Abstract
Background:
Given the lack of consistency in the literature regarding the reliability of the ankle-brachial index (ABI) as a valid screening tool and an independent risk indicator of cardiovascular events and mortality, we compared it with angiography as a reference standard test.
Methods:
This case-control study, conducted between 2010 and 2011 in Tehran Heart Center, recruited 362 angiographically confirmed cases of coronary artery disease (CAD) and 337 controls. A standard protocol was used to measure the ABI and different CAD risk factors.
Results:
A low ABI had specificity of 99.7%, positive predictive value of 95.8%, negative predictive value of 49.8%, sensitivity of 64%, likelihood ratio of 24.07, and odds ratio (OR) of 22.79 (95%CI: 3.06–69.76). The role of the associated risk factors was evaluated with OR (95%CI), with the variables including gender 3.15 (2.30–4.30), cigarette smoking 2.72 (1.86–3.99), family history 1.72 (1.17–2.51), diabetes 1.66 (1.15–2.4), and dyslipidemia 1.38 (1.02–1.88). In a multivariate model, the following variables remained statistically significantly correlated with CAD [OR (95%CI)]: ABI 13.86 (1.78–17.62); gender 3.69 (2.43–5.58); family history of CAD 2.18 (1.41–3.37); smoking 1.69 (1.08–2.64); age 1.04 (1.02–1.06).
Conclusions:
A low ABI had specificity of 99.7%; however, because of its low sensitivity (64%), we should consider CAD risk factors associated with a low ABI in order to use it as a first-line screening test.
Keywords: Ankle brachial index, Coronary artery disease, Risk assessment
Introduction
A low ankle-brachial index (ABI) is deemed a feasible, valid, and noninvasive indicator of atherosclerosis.1, 2 A low ABI suggests the development and severity of lower-limb peripheral arterial disease2 and can play a crucial role as an independent risk indicator of cardiovascular events and mortality.1 A reclassification of the cardiovascular risk assessment model to include the ABI led to the modification of the management of cardiovascular diseases in women (almost 36%) and in men (19%).1 The early detection and prevention of cardiovascular and cerebrovascular events require a precise identification of individual risks through the use of accurate indicators.
A low ABI (≤ 0.90) has been demonstrated to correlate with a twofold increase in cardiovascular mortality, ten-year mortality, and coronary events rate.1 A low ABI is considered a potential factor in the detection of symptomatic and asymptomatic patients afflicted with peripheral arterial disease and could, as such, be an accurate and effective screening test for peripheral arterial disease2 and coronary artery disease (CAD).3 The ABI might also provide more information regarding the prognosis of the long-term survival of patients with peripheral arterial disease.4
The ABI, alongside type-D personality, has been shown to predict the quality of life and depressive symptoms in patients with peripheral arterial disease.5 Both resting and post-exercise measures of the ABI were strong independent predictors of mortality6 in a cohort study of 3209 patients4 and increased risk of stroke or transient ischemic attack in the elderly in another study.7 There has been other evidence for a significant decline in the ABI after modifying the risk factors of CAD among a cohort of 5888 participants over 6 years of follow-up.8
Nonetheless, the reported evidence regarding the accuracy of the ABI suffers from a lack of consistency. This prospective investigation was, therefore, performed to evaluate whether the ABI could provide evidence for the prediction of CAD, and if so, to determine the accuracy of this indicator compared with coronary angiography.
Methods
A population of 699 was recruited in this case-control study with incidence cases (362) and controls (337) between 2010 and 2011 in Tehran Heart Center. The CAD cases were confirmed by angiographic findings, and the controls comprised patients with normal angiographic findings. The definition of obstructive CAD was decided as stenosis of 50% or more of the diameter of any coronary artery.9
For the measurement of risk factors, such well-known risk factors as age, gender, hypertension, diabetes, dyslipidemia, cigarette smoking, physical activity, body mass index, family history of CAD, and ABI were assessed via standardized methods.
