Turkish Journal of Medical Sciences Turk J Med Sci

(2016) 46:
http://journals.tubitak.gov.tr/medical/
TBTAK
Research Article doi:10.3906/sag-1410-124

Diagnostic accuracy of ankle-brachial index measurement in peripheral arterial disease

in Turkish adults: a comparison with angiography
1, 1 2 1 3
Tolga DOAN *, lker TAI , Uur BOZLAR , Birol YILDIZ , Cengizhan AIKEL ,
1 4 1 1
Selim SAYIN , Celalettin GNAY , Erol ARSLAN , Kenan SALAM
1
Department of Internal Medicine, Glhane School of Medicine, Ankara, Turkey
2
Department of Radiology, Glhane School of Medicine, Ankara, Turkey
3
Department of Public Health, Glhane School of Medicine, Ankara, Turkey
4
Department of Cardiovascular Surgery, Glhane School of Medicine, Ankara, Turkey

Received: 26.10.2014 Accepted/Published Online: 06.12.2015 Final Version: 00.00.2016

Background/aim: This study aimed to determine the sensitivity and specificity of the ankle-brachial index (ABI) measurement in the
diagnosis of peripheral artery disease (PAD) using angiography as the gold standard in a group of Turkish subjects.
Materials and methods: In this single-center, cross-sectional, and observational study, subjects who had been evaluated using aorta
and lower extremity arterial imaging by angiography subsequently underwent an ABI measurement. Data related to anthropometrics,
cardiovascular risk factors, and blood biochemistry were also recorded. The sensitivity and specificity of a low ABI (0.9) were calculated
in comparison with angiography.
Results: A total of 57 patients (age: 59.1 15.9, male/female: 47/10) were enrolled. Diabetes mellitus, coronary artery disease, and
cerebrovascular disease existed in 40.4%, 42.1%, and 15.8% of the participants, respectively. Three or more cardiovascular risk factors
were present in 54.4%. The angiographic diagnostic method was computerized tomography angiography in 57.9%, digital subtraction
angiography in 38.6%, and magnetic resonance angiography in 3.5% of the subjects. Presence of PAD on angiography was recorded
in 55 of 57 participants. Calculated mean ABI value was 0.6 0.2 in the overall group, and a low ABI (0.9) was found in 82.5% (n =
47). When compared to angiography, the sensitivity of a low ABI test was found to be 83.6% and the specificity was 50%. A positive
predictive value of 97.9% was calculated. When an ABI of 0.95 was used as the diagnostic threshold, the sensitivity of the ABI test
increased to 90.9%.
Conclusion: We found the ABI measurement to be a reliable diagnostic method for lower extremity PAD when compared to the gold
standard of angiographic procedures. Establishing a higher cutoff value (0.95) may improve the diagnostic power of the test in Turkish
patients.

Key words: Ankle-brachial index, peripheral arterial disease, sensitivity, specificity, angiography

