
2e4
and is also the most costly.
5
Accordingly, from a public
* Corresponding author. Department of Surgery, Central Hospital Karl-
stad, Rosenborgsgatan SE-65185 Karlstad, Sweden.
E-mail address: birgitta.sigvant@liv.se (B. Sigvant).
1078-5884/Ó2015 European Society for Vascular Surgery. Published by
Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ejvs.2015.10.022
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022
Eur J Vasc Endovasc Surg (2015) -,1e9

health perspective it is important to have solid data on the
epidemiology and the natural course of PAD.
All PAD stages are associated with an increased risk of CV
mortality and morbidity besides the probability of deterio-
ration of leg symptoms. The scientific support for this comes
largely from patient cohorts followed in the 1980s and early
1990s, before the era of risk factor modification and
endovascular treatment options. In particular, it seems that
data on natural course of PAD are based mainly on publi-
cations from the early 1990s.
6,7
The aim of this systematic review was to improve
assessment of risks associated with PAD by gathering all
current reliable data on long-term risk of leg symptom
deterioration and CV morbidity associated with PAD. A
secondary aim was to assess if these risks differed among
the sexes.
METHODS
Search strategy
An electronic literature search was conducted in December
2010 using the MEDLINE database. A second literature
search was performed on 24 April 2015 using MEDLINE Web
of Science, Science direct and Cochrane database. Three
more articles were identified for inclusion in the analysis.
The search strategy included basic medical subject head-
ings: peripheral vascular disease, intermittent claudication,
lowers extremity ischemia and poly-vascular disease, which
were followed by the headings: prognosis and natural his-
tory. The results were combined to capture all articles
encompassing these two topics. Retrieved papers were
limited to English language and human studies. Two in-
vestigators were responsible for the search and data
extraction.
Selection
The titles of the identified articles were reviewed for rele-
vance followed by assessment of the abstracts for topic
significance. Finally, the full texts of selected articles were
studied. Additional relevant citations identified from
reviewed articles were also gathered. All types of study
designs were accepted (Fig. 1). Inclusion criteria were:
cohort studies covering PAD patients, sample size 100
subjects, clear presentation of mortality and/or CV events
and endpoint data, and follow up 1 year.
Review articles, interventional studies, and publications
without relevant endpoints were excluded as were studies
in which PAD was diagnosed without the use of an objective
method and based on symptoms only. For studies reporting
data from the same population several times, the publica-
tion with the longest follow up time was used.
Data extraction and data processing
The content of selected studies was described and pre-
sented to summarize inclusion criteria, concomitant risk
factors, baseline medication use, and outcome. Follow up
time was estimated as mean follow up time or total follow
up time depending on available data. Three studies re-
ported median follow up time. For each included study the
following data were extracted: population characteristics
(sample size, number of men and women, mean age with
standard deviation when available), baseline characteris-
tics (smoking, diabetes mellitus (DM), hypertension (HT),
previous CV events, and medication), endpoints (CV and
non-CV mortality and all-cause mortality), CV events dur-
ing follow up, and PAD stage at baseline. The study pop-
ulations were classified into six groups based on baseline
disease stage (Table 1): (1) reference, no PAD (defined by
no/unknown PAD or 0.9 ABI 1.4); (2) asymptomatic
PAD (ABI <0.9 without clinical manifestations); (3) symp-
tomatic PAD (diagnosed IC or CLI); (4) ABI based group
(subjects with ABI <0.9 without information on symp-
toms); (5) total PAD cohort (all subjects with PAD
regardless of stage or with ABI <0.9); and (6) PAD subjects
separated by sex.
For analysis of incidence, incidence rates were collected
when reported, otherwise raw numbers and follow up time
were used.
The cohorts analyzed consisted of different studies. For
description of PAD stages, analyses were based on 23
studies that displayed baseline data on clinical diagnoses
(included in reference, asymptomatic, and/or symptomatic
PAD groups). The remaining studies based PAD diagnosis on
ABI without presenting symptoms. Data on disease pro-
gression was presented in eight studies, while revasculari-
zation and amputation rates were given in six and nine
studies, respectively. Only five studies separated analyses by
sex. In all but two studies (one using age only and one
unadjusted) adjustments were made for age, gender, BMI,
and risk factors as smoking, DM, HT, and prevalent CV
disease (Table 2).
Validation
All studies fulfilling the inclusion criteria were critically
assessed by two investigators to ensure reporting of rele-
vant information. Study design, selected population, sample
size, sex, and follow up time frame were summarized and
evaluated in concordance with PRISMA guidelines.
