Abstract
Background
The postexercise ankle‐brachial index (ABI) is useful in patients with suspected peripheral arterial disease (PAD) and a normal resting ABI. Our objective was to determine the independent predictors of an abnormal postexercise ABI.
Hypothesis
We hypothesized that the lowest ankle systolic pressure to calculate the resting ABI would be associated with an abnormal post‐exercise ABI.
Methods
Among 619 consecutive patients referred for suspected PAD, we calculated the postexercise ABI in patients with a normal resting ABI. An ABI <0.90 at rest was considered abnormal. We investigated 3 definitions of an abnormal postexercise ABI, defined as either <0.90, or >5% or >20% reduction compared with rest.
Results
Using multivariate analysis, the lowest ABI (calculated using the lowest and not the highest ankle systolic pressure) was consistently the most powerful independent predictor of an abnormal postexercise ABI. Patients with an abnormal lowest resting ABI were significantly more likely to have an abnormal postexercise ABI, as well as a significantly greater reduction in the ABI compared with rest. The lowest ABI had a high specificity (95%) but low sensitivity (82%) for a postexercise ABI <0.90.
Conclusions
An abnormal lowest ABI (calculated with the lowest ankle systolic pressure) is the most important independent predictor of an abnormal ABI response to exercise in patients with a conventionally normal ABI. All such patients should be exercised and their ABI measured postexercise.
Keywords: Exercise Testing, Ankle‐Brachial Index, Peripheral Arterial Disease
1. INTRODUCTION
Lower‐extremity peripheral arterial disease (PAD) is a major risk factor for future cardiovascular (CV) events and mortality.1, 2, 3, 4, 5, 6 A common clinical tool used to diagnose PAD is the ankle‐brachial index (ABI), which is calculated as the ratio of the highest systolic pressure in either the dorsalis pedis or posterior tibial artery to the highest systolic pressure in either the right or left brachial artery. The ABI has also been shown to be an independent predictor of CV events and mortality. An ABI of <0.90 has a sensitivity of 90% and specificity of 98% for the detection of a hemodynamically significant stenosis of ≥50% in the proximal lower limb.7, 8 Postexercise ABI measurement is recommended in patients with suspected PAD and a normal resting ABI.9, 10, 11
Our objective was to determine the independent predictors of an abnormal postexercise ABI. After determining that the lowest ABI (defined as the ABI calculated using the lowest ankle systolic pressure) was the most powerful independent predictor of an abnormal postexercise ABI, we sought to determine clinical usefulness of the lowest ABI in diagnosing PAD.
2. METHODS
2.1. Patient population
Between December 2, 2005, and October 27, 2016, there were 3948 patients who underwent 4854 PAD tests at our facility. We selected consecutive patients who were both eligible (ie, resting ABI ≥0.90 and clinical suspicion of PAD) and capable of undergoing exercise testing. After exclusions, there were 619 patients for the present analysis (Figure 1). There were 2297 patients with all 4 ABIs normal with no clinical indication for exercise testing; compared with patients who underwent exercise testing, these 2297 patients had higher resting ABIs, lower prevalence of smoking, higher prevalence of diabetes mellitus (DM), and lower prevalence of atypical leg pain and rest pain (see Supporting Information, Table 1, in the online version of this article). There were 449 patients who were eligible for exercise testing but were unable to exercise; compared with patients who subsequently underwent exercise testing, these 449 patients were older; had lower resting ABIs; had higher prevalence of dyslipidemia, hypertension, and DM; and had lower prevalence of smoking, lower prevalence of abnormal foot pulses, femoral bruit, claudication, and atypical leg pain (Figure 1 and Supporting Information, Table 1, in the online version of this article).
Figure 1.
Patient population from consecutive patients referred for suspicion of PAD. Reasons for not being able to exercise included inability to walk on treadmill (RN or MD deemed not capable of exercise, n =146), cannot walk on treadmill (n = 51), patient refusal (n = 49), cane or walker (n = 14), or medical reasons (angina, n = 129; COPD or asthma, n = 33; AF, n = 22; CHF, n = 5). Abbreviations: ABI, ankle‐brachial index; AF, atrial fibrillation; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; MD, medical doctor; PAD, peripheral arterial disease; RN, registered nurse
Table 1.