1. Introduction has emerged as a powerful, noninvasive diagnostic test

Peripheral artery disease (PAD) is associated with an
increased risk of morbidity and death (1). Lower extremity
PAD prevalence was reported as 20% in a multicenter
survey among Turkish adults aged 50 years or older
having moderate to high cardiovascular risk (2). However,
a recent study from our clinic showed a frequency of
PAD around 5% among internal medicine outpatients
(3). Nonetheless, awareness, detection, treatment, and
prevention of a vascular disease with a prevalence of
5%20% are particularly important. Although a definite
diagnosis of PAD can only be made using angiographic
procedures, the ankle-brachial index (ABI) measurement
* Correspondence: dr_tolgadogan94@yahoo.com
option in the identification of the disease (4,5).
Currently known sensitivity and specificity data regarding
ABI measurement mostly originate from the United
States, and ethnicity-based variations in ABI are largely
unknown. Moreover, the worldwide frequency of PAD
as detected by ABI varies markedly across countries and
regions, suggesting that the diagnostic accuracy of ABI
measurement may also differ. Therefore, the main objective
of this research project was to evaluate the sensitivity
and specificity of a low ABI value based on angiography
findings in subjects with suspected lower extremity arterial
stenosis.
1
 DOAN et al. / Turk J Med Sci
2. Materials and methods
2.1. Study population
This was a single-center, cross-sectional, and observational
study conducted at the Glhane Military Medical
Academys cardiovascular surgery, geriatrics, and internal
medicine clinics between March 2012 and May 2013.
The participants were selected among outpatients and
inpatients referred to the cardiovascular surgery clinic of
the institution. They were ordered routine angiography
due to a clinical judgement or a suspicious arterial
stenosis in the ultrasonographic evaluation, regardless
of the presence or absence of symptoms. Because of the
observational study design the decision of angiography was
made before enrollment. The main inclusion criteria were
having received angiography for aorta and lower extremity
arterial system short before enrollment and having no
extremity amputation. The presence of cancer or any other
terminal illness, and any condition that prevented giving
signed, informed consent, were the exclusion criteria.
2.2. Data collection
As the main procedure of this study, all included subjects
underwent an ABI measurement following angiography.
Demographic and anthropometric data including age,
height, body mass index, waist circumference, and hip
circumference measurements were recorded on a standard
chart. Smoking status, presence of hypertension, diabetes
mellitus, dyslipidemia, coronary artery disease (CAD),
cerebrovascular disease or stroke, and current medications
were also recorded. Blood test results obtained in the past
2 months, including complete blood count, glucose, renal
and liver function tests, and lipid profile, were recorded
from the hospital or the patients self-records. The total
number of major cardiovascular risk factors including
diabetes mellitus, dyslipidemia, hypertension, obesity, age,
and cigarette smoking were determined for each patient
within four categories as none, one risk factor, two risk
factors, and 3 risk factors.
2.3. The ankle-brachial index
The ABI measurement was performed using hand-
held Doppler (Hadeco, Japan) and four standard, fully
calibrated sphygmomanometers with a Velcro cuff (12
cm in size, ERKA, D-83646, Germany). With the patient
in supine position, after a resting period of at least 5
min, the systolic blood pressure readings were obtained
bilaterally from the brachial arteries, anterior tibial
arteries, and posterior tibial arteries. The first flow sound
heard while deflating the cuff was recorded as the target
reading. Two cycles of measurements were performed
starting from the right arm followed by the right leg, left
leg, and left arm. The mean of two readings was used for
each pulse in the calculation of ABI. Previously developed
improved standardization facilities were used during all
measurements (6). ABI was measured by an observer
blinded to the result of angiography.
2
ABI was calculated according to the Trans-Atlantic
Inter-Society Consensus Document on Management of
Peripheral Arterial Disease (TASC II) recommendations
(4). First, ABI was calculated separately for the right and
left limbs. The highest reading in either the anterior tibial
artery or the posterior tibial artery in each lower extremity
was divided by the highest brachial artery reading (right
or left). The lower one of the right and left ABI values was
considered as the final ABI result.
2.4. Categorical definitions of ABI
Subjects with an ABI value of 0.90 were considered to
have PAD (4). In addition, participants were also classified
into four other ABI categories: 0.91 to 1 borderline, 1.0
to 1.1 low normal, 1.10 to 1.4 normal, and >1.4 high
(7,8).
2.5. Angiographic assessments and radiological
classification
Computerized tomography angiography (CTA), digital
subtraction angiography (DSA), and magnetic resonance
arteriography (MRA) were the angiographic procedures
performed to identify PAD. Anatomical classification was
done by an experienced radiologist through reexamination
of imaging records of patients who were diagnosed with
PAD by angiography. Locations of stenoses were divided
into 3 main groups: aortoiliac, femoropopliteal, and
tibioperoneal (9). Aortic stenoses in the aortoiliac group
were indicated separately. The degree of any stenosis
was categorized into two categories: between 0% and
49%, indicating the absence of PAD, and 50% and above,
indicating the presence of PAD.
2.6. Statistical analyses
Commercially available SPSS software (version 15.0,
SPSS Inc., Chicago, Illinois) was used for statistical
analyses. Descriptive data were reported as percentages
for categorical variables and as mean standard deviation
for continuous data. The chi-square test or Fischers exact
test were used for comparisons of categorical variables.
Normality of distribution was tested by the one-sample
KolmogorovSmirnov test. Comparisons between
groups were performed using the MannWhitney U test.
Correlations were tested using the Pearson or Spearman
tests where appropriate. P < 0.05 was considered
statistically significant. Two different ABI cut-off values
were separately tested: 0.90 and 0.95. The ratio of
the detection of PAD as sensitivity and the ratio of the
detection of non-PAD as specificity were finally calculated.
3. Results
3.1. Demographics, clinical characteristics, and
laboratory findings
A total of 60 patients were initially enrolled, and 57 of them
(mean age: 59.1 15.9 years, males: 82.5%) were included
in the analyses. Three patients were excluded because of
 DOAN et al. / Turk J Med Sci