8
Statistical analyses
Study characteristics such as inclusion criteria, concomitant
risk factors, medication use, follow up time, and outcome
were tabulated. For each study the extracted data covered
both the entire reported population and sub-populations
defined by disease stage as described above. Incidence ra-
tio with confidence intervals was extracted when possible.
In studies not presenting incidence in events per person
years, person year exposure was computed from available
sources (mean follow up time, total follow up time). In-
cidences with confidence intervals (CI) for mortality end-
points were computed for each subgroup and plotted in
Forest plots with or without pooled estimates using a
random effects model.
9
Because of data heterogeneity, 95%
CI were calculated and random effect models used to pool
2B. Sigvant et al.
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022

data. The same strategy was used in the comparisons be-
tween baseline groups and between men and women.
Comparisons between disease stages were based on inci-
dence ratio within studies using the strategy described
above. To explore study heterogeneity meta-regression was
used, with population mean age and publication year as
explanatory variables. Analyses were also performed sepa-
rately for men and women.
All analyses were done using Stata version 13/MP4 (Stata
Corp LP, College Station, Texas, USA). For meta-analyses the
metan command was used, and meta-regressions were
made with metareg.
RESULTS
Study heterogeneity
The final analysis included 35 papers; the study selection
process is presented in Fig. 1. The most common reasons for
exclusion of studies were use of inadequate endpoints and
uncertain diagnosis of PAD. Patient cohorts varied greatly
(Table 2). A diversity of PAD stage definitions was used, and
while some studies were community based, others
recruited subjects eligible for treatment at hospital clinics.
Three studies used healthcare data registries. Study char-
acteristics such as inclusion criteria, sex, age distribution,
length of follow up, study period, region, and reporting of
events also varied (Tables 1 and 2). The meta-analysis
showed significant heterogeneity and as a result random
effect models were used for incidence and incidence ratio
(Figs. 2AeC, 3A,B, and 4A,B). The result of the meta-
regression did not give a clear picture of what may have
caused the between study heterogeneity, with one excep-
tion. The publication year clearly affected the rate of re-
vascularizations in the PAD group as presented in Fig. 2B.
Men were slightly over-represented (56%) in the studies
and mean age ranged from 56 to 81 years. Follow up time
varied from 1 to 13 years. Reporting of CV events also
differed. Some studies reported only all cause mortality and
two studies did not report fatal and non-fatal events
separately. For non-fatal events, a few studies reported first
event of each type (n¼4), others number of subjects
experiencing each event type (n¼6) or all events occurring
during follow up (n¼3). Presence of DM, HT, and smoking
habits also differed (p<.001) between PAD stage groups.
The prevalence of DM was 28% higher in the APAD group
than in the reference population, and in the PAD groups it
was 30% higher. Corresponding figures for smoking were
60% and 81%, respectively.
Disease progression and clinical course
Increasing age, low and high ABI, male sex, smoking, DM,
and concurrent CV disease were all predictors of disease
progression. Incidence rate (events per 1,000 person years)
leading to disease deterioration varied between cohorts.
The incidence rate for APAD patients deteriorating to IC was
15.1 (8.1e22.4), corresponding to a cumulative incidence
over 5 years of 7% (4e11%). For IC patients progressing to
CLI or IC deterioration, the 5 year cumulative incidence rate
was 46.6 (26.0e67.2) corresponding to 21% (12e29%) over
5 years (Fig. 2A). Only one study estimated incidence rate
for revascularization among APAD patients, and the rate
Table 1. Subgroups used for the analyses.
Subgroups Characteristics
1. Reference population, (0.9<ABI<1.4)
10,13,15e21,24,29,34e39
N¼28,452 (men 44%), mean age 67 years. The reference populations
within each study were in general younger, had less comorbidity, and
smoked less than PAD patients
2. APAD (ABI<0.9, without symptoms)
10,17,29,36,40
N¼1,789 (men 37%), mean age 70 years. Population and community
based mainly from recently performed studies
3. Symptomatic PAD (ABI <0.9 with
symptoms)
10,11,14,23,24,28,29,31,34e36,41e50
N¼57,322 (men 62%), mean age 67 years. IC and CLI patients (90%
and 10%). Eight studies did not report stage of PAD (n¼39,459)
4. ABI group (diagnosis based on ABI)
12,13,15,16,18e20,37e39,51
N¼7,636, (men 58%), mean age 70 years, 6,031 (74%) of subjects
were PAD patients referred to vascular clinics
5. Total disease cohort (APAD, symptomatic PAD, and ABI
subjects)
5,10e21,23,24,28,29,31,35e51
N¼66,753 (men 61%), mean age 67 years
6. Separate by sex
13,29,31,46,48
N¼21,567 (men 56%), mean age men and women 66.1 and 66.9
years, respectively
Peripheral Vascular
Disease, Lower Extremity
Ischemia, Polyvascular
Disease, Intermittent
Claudication + Prognosis +
Natural History
N=354
Excluded after abstract
screening
N=247
Articles selected for full text
retrieval
N=107
Excluded after full text screening
N= 72
Review articles 10
Irrelevant cohort or end point 30
Treatment studies 10
Small sample size 10
PAD diagnosis made by
questionnaires data 8
Same cohort in multiple publications 4
Articles eligible for
systematic
review
Figure 1. Flow chart showing how studies were selected. Humans,
English language, 1990e2015.