Baseline characteristics of the entire population and patients with a normal and abnormal postexercise ABI
| Entire Population, N = 619 | Normal Postexercise ABI ≥0.90, n = 371 | Abnormal Postexercise ABI <0.90, n = 248 | P Valuea | |
|---|---|---|---|---|
| Age, y | 64.2 ± 11.2 | 63.3 ± 12.5 | 65.6 ± 9.1 | <0.02 |
| Male sex | 396 (64) | 249 (67) | 150 (58) | <0.05 |
| Resting ABI, right leg | 1.11 ± 0.11 | 1.15 ± 0.09 | 1.00 ± 0.13 | <0.0001 |
| Resting ABI, left leg | 1.12 ± 0.11 | 1.14 ± 0.10 | 0.99 ± 0.11 | <0.0001 |
| Lowest ABI | 1.01 ± 0.13 | 1.05 ± 0.11 | 0.87 ± 0.13 | <0.0001 |
| Exercise time, s | 240.6 ± 86.8 | 260.6 ± 80.6 | 210.6 ± 89.5 | <0.0001 |
| Risk factors | ||||
| Dyslipidemia | 442 (71) | 253 (68) | 189 (76) | <0.05 |
| HTN | 422 (68) | 242 (65) | 180 (73) | 0.06 |
| Current smoker | 209 (34) | 95 (26) | 114 (46) | <0.0001 |
| DM | 178 (29) | 102 (27) | 76 (31) | 0.42 |
| Prior MI | 75 (12) | 47 (13) | 46 (19) | 0.05 |
| Prior PCI/CABG | 145 (23) | 80 (22) | 64 (26) | 0.24 |
| Prior stroke/TIA | 58 (9) | 34 (9) | 24 (10) | 0.89 |
| Erectile dysfunction | 51 (13) | 34 (14) | 17 (12) | 0.37 |
| Prior vascular surgery | 14 (2) | 2 (0.5) | 12 (4.8) | <0.0005 |
| Signs and symptoms | ||||
| Abnormal foot pulses | 263 (43) | 129 (35) | 128 (52) | <0.0001 |
| Femoral bruit | 231 (37) | 84 (23) | 147 (59) | <0.0001 |
| Claudication | 210 (34) | 86 (23) | 124 (50) | <0.0001 |
| Atypical leg pain | 106 (17) | 57 (15) | 49 (20) | 0.16 |
| Rest pain | 170 (27) | 79 (21) | 91 (37) | <0.0001 |
| Any lower‐limb ulceration | 4 (0.6) | 1 (0.3) | 3 (1.2) | 0.31 |
| Medications | ||||
| RAAS inhibition | 345 (56) | 191 (51) | 154 (71) | <0.02 |
| Antiplatelet | 392 (63) | 220 (59) | 172 (69) | <0.02 |
| OAC | 38 (6) | 27 (7) | 11 (4) | 0.17 |
| Statin | 394 (64) | 228 (61) | 166 (67) | 0.17 |
Abbreviations: ABI, ankle‐brachial index; CABG, coronary artery bypass grafting; DM, diabetes mellitus; HTN, hypertension; MI, myocardial infarction; OAC, oral anticoagulant; PCI, percutaneous coronary intervention; RAAS, renin‐angiotensin‐aldosterone system; SD, standard deviation; TIA, transient ischemic attack.
Data are presented as n (%) or mean ± SD.
For normal vs abnormal postexercise ABI.
2.2. Physiologic PAD testing
PAD testing was performed using the 4‐cuff method. Proximal and distal vessels were insonated using a 4‐MHz and 8‐MHz transducer, respectively, for continuous‐wave Doppler. A normal and abnormal ABI were defined as ≥0.90 and <0.90, respectively. The ABI was measured using the 8‐MHz continuous‐wave Doppler transducer. The Edinburgh questionnaire was used to determine the presence or absence of claudication.12 The examination of the peripheral pulses and auscultation for iliofemoral and femoral bruits were performed by a registered nurse (C.T.) specifically trained in vascular examination.