incomplete angiography images. Basic characteristics of Table 1. Demographic characteristics and laboratory findings.
the study population are presented in Table 1.
Hypertension was present in 27 (47.4%) patients, diabetes Variable Mean SD (minmax)
mellitus in 23 (40.4%), obesity in 6 (10.5%; BMI 30),
dyslipidemia in 33 (57.9%), CAD in 24 (42.1%), and Age (years) 59.1 15.9 (2082)
cerebrovascular disease in 9 (15.8%). Smoking was Sex, male, n (%) 47 82.5
reported in 26 (45%) patients, or in 47 (82.5%) when ex- Anthropometrics Mean SD (minmax)
smokers were included. A total of 45 patients (80.4%) Height (cm) 169.6 9.1 (153185)
were under treatment with cardiovascular medicines, 20 Body weight (kg) 74.0 11.4 (52100)
patients (35.7%) were on antidiabetic drugs, 41 patients
Body mass index (kg/m ) 2
25.9 4.3 (16.737.5)
(73.2%) were taking antithrombotics or warfarin, and 6
patients (10.7%) were on psychiatric medications. Waist circumference (cm) 93.8 12.6 (65137)
3.2. Risk factors for cardiovascular disease Hip circumference (cm) 98.0 10.6 (78147)
The total number of male patients aged >45 years and Waist to hip ratio 1.0 0.8 (0.81.2)
female patients aged >55 years was 43 (75.4%). Nine Blood tests
patients (15.8%) had none or a single risk factor, 17 Hemoglobin (g/dL) 13.9 4.9 (8.017.1)
(29.8%) had 2 risk factors, and 31 (54.4%) had 3 or more
Leukocyte (103/L) 7.5 2.2 (3.212.8)
risk factors.
3.3. ABI vs. angiographic diagnosis of PAD MCV (fL) 84.7 12.1 (56104)
The mean ABI value was 0.6 0.2 in the whole study group, Platelets (103/L) 273.8 96.1 (69523)
with no sex difference (ABI: 0.6 0.2 in males vs. 0.6 0.3 Sedimentation (mm/h) 28.0 29.3 (2131)
in females). According to the predefined ABI classification Glucose (mg/dL) 126.1 67.8 (59340)
(see Section 2), 82.5% of the patients had PAD, 12.3% had Urea (mg/dL) 47.4 32.3 (12172)
a borderline value, 1.8% had a low normal ABI, and 3.5%
Creatinine (mg/dL) 1.3 1.2 (0.68.8)
had a normal ABI (Table 2). The number of subjects who
were classified as having severe PAD with an ABI value of AST (U/L) 20.1 6.2 (1146)
0.4 was 7 (12.3%). There was no correlation between ABI ALT (U/L) 17.4 8.2 (753)
values and laboratory results. GGT (U/L) 37.4 36.5 (7192)
All participants had been evaluated by at least one Total cholesterol (mg/dL) 180.5 48.3 (67277)
angiographic diagnostic method; 33 patients (57.9%) Triglyceride (mg/dL) 149.2 120.5 (40471)
were evaluated by CTA, 22 patients (38.6%) by DSA,
HDL-cholesterol (mg/dL) 47.0 25.3 (2067)
and 2 (3.5%) patients by MRA (Table 2). After a careful
reexamination of the images and their reports, some degree LDL-cholesterol (mg/dL) 113.8 37.6 (36182)
of stenosis was detected in all 57 patients. However, 55 Albumin (g/dL) 4.1 0.4 (2.94.9)
(96.5%) patients were classified as having PAD according Total protein (mg/dL) 7.0 0.4 (6.07.9)
to angiography, while 2 (3.5%) subjects were not found to Sodium (mEq/L) 140.0 2.4 (135146)
have angiographically proven PAD (Table 2). Potassium (mEq/L) 4.3 0.4 (3.25.4)
3.4. Sensitivity and specificity of ABI in the diagnosis of
Calcium (mg/dL) 9.6 0.5 (8.510.4)
PAD with respect to angiography
A low ABI value was found in 46 (83.6%) out of 55 patients LDH (IU/L) 354.1 103.3 (172667)
who had angiographically detected PAD. One of the two
AST: Aspartate aminotransferase; ALT: alanine aminotransferase;
patients with no PAD according to angiography had a low
GGT: gamma glutamyl transferase; HDL: high-density
ABI value (false positive). Finally, the sensitivity of the ABI lipoprotein; LDH: lactate dehydrogenase; LDL: low-density
test to detect angiographically proven PAD was 83.6%, but lipoprotein; MCV: mean corpuscular volume.
the specificity was 50%. A strong positive predictive value
of 97.9% of a low ABI to identify angiographically detected
PAD was calculated. unchanged (50%) due to an insufficient number of
When the same analysis was applied to an ABI 0.95 participants with nondiagnostic angiography for PAD.
threshold for PAD diagnosis, 50 (90.9%) of the 55
angiographically detected PAD patients were categorized 4. Discussion
as having PAD. In this case, both the sensitivity and the This study on a group of Turkish people revealed the ABI
positive predictive value improved (90.9% and 98%, measurement to be a reliable test in the diagnosis of PAD
respectively) (Table 3). However, specificity remained based on angiographically identified disease. To the best