Meta-Analysis of Mortality and Disease Progression 3
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022

was highly variable over time in studies assessing IC
revascularization (Fig. 2B). The incidence rate for amputa-
tion also varied between 4% and 27% (Fig. 2C), where
95.4% of analyzed patients had IC and 4.6% CLI.
Mortality and CV events
All cause mortality rates were reported in 33 of the included
studies and CV mortality in 27, and the same comorbidities
and risk factors predicted PAD disease progression and CV
events. PAD groups had higher incidence of all cause mortality
when compared with the reference group (Fig. 3A). The inci-
dence of all cause mortality in the reference group was 27.6
(21.6e31.8) corresponding to 5 year cumulative incidence of
13% (10e15%). In the APAD group the corresponding inci-
dence was 41.3 (37.4e45.1) and the 5 year cumulative inci-
dence 19% (17e20%). The incidence and 5 year cumulative
Table 2. Baseline characteristics of included studies.
Study Publication
year
Number
(men %)
Age, mean
(SD)
Follow up
(years)
Study population
McKenna
15
1990 744 (46.1) 66.2 (11.6) 3.28
a
PAD patients referred to a vascular clinic
Dormandy
11
1991 1969 (79.8) 63.2 (9.1) 1
b
Treatment study of IC subjects
Criqui
29
1992 475 (45.7) 66 10
b
Population based (Lipid Research Clinic)
McGrae
McDermott
34
1994 422 (53.8) 67.6 4.3
b
Patients referred to vascular lab without previous
intervention
Leng
35
1996 1498 (51.4) 67.4 (0.3) 5
b
Population based (Edinburgh Artery Study)
Sikkink
12
1997 154 (66.2) 63 5
c
Medical records and ABI<0.9 without previous vascular
intervention
Brevetti
36
1998 110 (88.2) 63.6 (0.8) 2
c
Outpatients with IC without previous vascular
intervention
Jager
37
1999 631 (48.7) 64.5 (6.9) 5
b
Population based cohort study of disturbance of glucose
intolerance (Hoorn Study)
Kobayashi
38
2000 137 (91.2) 69 (8) 4.17
c
IC patients referred to a hospital
Muluk
39
2001 2777 (100) 64.7 (8.4) 3.92
c
IC patients at a Veterans Medical Centre
Vickrey
40
2001 10846 (53.7) 67.3 (9.2) 1.26
c
Patients from a healthcare database. Baseline groups
based on ICD codes (Stroke, MI or PAD)
Jude
23
2001 136 (59.6) 64.7 (10.8) 4.5
c
Symptomatic PAD patients referred for an angiogram
Pasqualini
41
2001 297 (73.7) 70.4 (8.7) 4
c
Symptomatic PAD outpatients referred to an out clinic
Jönsson
42
2002 240 (57.1) 69.2 (8.9) 12
b
Population based on ABI and IC
Murabito
16
2003 674 (37.5) 80.8 4
b
Population based (part of the Framingham cohort)
Fiotti
24
2003 669 (80.7) 64 13
a
IC patients referred to a vascular clinic, treated with
antiplatelet agents
Hooi
10
2004 3634 (46.9) 59.1 7.2
b
Population based
Resnick
18
2004 3989 (60.1) 56.1 8.3
c
Native Americans (The Strong Heart Study)
Lee
17
2004 1507 54e74 yrs 12 Population based (Edinburgh Artery Study)
Caro
43
2005 16440 (54.9) 67.3 (9.2) 5.9
c
Patients from a healthcare database. Baseline groups
based on ICD codes
Faglia
28
2006 564 (64.9) 70.1 (9.5) 3.4
c
Hospitalized patients with CLI and diabetes mellitus
O’Hare
13
2006 5682 (42.4) 73 (5.2) 11.1
a
Community based on ABI
Feringa
44
2006 2420 (72.0) 67 (11) 8
a
PAD patients referred to a vascular clinic (ABI based)
Steg
45
2007 8581 (63.8) 69 (10) 1
b
Outpatients with established PAD, ABI<0.9
Sutton-Tyrrell
19
2008 2682 (46.8) 73.6 (2.9) 6.7
c
Community based on ABI
Sprengers
14
2009 800 (69.0) 59.5 (12.7) 4.7
c
PAD patients referred to a vascular clinic
Vaartjes
31
2009 4158 (61.1) 66.4 (12.6) 3.78
c
Patients from a healthcare database. Baseline groups
based on ICD codes
Surinach
46
2009 763 (NA) NA 1.67
c
Register data on outpatients with any CV symptomatic
disease (FRENA)
Souminen
47
2009 1974 (58.4) 69.7 (11.8) 3.25
c