2.3. Exercise testing
Our PAD exercise protocol is a variation of the standard Carter protocol.13 The treadmill is programmed to a speed of 2 mph at an incline of 12% for 5 minutes. The exercise test is stopped either for symptoms or when the 5 minutes is completed. Postexercise, the brachial systolic pressure in the arm with the highest resting pressure is re‐measured, followed by both ankle pressures. The highest resting ABI is used when calculating the post‐ABI measurements. This sequence is repeated for a total of 5 postexercise measurements. The sequence is completed as quickly as practically possible. There is no electrocardiographic or blood pressure monitoring during the 5 minutes of exercise. We defined an abnormal postexercise ABI as one of the following: (1) postexercise ABI <0.90 based on current Canadian Cardiovascular Society guidelines11; (2) reduction in postexercise ABI >5% based on previously published report indicating that this is associated with increased CV events and mortality5; and (3) reduction in postexercise ABI >20% according to current American and European guidelines.9, 10
2.4. Statistical analysis
Multivariate analysis using stepwise linear regression was used to determine independent predictors of the postexercise ABI. We used the following variables in the linear regression model: age, sex, claudication, femoral bruit, abnormal foot pulses, lowest ABI (calculated using the lowest ankle systolic pressure), lowest toe‐brachial index, atypical leg pain, exercise time, smoking, hypertension, dyslipidemia, DM, previous coronary artery bypass grafting or percutaneous coronary intervention, previous peripheral revascularization (carotid or lower extremity), previous myocardial infarction, previous stroke or transient ischemic attack, erectile dysfunction, and medical treatment with warfarin or non–vitamin K antagonist oral anticoagulant, antiplatelet therapy (aspirin/clopidogrel/ticagrelor/prasugrel), statin, and renin angiotensin aldosterone system (RAAS) blockers (angiotensin‐converting enzyme inhibitor or angiotensin receptor blocker). The unpaired t test was used to detect differences between means. The Fisher exact test was used to detect differences between proportions. One‐way ANOVA with Tukey intercomparison testing was used to compare group data. Two‐way ANOVA was used to compare postexercise ABI in patients with a normal vs abnormal lowest resting ABI. Data were analyzed using MedCalc version 15.8 (MedCalc Software, Ostend, Belgium) and GraphPad Prism version 5.0 (GraphPad Software, La Jolla, CA). The level of significance was set at P < 0.05.
3. RESULTS
3.1. Patient population
There were 396 males and 223 females, with a mean age of 64.2 ± 11.2 years. The mean resting ABI was 1.12 ± 0.11 on the left and 1.11 ± 0.11 on the right, and the mean lowest ABI was 1.01 ± 0.13. Overall, there was a relatively high prevalence of risk factors for PAD (Table 1). The baseline characteristics of patients with a normal vs abnormal postexercise ABI defined as <0.90 are also shown in Table 1. Patients with an abnormal postexercise ABI were older, less likely to be female, had a lower resting ABI, had higher rates of abnormal foot pulses, femoral bruits, claudication, rest pain, lower exercise time, higher rates of smoking and dyslipidemia, higher rates of prior vascular surgery (carotid or lower extremity), and higher rates of use of antiplatelet agents and RAAS blockers.
3.2. Multivariate analysis
Independent predictors of an abnormal postexercise ABI are shown in Table 2, which describes the 3 different definitions of an abnormal postexercise ABI (postexercise ABI <0.90, postexercise reduction >5% and >20%). The lowest ABI was consistently the most powerful predictor of an abnormal postexercise ABI. Twenty‐six percent (n = 160) of patients had an abnormal lowest resting ABI.
Table 2.
Multivariate analysis of variables that predict an abnormal postexercise ABI according to a postexercise ABI <0.90, and reduction in the postexercise ABI >5% and >20%a
| Variable | T Value | P Value |
|---|---|---|
| Abnormal postexercise ABI <0.90 | ||
| Lowest ABI | –13.127 | <0.0001 |
| Femoral bruit | 6.125 | <0.0001 |
| Claudication | 3.282 | <0.002 |
| Current smoker | 2.318 | <0.02 |
| Previous PAD surgery | 2.894 | <0.004 |
| Exercise time | –2.661 | <0.008 |
| Reduction in postexercise ABI >5% | ||
| Lowest ABI | –9.079 | <0.0001 |
| Femoral bruit | 2.335 | <0.02 |
| Reduction in postexercise ABI >20% | ||
| Lowest ABI | –12.962 | <0.0001 |
| Antiplatelet agent | 2.656 | <0.009 |
| Anticoagulant | 2.271 | <0.03 |
Abbreviations: ABI, ankle‐brachial index; PAD, peripheral arterial disease.