3
 DOAN et al. / Turk J Med Sci

Table 2. ABI and angiography-based diagnosis of PAD. 84% and 3 risk factors in 54.4% of our participants, with
coexisting hypertension, diabetes mellitus, or CAD in more
ABI Mean SD than 40%, the present study population was quite similar
to those that were previously studied by different authors.
Overall 0.6 0.2 However, the enrollment of a small number of subjects
Female (n = 10) 0.6 0.3 with normal or nondiagnostic angiography for PAD
Male (n = 47) 0.6 0.2 appeared to be a drawback of our study. While ordering
ABI categories n (%) angiography solely for study purposes in asymptomatic
PAD (ABI 0.9) 47 (82.5) subjects with a normal ABI was not possible due to ethical
considerations, it is also not recommended for most cases
Borderline (ABI: 0.911) 7 (12.3)
with a low ABI value (4).
Low-normal (ABI: 1.01.1) 1 (1.8) The ABI has been a sensitive, reliable, and cost-effective
Normal (ABI: 1.111.40) 2 (3.5) method in the diagnosis of PAD (4,10). It has also been
High (ABI >1.4) 0 demonstrated to be useful in the detection of generalized
Angiographic methods n (%) atherosclerosis (11). The presence of a low ABI or
CTA 33 (57.9) subclinical atherosclerosis among older patients without
a history of CAD was shown to be a significant predictor
DSA 22 (38.6)
of all-cause mortality and cardiovascular morbidity (12).
MRA 2 (3.5) Thus, routine, effective use of the ABI measurement can
Angiographic diagnosis of PAD n (%) improve patient management in any population.
PAD (+) 55 (96.5) When compared to DSA, CTA had 96.4% sensitivity and
PAD (-) 2 (3.5) 98.4% specificity for the detection of peripheral arterial
stenoses (13), while the sensitivity and specificity of MRA
ABI: Ankle-brachial index; CTA: computerized tomography were documented to be almost 100% (14). The less invasive
angiography; DSA: digital subtraction angiography; MRA: CTA and MRA have increasingly been preferred to the
magnetic resonance angiography; PAD: peripheral arterial more invasive DSA due to patient comfort and low risk
disease. of complications. Accordingly, despite being considered
a gold-standard imaging method, DSA could only be
performed in less than one-third of our patients.
of our knowledge, this is the first report from Turkey Based on a threshold value of 0.9, the ABI was reported to
presenting evidence for the utility of noninvasive ABI have 79%95% sensitivity and 85%100% specificity with
measurement in the detection of lower extremity PAD. respect to angiographic methods in the diagnosis of PAD
Smoking, older age, diabetes mellitus, hyperlipidemia, and (15). In our study, the sensitivity of the ABI was found to
hypertension were linked to increased PAD incidence, be 83.6% using this threshold, which seems acceptable and
with the presence of at least one cardiovascular disease risk in line with the literature. Although the main objective of
factor in up to 95% of individuals with PAD (1). Given the our study was to determine the sensitivity, the specificity
identification of at least one cardiovascular risk factor in of ABI could be evaluated for only 2 cases that had <50%