PAD patients referred to a vascular clinic
Taute
48
2009 109 (80.7) 60.8 (8.8) 8.67
c
Outpatients with IC
Diehm
49
2009 6821 (42.0) 73 (5.2) 5
b
Primary care attendees
Li
50
2010 3732 (52.5) 61.9 (9.5) 3.14
c
Hospitalized patients with PAD (The China ABI cohort)
Pasqualini
20
2012 654 (45.7) 74.8 (10.1) 1.6
c
Patients referred to an Internal Medicine clinic
Abola
51
2012 7996 (71.1) 65.4 (9.9) 3 Outpatients with any CV disease or multiple risk factors
(REACH)
Mueller
21
2014 487 (69.8) 70.0 (NA) 5 Symptomatic PAD patients referred to hospital
ABI ¼ankle brachial index; CLI ¼critical limb ischemia; IC ¼intermittent claudication; NA ¼not available; PAD ¼peripheral arterial
disease.
a
Median follow up.
b
Population follow up.
c
Mean follow up.
4B. Sigvant et al.
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022

incidence for symptomatic patients were 62.5 (51.2e73.7)
and 27% (23e31%), respectively. For the ABI group these
numbers were 75.6 (62.9e88.4) and 32% (27e36%).
Fig. 3A presents the incidence ratio for all cause mortality
in relation to the reference population, showing a similar
ratio for all groups with the exception of one study.
The incidence rate for CV mortality was 10.1 (7.5e12.8)
in the reference group corresponding to 5 year cumulative
incidence of 5% (4e6%). For APAD patients, the corre-
sponding figures were incidence rate 19.8 (14.7e24.9),
cumulative incidence 9% (7e12%); for the symptomatic
PAD group 28.3 (19.8e36.8), 13% (9e17%); and for the ABI
group 45.4 (34.3e56.6), 20% (16e25%). The incidence ratio
for CV mortality is presented in Fig. 3B.
Sex differences
Five of the included studies reported results separated by
sex. Men had significantly higher all cause mortality than
women, but pooled results were unreliable (Fig. 4A). Three
studies assessed CV mortality in men and women, where
only one observed higher incidence ratio in men
13
(Fig. 4B).
Data on PAD progression were not separated by sex in the
included studies.
Time trends
All study cohorts recruited subjects from both the general
population and from outpatient clinic settings (Table 2). Co-
horts recruited before the year 2000 had a younger mean age
(67.0 years, SD 9.1 years) than those performed after the year
2000 (69.0 years, SD 9.1). Incidence rates for CV mortality in
cohorts recruited before the year 2000 were consistent, with
one exception.
15
No clear trends were identified in the studies
recruiting patients after that time (Fig. 5).
DISCUSSION
A main finding in this systematic review is that even mild
PAD has a rather aggressive clinical course in respect to leg
problems. IC is commonly regarded as being quite stable in
Figure 2. Incidence (event/1000 person year) of (A) Deterioration of leg symptoms, asymptomatic PAD to IC and IC to CLI, or progression of
IC (in the study by Dormandy et al, were pts revascularized because “deterioration of arterial disease in legs requiring intervention”). (B)
Revascularized PAD patients. (C) Amputation in the PAD cohort.
Meta-Analysis of Mortality and Disease Progression 5
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022

its clinical course, with symptomatic deterioration of 7e9%
during the first year after diagnosis, followed by 2e3% per
year thereafter. The amputation rate is stated in the TASC
document to be around 1e3% over a 5 year period but it is
unclear where this estimation originates from.
22
The pre-
sent review found a more aggressive disease progression for
IC, with an amputation rate between 4% and 27%, likely to
be at the upper end of this range. The comparison was
rather inconclusive, however, because amputation rates are
largely dependent on cohort composition and duration of
follow up.