N = 619.
3.3. Postexercise ABI in patients with an abnormal lowest resting ABI
There was significant correlation between the conventional and lowest resting ABI in the right leg (P < 0001, r 2 = 0.74; Figure 2A); similar data were found in the left leg (P < 0.0001, r 2 = 0.76; not shown). Given the lowest resting ABI was the most powerful independent predictor of an abnormal postexercise ABI, we sought to further characterize this group of patients. Patients with an abnormal lowest ABI had lower exercise time (203.4 ± 94.5 vs 253.5 ± 80.0, P < 0.0001), higher rates of smoking (41% vs 31%, P < 0.05), were more likely to have abnormal foot pulses (60% vs 30%, P < 0.0001), femoral bruits (58% vs 30%, P < 0.0001), claudication (54% vs 27%, P < 0.0001), rest pain (39% vs 24%), and more likely to be taking RAAS blockers (63% vs 53%, P < 0.05). ABI values at rest and postexercise for patients with a normal and abnormal lowest resting ABI are shown in Figure 2B. Patients with an abnormal lowest resting ABI had a greater reduction in the postexercise ABI compared with patients with a normal resting lowest ABI (P < 0.0001; Figure 2C). The proportions of patients with an abnormal postexercise ABI, defined as either a postexercise ABI <0.90 or a reduction >5% or 20%, are shown in Figure 2D. Overall, patients with an abnormal lowest resting ABI had a significantly higher proportion with an abnormal postexercise ABI (P < 0.0001).
Figure 2.
(A) Correlation between the conventional and lowest resting ABI in the right leg. (B) Resting and post‐exercise ABI measurements in patients with a normal and abnormal lowest resting ABI, calculated using the lowest ankle systolic pressure. (C) Percent reduction in ABI after exercise in patients with a normal vs. abnormal lowest resting ABI. (D) Proportion of patients with an abnormal post‐exercise ABI according to post exercise ABI <.90 or >5% or >20% reduction. Abbreviations: ABI, ankle‐brachial index; ANOVA, analysis of variance
3.4. Diagnostic accuracy of the lowest ABI to predict an abnormal postexercise ABI
The diagnostic accuracy of the lowest resting ABI is shown in Table 3. Overall, the lowest resting ABI had a relatively high specificity and low sensitivity for an abnormal postexercise ABI. The sensitivity and specificity of the lowest ABI for an abnormal postexercise ABI was highest (81.7% and 94.6%, respectively) when using the definition of postexercise ABI <0.90. The sensitivity of the lowest ABI for an abnormal postexercise ABI was lowest (46.0%) when the definition of an abnormal postexercise ABI was >5% reduction, and the specificity was lowest (83.4%) when using the definition of >20% reduction.
Table 3.
Diagnostic accuracy of the lowest resting ABI to predict an abnormal postexercise ABI (95% CI)
| Postexercise ABI <0.90 | >5% Reduction in ABI Postexercise | >20% Reduction in ABI Postexercise | |
|---|---|---|---|
| Sensitivity | 81.7 (77.2‐85.7) | 46.0 (30.9‐41.2) | 70.4 (60.8‐78.8) |
| Specificity | 94.6 (92.9‐96.0) | 86.1 (81.2‐90.0) | 83.4 (79.9‐86.5) |
| PLR | 15.15 (11.48‐19.98) | 2.58 (1.87‐3.57) | 4.23 (3.36‐5.32) |
| NLR | 0.19 (0.15‐0.24) | 0.74 (0.68‐0.82) | 0.36 (0.27‐0.48) |
| PPV | 84.8 (80.4‐88.5) | 75.9 (68.4‐82.1) | 47.2 (39.3‐55.2) |
| NPV | 93.4 (91.6‐94.9) | 52.6 (47.9‐57.3) | 93.0 (90.3‐95.1) |
Abbreviations: ABI, ankle‐brachial index; CI, confidence interval; NLR, negative likelihood ratio; NPV, negative predictive value; PLR, positive likelihood ratio; PPV, positive predictive value.