Table 3. Sensitivity and specificity of a low ABI in the diagnosis of PAD with respect to
angiography.

Diagnosis according to ABI Diagnosis according to angiography

PAD absent PAD present Total

Absent 1 9 10
Threshold 0.9
Present 1 46 47
Absent 1 5 6
Threshold 0.95
Present 1 50 51
Total 2 55 57

ABI: Ankle-brachial index; PAD: peripheral arterial disease.

4
 DOAN et al. / Turk J Med Sci
stenosis on angiography. Due to a small sample size,
the specificity value of 50% for a low ABI in our study
population lacks adequate accuracy.
The sensitivity and specificity of ABI measurement for
lower extremity PAD diagnosis revealed inconsistent
findings in several studies, depending on various factors
including observer, equipment, and methods (16). In a
study by Khusrow et al., conducted on 107 patients and
based on DSA and ABI in the diagnosis of PAD, the
sensitivity and specificity of the ABI using higher or lower
ankle pressures were reported as 69%84% and 83%64%,
respectively (17). The calculation of ABI using lower ankle
pressure was associated with improved sensitivity, while
it decreased specificity values in other studies (18,19).
However, the American Heart Association and TASC2
guidelines continue with the recommendation of using
the higher ankle pressure in the calculation of ABI, which
was also the chosen method in the present investigation.
In a recent study by Nam et al., CTA, DSA, and ABI
were serially performed in 79 patients with intermittent
claudication. The authors reported relatively lower levels
of sensitivity and specificity of ABI of 61% and 87%,
respectively (19). The performance of ABI testing was
also evaluated according to the observer, which generally
revealed a high reproducibility of the method (10,20,21).
ABI measurement was performed meticulously under
improved conditions for each patient in the present study.
The same trained and experienced observer performed
all measurements using the same 8-Hz Doppler and
four Velcro-cuff (12 cm width and 2940 cm length)
sphygmomanometers wrapped on four extremities at
the same time. The environment was also arranged
specifically with a comfortable stretcher and two metal
armrests placed at 30 to ensure the comfort of the upper
extremities. Measurements were performed twice for each
vessel in two rounds and the mean value was used for each
vessel reading. Therefore, unfavorable factors that could
affect the result of ABI measurement were considered to
References
1. Selvin E, Erlinger TP. Prevalence of and risk factors for
peripheral arterial disease in the United States: results from
the National Health and Nutrition Examination Survey, 1999-
2000. Circulation 2004; 110: 738-743.
2. Bozkurt AK, Tasci I, Tabak O, Gumus M, Kaplan Y. Peripheral
artery disease assessed by ankle-brachial index in patients with
established cardiovascular disease or at least one risk factor for
atherothrombosis - CAREFUL study: a national, multi-center,
cross-sectional observational study. BMC Cardiovascular
Disorders 2011; 11: 4.
be minimized.
Given the decrease in the sensitivity of ABI measurement
in diabetics and the elderly population who frequently
have calcified vessels, ABI has been considered less
reliable in calcified and incompressible arteries leading to
falsely higher values (15). As an example, the sensitivity
of >90% in patients with PAD was reported to decrease
to 50% in a group with diabetic neuropathy and to 30%