23,24
Unfortunately, no reliable source data were
identified on the amputation risk for APAD patients, but a
substantial risk was found for deterioration to IC with 15.1
(8.1e22.4) events (diagnosed IC) per 1,000 person years,
corresponding to a 5 year cumulative incidence 7% (4e
11%). Also this risk was higher than anticipated. To put this
in perspective, the incidence rates/100,000 for newly
diagnosed breast or colonic tumors in 2013 were 43 and
190, respectively, in Sweden.
25
This observation emphasizes
the need for exercise advice and perhaps even preventive
medication in asymptomatic patients.
26,27
CLI, on the other hand, is regarded as a condition with a
high risk of disability and amputation but data on its natural
history are scarce. Data on revascularization and amputa-
tion applied to only 1% and 6% of subjects in this study,
unfortunately, making it impossible to evaluate. Faglia
et al.
28
noted an amputation rate of 15% in a diabetic
cohort of CLI during 3.4 years of follow up, but overall data
are scarce and diverging, and this needs to be addressed in
future studies.
The lack of consistency when reporting mortality data
for PAD patients makes comparisons of CV risk difficult
Figure 4. (A) Comparison of all cause mortality (men relative to women) for symptomatic PAD cohort.
13,21,30,33
(B) Comparison of car-
diovascular mortality (men relative to women) for symptomatic PAD cohort.
13,21,30,33
Figure 3. Group comparisons of (A) all cause mortality (HR and 95% CI), and (B) cardiovascular mortality (HR and 95% CI).
6B. Sigvant et al.
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022

and studies included in this meta-analysis often use all
cause and CV mortality interchangeably. Because PAD is
an atherosclerotic disease and CV mortality is more likely
to be influenced by prevention, focus should be on CV
rather than all cause mortality. As expected, CV mortality
shows a pattern with slightly increasing incidence with
more severe PAD stage in the studies covered in this
analysis.
The study by Criqui et al. appears to be an outlier in
respect of mortality data for PAD.
29
For example, the APAD
group in that study displayed the highest relative incidence
for all cause mortality compared with all other studies. In
the present analysis the combined relative incidence was
1.68 for APAD patients. Furthermore, the reference popu-
lation in Criqui’s study was healthier than other reference
populations in the included studies in this analysis. These
circumstances may be one important explanation for the
notion of a general high mortality risk in APAD subjects.
An additional important observation in this analysis was
the lack of consistency among studies when reporting
mortality, but despite this, CV mortality appears to be only
moderately increased for PAD patients compared with
reference populations. It is common in the literature that
mortality or morbidity risks are the focus rather than the leg
problems. For example, in a review by Diehm et al. it was
stated that IC subjects should “worry less about their limbs
and focus on the MI and stroke risk,”but the findings in the
current meta-analysis indicate that this assumption might
need to be reconsidered.
30
A secondary aim in this study was to gain more solid
information on sex differences in the natural course of PAD.
Unfortunately, few studies provide data separated by sex,
but the included studies indicate increased all cause and CV
mortality in men. Differences in smoking behavior, with a
higher smoking prevalence among men, may explain this
divergence in mortality risk.
31
No data were available on
disease progression; this should be the focus of upcoming
studies.
A reduction in mortality was expected in more recently
performed studies, but long-term trends for PAD were
inconsistent in this analysis. Even though there are in-
dications of a decline in CV mortality in western countries
since the end of the twentieth century related to behavioral
changes in smoking, physical activity, and pharmacologic
treatment, this was not apparent in this analysis.
32
One
explanation may be the increased life expectancy.
33
The risk
of CV death may be consistent, but occurs later in life, so
when populations with PAD become older the risk will be
preserved. The observed changes over time may be a result
of this shift. Another plausible explanation is differences in
type of recruited studied populations over time. Earlier
cohorts were younger and population based to a large
extent, whereas more recent studies consisted of pop-
ulations with a more advanced disease burden.
Limitations
The main limitation of this systematic review is the vari-
ability in study cohorts and outcome measures of the
included studies, creating a significant heterogeneity and
unreliable pooling of data. Therefore, parts of the present
analysis are descriptive, and as such a more liberal way of
doing the analyses has been chosen. Another limitation is
the lack of data on CLI, and the decision to focus the
analysis on real-life cohort studies only made it harder to
find reliable data on CLI. For example, allowing randomized
clinical trial (RCT) would increase the amount of available
data but would reduce external validity and cause practical
problems with the active RCT treatment arms. The meta-
regressions indicated weak trends over publication year
and population age, but the confidence bands were too
wide for meaningful interpretations. It is still thought that
the pooled results are meaningful, but it is left to the reader
to interpret the pooled data with caution. Assessment of
the methodological quality of the included studies was not
made. Finally, the diagnostic methods, for example ABI, are
straightforward to use. These limitations withstanding, it is
believed that this study provides useful and partly novel
information on the natural course of limb and life in PAD
patients.