4. DISCUSSION
The present observational study demonstrates that in a population referred to an outpatient cardiology facility with a normal resting ABI, the lowest resting ABI (calculated using the lowest ankle systolic pressure) was the most powerful predictor of an abnormal postexercise ABI. An abnormal lowest ABI was also associated with a greater reduction in the postexercise ABI, which previously has been shown to be an independent predictor of lower‐limb revascularization14 and adverse outcomes3, 15, 16, 17 and provides additional prognostic information in patients referred for stress testing with suspected coronary disease.18
In terms of practicality, the lowest ABI had a variable sensitivity and specificity depending on the definition of an abnormal postexercise ABI. An abnormal lowest resting ABI was highly specific (ie, can rule in an abnormal postexercise ABI, defined as <0.90), whereas the relatively low sensitivity was not sufficient to rule out PAD or exclude patients with a normal resting lowest ABI from exercise testing. These data suggest that an abnormal lowest resting ABI should raise the suspicion that a patient may have an abnormal postexercise ABI. Conversely, a normal lowest resting ABI is not sensitive to rule out an abnormal postexercise ABI.
We analyzed the prevalence of an abnormal lowest ABI based on different recommendations. Current Canadian guidelines suggest that an abnormal postexercise ABI is <0.90,11 whereas the American Heart Association (AHA) recommends an abnormal postexercise ABI is a >30 mm Hg fall in the ankle systolic pressure or >20% fall compared with rest (IIa recommendation),10 and European guidelines indicate that a reduction >20% compared with rest is abnormal.9 The >20% reduction is based on data from a study by Ouriel et al,7 which used normal patients age <30 years to define a normal postexercise ABI as <5% decrease compared with rest, whereas the abnormal value of >20% reduction was from patients with claudication and a mean resting ABI of 0.60. However, others have shown that in patients with a normal resting ABI >0.90 there is an increased risk of cardiac and all‐cause mortality in patients with a postexercise ABI reduction 6% to 24%.5 In our study, >47% and >76% of patients with a normal and abnormal resting lowest ABI, respectively, had a postexercise reduction >5%, which suggests that defining an abnormal postexercise ABI reduction >20% may exclude a substantial proportion of patients with underlying PAD or increased CV risk.
It has previously been reported that the lowest ABI confers improved sensitivity at the expense of lower specificity for ≥50% stenosis confirmed with angiography19 or ≥70% using duplex ultrasound.20 In patients referred for coronary angiography for chest pain, an abnormal lowest ABI was associated with similar rates of event free‐survival compared with patients with a conventionally abnormal ABI, thereby identifying more patients at increased risk of CV events.21
We are unaware of prior investigations regarding possible interventions for risk reduction in patients with a normal resting ABI and abnormal postexercise ABI, nor in patients with an abnormal postexercise ABI. The Aspirin for Asymptomatic Atherosclerosis (AAA) trial was a double‐blind, randomized, placebo‐controlled trial that evaluated the effect of once‐daily 100 mg aspirin in 3350 asymptomatic patients in Scotland with no history of CV disease and resting ABI ≤0.95 using the lowest ankle systolic pressure.22 The trial did not meet the primary endpoint (reduced fatal or nonfatal coronary event, stroke, or revascularization), whereas aspirin treatment increased the risk of major hemorrhage. It is important to highlight that this was a primary prevention trial in low risk, asymptomatic patients screened from the general population, and it suggests that screening for PAD is not useful in the absence of an intervention to reduce risk. Conversely, an observational study from Spain with 5480 asymptomatic patients with an ABI ≤0.95 found that statin use was associated with a 20% relative risk reduction in major adverse CV events and mortality (it is unclear which ankle pressure was used in this study, as the ABI was retrospectively taken from measurements in the offices of general practitioners).23 In contrast to the AAA trial, the Spanish study consisted of patients with a high prevalence of risk factors for CV events and mortality, similar to the patients in our study.
4.1. Caution in using the lowest ABI to diagnose PAD
The lowest ABI identifies more patients with abnormal lower‐limb perfusion. Our facility has performed ABI testing in 3641 patients. Use of the lowest ABI increases the prevalence of PAD from 21.2% to 27.7%. The Multi‐Ethnic Study of Atherosclerosis (MESA) has shown that using the lowest ABI increases the prevalence of PAD from 3.7% to 14.6% in females and 3.4% to 9.3% in males.24 The difference in baseline prevalence is likely because that MESA consists of asymptomatic people in the general population, whereas our population was referred for suspicion of PAD or was screened because of a high prevalence of PAD risk factors (age >70 years, DM, or smokers age 50–69 years). One major concern with the use of the lowest ABI to diagnose PAD is overdiagnosis and overtreatment, with potential imbalance between risk of side effects of treatment and benefit in terms of prevention.