in asymptomatic subjects (22,23). Although a high ABI is
common among diabetics and/or the elderly population,
these patients are considered to have the same risk as
those having a low ABI value, even when they have no
symptoms. Although the average age of our participants
was 59, diabetes mellitus was recorded in 40% of the whole
study group, but we detected no high ABI values in these
participants who had angiographically proven PAD. A
high ABI was also less common in our previous studies
on Turkish subjects compared to other populations (3).
However, whether ethnic differences in the rate of vessel
calcinosis exist has not been clearly reported to date.
Although an ABI value below 0.90 is currently accepted
as a diagnostic for the presence of PAD, a threshold
index value of 0.95 was found to improve the diagnostic
performance in some studies (24,25). Accordingly, in our
study population, the sensitivity of the ABI test increased
from 83.6% to 90.9% when the diagnostic cut off was set at
0.95. Whether some higher ABI threshold values provide
better diagnostic performance in Turkish subjects with
PAD needs to be further tested.
In conclusion, this study on Turkish subjects showed that
the sensitivity of a low ABI value to detect angiographically
proven PAD was in the acceptable range, with an impressive
positive predictive value. Moreover, using the value of ABI
0.95 as the diagnostic threshold markedly improved the
sensitivity ratio. Nevertheless, due to the low number of
participants without PAD as confirmed by angiography,
clearer specificity values could not be obtained.
3. Gezer M, Tasci I, Demir O, Acikel C, Cakar M, Saglam K,
Kutsi Kabul H, Fatih Bulucu M, Acar R, Ozturk K et al. Low
frequency of a decreased ankle brachial index and associated
conditions in the practice of internal medicine in a Turkish
population sample. Int Angiol 2012; 31:454-461.
4. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA,
Fowkes FG, Bell K, Caporusso J, Durand-Zaleski I, Komori K et
al. Inter-Society Consensus for the Management of Peripheral
Arterial Disease (TASC II). Eur J Vasc Endovasc 2007; 33: 1-75.
5
 DOAN et al. / Turk J Med Sci
5. McDermott MM, Greenland P, Liu K, Guralnik JM, Criqui
MH, Dolan NC, Chan C, Celic L, Pearce WH, Schneider JR
et al. Leg symptoms in peripheral arterial disease: associated
clinical characteristics and functional impairment. JAMA-J
Am Med Assoc 2001; 286: 1599-1606.
6. Tasci I. Best practice in ankle brachial index measurement. J
Wound Ostomy Cont 2012; 39: 238.
7. Hirsch AT, Criqui MH, Treat-Jacobson D, Regensteiner JG,
Creager MA, Olin JW, Krook SH, Hunninghake DB, Comerota
AJ, Walsh ME et al. Peripheral arterial disease detection,
awareness, and treatment in primary care. JAMA-J Am Med
Assoc 2001; 286: 1317-1324.
8. McDermott MM, Liu K, Criqui MH, Ruth K, Goff D, Saad
MF, Wu C, Homma S, Sharrett AR. Ankle-brachial index and
subclinical cardiac and carotid disease: the multi-ethnic study
of atherosclerosis. Am J Epidemiol 2005; 162: 33-41.
9. Haimovici H. Arteriographic patterns of atherosclerotic
occlusive disease of the lower extremity. In: Ascher E, editor.
Haimovicis Vascular Surgery. 6th ed. New York, NY, USA:
Wiley-Blackwell Publications; 2012. pp. 56-78.
10. Demir O, Tasci I, Acikel C, Saglam K, Gezer M, Acar R, Yildiz
B, Bulucu F, Kabul HK, Dogan MI et al. Individual variations