CONCLUSION
This attempt of a meta-analysis demonstrates that the risk
for the limb is underestimated in PAD patients while they
may have a moderate risk of CV related morbidity. This
observation is valid for IC subjects and for APAD. Unfortu-
nately there is limited information on risk available for CLI
Figure 5. Incidence rates of cardiovascular mortality for the total
disease cohort and study periods (incidence/1,000 person years).
Meta-Analysis of Mortality and Disease Progression 7
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022

patients and sex differences. These findings should be
considered when evaluating patients for treatment and
interventions.
CONFLICT OF INTEREST
None.
FUNDING
None.
APPENDIX A. SUPPLEMENTARY DATA
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.ejvs.2015.10.022.
REFERENCES
1Sigvant B, Wiberg-Hedman K, Bergqvist D, Rolandsson O,
Andersson B, Persson E, et al. A population-based study of
peripheral arterial disease prevalence with special focus on
critical limb ischemia and sex differences. J Vasc Surg
2007;45(6):1185e91.
2Longworth L, Buxton MJ, Sculpher M, Smith DH. Estimating
utility data from clinical indicators for patients with stable
angina. Eur J Health Econ 2005;6(4):347e53.
3Letterstal A, Forsberg C, Olofsson P, Wahlberg E. Risk attitudes
to treatment among patients with severe intermittent claudi-
cation. J Vasc Surg 2008;47(5):988e94.
4Xie J, Wu EQ, Zheng ZJ, Sullivan PW, Zhan L, Labarthe DR. Pa-
tient reported health status in coronary heart disease in the
United States: age, sex, racial, and ethnic differences. Circula-
tion 2008;118(5):491e7.
5Smolderen KG, Wang K, de Pouvourville G, Bruggenjurgen B,
Rother J, Zeymer U, et al. Two-year vascular hospitalisation
rates and associated costs in patients at risk of athero-
thrombosis in France and Germany: highest burden for pe-
ripheral arterial disease. Eur J Vasc Endovasc Surg 2012;43(2):
198e207.
6Adam DJ, Bradbury AW. TASC II document on the management
of peripheral arterial disease. Eur J Vasc Endovasc Surg
2007;33(1):1e2.
7Tendera M, Aboyans V, Bartelink ML, Baumgartner I,
Clement D, Collet JP, et al. ESC guidelines on the diagnosis and
treatment of peripheral artery diseases: document covering
atherosclerotic disease of extracranial carotid and vertebral,
mesenteric, renal, upper and lower extremity arteries: the Task
Force on the Diagnosis and Treatment of Peripheral Artery
Diseases of the European Society of Cardiology (ESC). Eur Heart
J2011;32(22):2851e906.
8Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC,
Ioannidis JP, et al. The PRISMA statement for reporting sys-
tematic reviews and meta-analyses of studies that evaluate
health care interventions: explanation and elaboration. PLoS
Med 2009;6(7):e1000100.
9Harris RJ, Deeks JJ, Harbord RM, Altman DG, Sterne JAC.
Metan: fixed- and random-effects meta analysis. Stata J
2008;8(1):3e28.
10 Hooi JD, Kester AD, Stoffers HE, Rinkens PE, Knottnerus JA, van
Ree JW. Asymptomatic peripheral arterial occlusive disease
predicted cardiovascular morbidity and mortality in a 7-year
follow up study. J Clin Epidemiol 2004;57(3):294e300.
11 Dormandy JA, Murray GD. The fate of the claudicantea pro-
spective study of 1969 claudicants. Eur J Vasc Surg 1991;5(2):
131e3.
12 Sikkink CJ, van Asten WN, van’t Hof MA, van Langen H, van der
Vliet JA. Decreased ankle/brachial indices in relation to
morbidity and mortality in patients with peripheral arterial
disease. Vasc Med (London, England) 1997;2(3):169e73.
13 O’Hare AM, Katz R, Shlipak MG, Cushman M, Newman AB.
Mortality and cardiovascular risk across the ankle-arm index
spectrum: results from the Cardiovascular Health Study. Cir-
culation 2006;113(3):388e93.
14 Sprengers RW, Janssen KJ, Moll FL, Verhaar MC, van der
Graaf Y. Prediction rule for cardiovascular events and mortality
in peripheral arterial disease patients: data from the prospec-
tive second manifestations of ARTerial disease (SMART) cohort
study. J Vasc Surg 2009;50(6):1369e76.
15 McKenna M, Wolfson S, Kuller L. The ratio of ankle and arm
arterial pressure as an independent predictor of mortality.
Atherosclerosis 1991;87(2e3):119e28.