Nonetheless, PAD is an underdiagnosed and poorly managed disease, and it has been estimated that only one‐quarter of patients with PAD receive guideline‐recommended therapy.25 A recent study found that patients with PAD exhibit more extensive coronary atherosclerosis, which may contribute to the increased CV morbidity and mortality observed in patients with PAD.26 These data underscore the need for aggressive risk‐factor modification in patients with PAD.
4.2. Study limitations
We have not correlated these data with hard outcomes. Future studies are needed to determine if use of the lowest ankle pressure to calculate the ABI aids in predicting patients at increased risk of CV morbidity and mortality, including lower‐limb revascularization and amputation.
5. CONCLUSION
Among patients eligible and able to undergo exercise testing with a normal conventional resting ABI, 15% have an abnormal resting lowest ABI. The lowest ABI is the most powerful independent predictor for an abnormal postexercise ABI and has a high specificity but lower sensitivity for an abnormal postexercise ABI <0.90.
Supporting information
Table S1. Baseline characteristics of the population (N=619) and patients who were eligible for exercise testing but unable to exercise (N=449) and patients with no clinical indication for exercise testing (N=2,297).
ACKNOWLEDGMENTS
The authors express special thanks to Drs Brouillard and Jurt for their continued clinical support and cooperation.
Conflicts of interest
The authors declare no conflicts of interest.
Armstrong DWJ, Tobin C, Matangi MF. Predictors of an abnormal postexercise ankle brachial index: Importance of the lowest ankle pressure in calculating the resting ankle brachial index. Clin Cardiol. 2017;40:1163–1168. 10.1002/clc.22808
REFERENCES
- 1. Criqui MH, Langer RD, Fronek A, et al. Mortality over a period of 10 years in patients with peripheral arterial disease. N Engl J Med. 1992;326:381–386. [DOI] [PubMed] [Google Scholar]
- 2. Diehm C, Allenberg JR, Pittrow D, et al; German Epidemiological Trial on Ankle Brachial Index Study Group. Mortality and vascular morbidity in older adults with asymptomatic versus symptomatic peripheral artery disease. Circulation. 2009;120:2053–2061. [DOI] [PubMed] [Google Scholar]
- 3. Heald CL, Fowkes FG, Murray GD, et al. Risk of mortality and cardiovascular disease associated with the ankle‐brachial index: systematic review. Atherosclerosis. 2006;189:61–69. [DOI] [PubMed] [Google Scholar]
- 4. Resnick HE, Lindsay RS, McDermott MM, et al. Relationship of high and low ankle brachial index to all‐cause and cardiovascular disease mortality: the Strong Heart Study. Circulation. 2004;109:733–739. [DOI] [PubMed] [Google Scholar]
- 5. Feringa HH, Bax JJ, van Waning VH, et al. The long‐term prognostic value of the resting and postexercise ankle‐brachial index. Arch Intern Med. 2006;166:529–535. [DOI] [PubMed] [Google Scholar]
- 6. Doobay AV, Anand SS. Sensitivity and specificity of the ankle–brachial index to predict future cardiovascular outcomes: a systematic review. Arterioscler Thromb Vasc Biol. 2005;25:1463–1469. [DOI] [PubMed] [Google Scholar]
- 7. Ouriel K, McDonnell AE, Metz CE, et al. A critical evaluation of stress testing in the diagnosis of peripheral vascular disease. Surgery. 1982;91:686–693. [PubMed] [Google Scholar]
- 8. Yao ST, Hobbs JT, Irivne WT. Ankle systolic pressure measurements in arterial disease affecting the lower extremities. Br J Surg. 1969;56:676–679. [DOI] [PubMed] [Google Scholar]
- 9. Tendera M, Aboyans V, Bartelink ML, et al; ESC Committee for Practice Guidelines . ESC Guidelines on the diagnosis and treatment of peripheral artery diseases. Eur Heart J. 2011;32:2851–2906. [DOI] [PubMed] [Google Scholar]