in ankle brachial index measurement among Turkish adults.
Vascular 2016; 24: 53-58.
11. Criqui MH, Denenberg JO. The generalized nature of
atherosclerosis: how peripheral arterial disease may predict
adverse events from coronary artery disease. Vasc Med 1998;
3: 241-245.
12. Doobay AV, Anand SS. Sensitivity and specificity of the ankle-
brachial index to predict future cardiovascular outcomes: a
systematic review. Arterioscl Throm Vas 2005; 25: 1463-1469.
13. Ota H, Takase K, Igarashi K, Chiba Y, Haga K, Saito H,
Takahashi S. MDCT compared with digital subtraction
angiography for assessment of lower extremity arterial
occlusive disease: importance of reviewing cross-sectional
images. Am J Roentgenol 2004; 182: 201-209.
14. Collins R, Burch J, Cranny G, Aguiar-Ibanez R, Craig D,
Wright K, Berry E, Gough M, Kleijnen J, Westwood M.
Duplex ultrasonography, magnetic resonance angiography,
and computed tomography angiography for diagnosis and
assessment of symptomatic, lower limb peripheral arterial
disease: systematic review. BMJ 2007; 334: 1257.
15. Hirsch AT, Haskal ZJ, Hertzer NR, Bakal CW, Creager
MA,Halperin JL,Hiratzka LF,Murphy WR,Olin JW,Puschett
JBet al. ACC/AHA guidelines for the management of patients
with peripheral arterial diseases. J Vasc Interv Radiol, 2006; 17:
1383-1397.
16. Endres HG, Hucke C, Holland-Letz T, Trampisch HJ. A new
efficient trial design for assessing reliability of ankle-brachial
index measures by three different observer groups. BMC
Cardiovascular Disorders 2006; 6: 33.
17. Niazi K, Khan TH, Easley KA. Diagnostic utility of the two
methods of ankle brachial index in the detection of peripheral
arterial disease of lower extremities. Catheter Cardio Inte 2006;
68: 788-792.
18. Schroder F, Diehm N, Kareem S, Ames M, Pira A, Zwettler
U, Lawall H, Diehm C. A modified calculation of ankle-
brachial pressure index is far more sensitive in the detection of
peripheral arterial disease. J Vasc Surg 2006; 44: 531-536.
19. Nam SC, Han SH, Lim SH, Hong YS, Won JH, Bae JI, Jo J.
Factors affecting the validity of ankle-brachial index in the
diagnosis of peripheral arterial obstructive disease. Angiology
2010; 61: 392-396.
20. Baker JD, Dix DE. Variability of Doppler ankle pressures with
arterial occlusive disease: an evaluation of ankle index and
brachial-ankle pressure gradient. Surgery 1981; 89: 134-137.
21. deGraaff JC, Ubbink DT, Legemate DA, de Haan RJ, Jacobs MJ.
Interobserver and intraobserver reproducibility of peripheral
blood and oxygen pressure measurements in the assessment of
lower extremity arterial disease. J Vasc Surg 2001; 33: 1033-
1040.
22. Williams DT, Harding KG, Price P. An evaluation of the efficacy
of methods used in screening for lower-limb arterial disease in
diabetes. Diabetes Care 2005; 28: 2206-2210.
23. Sahli D, Eliasson B, Svensson M, Blohme G, Eliasson M,
Samuelsson P, Ojbrandt K, Eriksson JW. Assessment of toe
blood pressure is an effective screening method to identify
diabetes patients with lower extremity arterial disease.
Angiology 2004; 55: 641-651.
24. Stoffers HE, Rinkens PE, Kester AD, Kaiser V, Knottnerus JA.
The prevalence of asymptomatic and unrecognized peripheral
arterial occlusive disease. Int J Epidemiol 1996; 25: 282-290.
25. Guo X, Li J, Pang W, Zhao M, Luo Y, Sun Y, Hu D. Sensitivity and
specificity of ankle-brachial index for detecting angiographic
stenosis of peripheral arteries. Circ J 2008; 72: 605-610.