16 Murabito JM, Evans JC, Larson MG, Nieto K, Levy D, Wilson PW.
The ankle brachial index in the elderly and risk of stroke, cor-
onary disease, and death: the Framingham Study. Arch Int Med
2003;163(16):1939e42.
17 Lee AJ, Price JF, Russell MJ, Smith FB, van Wijk MC, Fowkes FG.
Improved prediction of fatal myocardial infarction using the
ankle brachial index in addition to conventional risk factors:
the Edinburgh Artery Study. Circulation 2004;110(19):3075e
80.
18 Resnick HE, Lindsay RS, McDermott MM, Devereux RB,
Jones KL, Fabsitz RR, et al. Relationship of high and low ankle
brachial index to all-cause and cardiovascular disease mortal-
ity: the Strong Heart Study. Circulation 2004;109(6):733e9.
19 Sutton-Tyrrell K, Venkitachalam L, Kanaya AM, Boudreau R,
Harris T, Thompson T, et al. Relationship of ankle blood pres-
sures to cardiovascular events in older adults. Stroke
2008;39(3):863e9.
20 Pasqualini L, Schillaci G, Pirro M, Vaudo G, Leli C, Colella R,
et al. Prognostic value of low and high ankle brachial index in
hospitalized medical patients. Eur J Int Med 2012;23(3):240e4.
21 Mueller T, Hinterreiter F, Luft C, Poelz W, Haltmayer M,
Dieplinger B. Mortality rates and mortality predictors in
patients with symptomatic peripheral artery disease strati-
fied according to age and diabetes. J Vasc Surg 2014;59(5):
1291e9.
22 Norgren L, Hiatt WR, Harris KA, Lammer J. TASC II section F on
revascularization in PAD. J Endovasc Ther 2007;14(5):743e4.
23 Jude EB, Oyibo SO, Chalmers N, Boulton AJ. Peripheral arterial
disease in diabetic and nondiabetic patients: a comparison of
severity and outcome. Diabetes Care 2001;24(8):1433e7.
24 Fiotti N, Altamura N, Cappelli C, Schillan M, Guarnieri G,
Giansante C. Long term prognosis in patients with peripheral
arterial disease treated with antiplatelet agents. Eur J Vasc
Endovasc Surg 2003;26(4):374e80.
25 Ludvigsson JF, Andersson E, Ekbom A, Feychting M, Kim JL,
Reuterwall C, et al. External review and validation of the
Swedish national inpatient register. BMC Public Health
2011;11:450.
26 Sigvant B, Wiberg-Hedman K, Bergqvist D, Rolandsson O,
Wahlberg E. Risk factor profiles and use of cardiovascular drug
prevention in women and men with peripheral arterial disease.
Eur J Cardiovasc Prev Rehabil 2009;16(1):39e46.
27 Sigvant B, Henriksson M, Lundin F, Wahlberg E. Asymptomatic
peripheral arterial disease: is pharmacological prevention of
8B. Sigvant et al.
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022

cardiovascular risk cost-effective? Eur J Cardiovasc Prev Rehabil
2011;18(2):254e61.
28 Faglia E, Clerici G, Clerissi J, Gabrielli L, Losa S, Mantero M,
et al. Early and five-year amputation and survival rate of dia-
betic patients with critical limb ischemia: data of a cohort
study of 564 patients. Eur J Vasc Endovasc Surg 2006;32(5):
484e90.
29 Criqui MH, Langer RD, Fronek A, Feigelson HS, Klauber MR,
McCann TJ, et al. Mortality over a period of 10 years in patients
with peripheral arterial disease. New Engl J Med 1992;326(6):
381e6.
30 Diehm C, Kareem S, Lawall H. Epidemiology of peripheral
arterial disease. Vasa 2004;33(4):183e9.
31 Vaartjes I, de Borst GJ, Reitsma JB, de Bruin A, Moll FL,
Grobbee DE, et al. Long-term survival after initial hospital
admission for peripheral arterial disease in the lower extrem-
ities. BMC Cardiovasc Dis 2009;9:43.
32 Kesteloot H, Sans S, Kromhout D. Dynamics of cardiovascular
and all-cause mortality in Western and Eastern Europe be-
tween 1970 and 2000. Eur Heart J 2006;27(1):107e13.
33 Christensen K, Doblhammer G, Rau R, Vaupel JW. Ageing
populations: the challenges ahead. Lancet 2009;374(9696):
1196e208.
34 McDermott MM, Feinglass J, Slavensky R, Pearce WH. The ankle
brachial index as a predictor of survival in patients with pe-
ripheral vascular disease. J Gen Intern Med 1994;9(8):445e9.