- 10. Aboyans V, Criqui MH, Abraham P, et al; American Heart Association Council on Peripheral Vascular Disease . Measurement and interpretation of the ankle‐brachial index: a scientific statement from the American Heart Association [published correction appears in Circulation. 2013;127:e264]. Circulation. 2012;126:2890–2909. [DOI] [PubMed] [Google Scholar]
- 11. Abramson BL, Huckell V, Anand S, et al; Canadian Cardiovascular Society . Canadian Cardiovascular Society Consensus Conference: peripheral arterial disease—executive summary. Can J Cardiol. 2005;21:997–1006. [PubMed] [Google Scholar]
- 12. Leng GC, Fowkes FG. The Edinburgh Claudication Questionnaire: an improved version of the WHO/Rose Questionnaire for use in epidemiological surveys. J Clin Epidemiol. 1992;45:1101–1109. [DOI] [PubMed] [Google Scholar]
- 13. Creager MA, Beckman JA, Loscalzo J. Vascular Medicine: A Companion to Braunwald's Heart Disease. 2nd ed Oxford, UK: Elsevier Health Sciences; 2012. [Google Scholar]
- 14. Hammad TA, Strefling JA, Zellers PR, et al. The effect of post‐exercise ankle‐brachial index on lower‐extremity revascularization. JACC Cardiovasc Interv. 2015;8:1238–1244. [DOI] [PubMed] [Google Scholar]
- 15. de Liefde II, Klein J, Bax JJ, et al. Exercise ankle‐brachial index adds important prognostic information on long‐term outcome only in patients with a normal resting ankle‐brachial index. Atherosclerosis. 2011;216:365–369. [DOI] [PubMed] [Google Scholar]
- 16. de Liefde II, Verhagen HJM, Stolker RJ, et al. The value of treadmill exercise test parameters together in patients with known or suspected peripheral arterial disease. Eur J Prev Cardiol. 2012;19:192–198. [DOI] [PubMed] [Google Scholar]
- 17. Sheikh MA, Bhatt DL, Li J, et al. Usefulness of postexercise ankle‐brachial index to predict all‐cause mortality. Am J Cardiol. 2011;107:778–782. [DOI] [PubMed] [Google Scholar]
- 18. Narula A, Benenstein RJ, Duan D, et al. Ankle‐brachial index testing at the time of stress testing in patients without known atherosclerosis. Clin Cardiol. 2016;39:24–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. 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 Cardiovasc Interv. 2006;68:788–792. [DOI] [PubMed] [Google Scholar]
- 20. Schröder F, Diehm N, Kareem S, et al. 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. [DOI] [PubMed] [Google Scholar]
- 21. Espinola‐Klein C, Rupprecht HJ, Bickel C, et al. Different calculations of ankle‐brachial index and their impact on cardiovascular risk prediction. Circulation. 2008;118:961–967. [DOI] [PubMed] [Google Scholar]
- 22. Fowkes FG, Price JF, Stewart MC, et al; Aspirin for Asymptomatic Atherosclerosis Trialists. Aspirin for prevention of cardiovascular events in a general population screened for a low ankle brachial index: a randomized controlled trial. JAMA. 2010;303:841–848. [DOI] [PubMed] [Google Scholar]
- 23. Ramos R, García‐Gil M, Comas‐Cufí M, et al. Statins for prevention of cardiovascular events in a low‐risk population with low ankle brachial index. J Am Coll Cardiol. 2016;67:630–640. [DOI] [PubMed] [Google Scholar]
- 24. Allison MA, Aboyans V, Granston T, et al. The relevance of different methods of calculating the ankle‐brachial index: the Multi‐Ethnic Study of Atherosclerosis. Am J Epidemiol. 2010;171:368–376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Bhatt DL, Steg PG, Ohman EM, et al. International prevalence, recognition, and treatment of cardiovascular risk factors in outpatients with atherothrombosis. JAMA. 2006;295:180–189. [DOI] [PubMed] [Google Scholar]
- 26. Hussein AA, Uno K, Wolski K, et al. Peripheral arterial disease and progression of coronary atherosclerosis. J Am Coll Cardiol. 2011;57:1220–1225. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. Baseline characteristics of the population (N=619) and patients who were eligible for exercise testing but unable to exercise (N=449) and patients with no clinical indication for exercise testing (N=2,297).