35 Leng GC, Fowkes FG, Lee AJ, Dunbar J, Housley E, Ruckley CV. Use
of ankle brachial pressure index to predict cardiovascular events
and death: a cohort study. BMJ 1996;313(7070):1440e4.
36 Brevetti G, Martone VD, Perna S, Cacciatore F, Corrado S, Di
Donato A, et al. Intermittent claudication and risk of cardio-
vascular events. Angiology 1998;49(10):843e8.
37 Jager A, Kostense PJ, Ruhe HG, Heine RJ, Nijpels G, Dekker JM,
et al. Microalbuminuria and peripheral arterial disease are in-
dependent predictors of cardiovascular and all-cause mortality,
especially among hypertensive subjects: five-year follow up of
the Hoorn Study. Arterioscler Thromb Vasc Biol 1999;19(3):
617e24.
38 Kobayashi M, Shindo S, Kubota K, Kojima A, Ishimoto T, Iyori K,
et al. Causes of late mortality in patients with disabling inter-
mittent claudication. Jpn Circ J 2000;64(12):925e7.
39 Muluk SC, Muluk VS, Kelley ME, Whittle JC, Tierney JA,
Webster MW, et al. Outcome events in patients with claudi-
cation: a 15 year study in 2777 patients. J Vasc Surg
2001;33(2):251e7. discussion 7e8.
40 Vickrey BG, Rector TS, Wickstrom SL, Guzy PM, Sloss EM,
Gorelick PB, et al. Occurrence of secondary ischemic events
among persons with atherosclerotic vascular disease. Stroke
2002;33(4):901e6.
41 Pasqualini L, Schillaci G, Vaudo G, Innocente S, Ciuffetti G,
Mannarino E. Predictors of overall and cardiovascular mortality
in peripheral arterial disease. Am J Cardiol 2001;88(9):1057e60.
42 Jonsson B, Skau T. Ankle brachial index and mortality in a
cohort of questionnaire recorded leg pain on walking. Eur J
Vasc Endovasc Surg 2002;24(5):405e10.
43 Caro J, Migliaccio-Walle K, Ishak KJ, Proskorovsky I. The
morbidity and mortality following a diagnosis of peripheral
arterial disease: long-term follow up of a large database. BMC
Cardiovasc Dis 2005;5:14.
44 Feringa HH, van Waning VH, Bax JJ, Elhendy A, Boersma E,
Schouten O, et al. Cardioprotective medication is associated
with improved survival in patients with peripheral arterial
disease. J Am Coll Cardiol 2006;47(6):1182e7.
45 Steg PG, Bhatt DL, Wilson PW, D’Agostino Sr R, Ohman EM,
Rother J, et al. One-year cardiovascular event rates in
outpatients with atherothrombosis. Jama 2007;297(11):
1197e206.
46 Surinach JM, Alvarez LR, Coll R, Carmona JA, Sanclemente C,
Aguilar E, et al. Differences in cardiovascular mortality in
smokers, past-smokers and non-smokers: findings from the
FRENA registry. Eur J Int Med 2009;20(5):522e6.
47 Suominen V, Uurto I, Saarinen J, Venermo M, Salenius J. PAD as
a risk factor for mortality among patients with elevated ABIea
clinical study. Eur J Vasc Endovasc Surg 2010;39(3):316e22.
48 Taute BM, Thommes S, Taute R, Rapmund I, Lindner K,
Podhaisky H. Long-term outcome of patients with mild inter-
mittent claudication under secondary prevention. Vasa
2009;38(4):346e55.
49 Diehm C, Allenberg JR, Pittrow D, Mahn M, Tepohl G,
Haberl RL, et al. Mortality and vascular morbidity in older
adults with asymptomatic versus symptomatic peripheral ar-
tery disease. Circulation 2009;120(21):2053e61.
50 Li X, Luo Y, Xu Y, Li J, Hu D. Relationship of ankle brachial index
with all-cause mortality and cardiovascular mortality after a 3
year follow up: the China ankle brachial index cohort study.
J Hum Hypertens 2010;24(2):111e6.
51 Abola MT, Bhatt DL, Duval S, Cacoub PP, Baumgartner I, Keo H,
et al. Fate of individuals with ischemic amputations in the
REACH Registry: three-year cardiovascular and limb-related
outcomes. Atherosclerosis 2012;221(2):527e35.
Meta-Analysis of Mortality and Disease Progression 9
Please cite this article in press as: Sigvant B, et al., The Risk of Disease Progression in Peripheral Arterial Disease is Higher than Expected: A Meta-Analysis
of Mortality and Disease Progression in Peripheral Arterial Disease, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/
10.1016/j.ejvs.2015.10.022