The ABI measurements were done using a standard sphygmomanometer and a Doppler device, after 30 minutes of rest in a supine position. The detection of both upper- and lower- limb blood pressures was done by placing the blood pressure cuffs over the brachial arteries and malleoluses. For each right and left brachial artery and right and left dorsalis pedis artery, the cuff was inflated to 20 mmHg over the detectable systolic blood pressure and deflated appropriately. The ABI was calculated in accordance with the American Heart Association definition1 as the ratio of the higher systolic blood pressure of each ankle-brachial artery. The ABI was considered attenuated if the quotient was < 0.9. All the ABI measurements were performed by 2 well trained physicians. In this study, the ABI was treated as a categorical variable of an ABI ≤ 0.9 and an ABI > 0.9.
Univariate analysis was performed by applying the Pearson chi-squared test and the independent sample T-test as well as the estimation of the odds ratio and 95% confidence interval (95%CI) and measures of accuracy. The odds ratio, 95%CI, and likelihood ratio were calculated for different endpoints of CAD. Bivariate analysis was carried out by performing age- and sex-adjusted Mantel-Haenszel odds ratio and 95%CI. Multivariate analysis was conducted via the logistic regression model to evaluate multivariate adjusted relationships between CAD and independent variables and for adjusting potential confounders. Statistical comparisons were considered significant at a p value < 0.05. Measure of validity was done to assess the accuracy of the ABI (≤ 0.90) compared to angiographic results through the calculation of the sensitivity, specificity, and predictive values.
Results
This research was performed on 362 cases and 337 controls. Table 1 compares the different risk factors and systolic blood pressures of both extremities between the two groups of cases and controls.
Table 1.
Characteristics of the patients*
| Cases (n = 362) | Controls (n = 337) | P value | |
|---|---|---|---|
| Age (y) | 60.01±0.01 | 55.48±10.69 | < 0.001 |
| Gender | < 0.001 | ||
| Female | 106 (29.3) | 191 (56.7) | |
| Male | 256 (70.7) | 146 (43.3) | |
| Diabetes mellitus | 93 (25.7) | 58 (17.2) | 0.006 |
| Hypertension | 134 (37.0) | 117 (34.7) | 0.521 |
| Dyslipidemia | 111 (30.7) | 119 (35.5) | 0.034 |
| Cigarette smoking | 156 (43.1) | 47 (13.9) | < 0.001 |
| Family history of CAD | 88 (24.3) | 53 (15.7) | 0.005 |
| Physically active | 25 (6.9) | 37 (11.0) | 0.058 |
| Triglyceride (mg/dl) | 175.62±26.40 | 151.24±66.36 | 0.002 |
| LDL-C (mg/dl) | 111.42±47.45 | 108.35±40.58 | 0.368 |
| HDL-C (mg/dl) | 43.81±10.90 | 49.58±27.53 | < 0.001 |
| Total cholesterol (mg/dl) | 186.48±46.44 | 178.79±62.03 | 0.062 |
| Fasting blood glucose (mg/dl) | 115.18±62.51 | 104.04±30.78 | 0.003 |
| ABI | 1.06±0.12 | 1.09±0.08 | < 0.001 |
| Body mass index (kg/m2) | 27.36±3.90 | 28.29±5.08 | 0.007 |
| Right-hand SBP (mmHg) | 126.83±19.35 | 126.80±17.44 | 0.985 |
| Left-hand SBP (mmHg) | 126.46±19.02 | 125.94±16.71 | 0.706 |
| Right-leg SBP (mmHg) | 134.2±20.96 | 137.91±17.89 | 0.013 |
| Left-leg SBP (mmHg) | 134.59±21.35 | 138.5±18.17 | 0.011 |
Data are presented as mean±SD or n (%).
CAD, Coronary artery disease; LDL-C, Low-density lipoprotein; HDL-C, High-density lipoprotein; ABI, Ankle brachial index; SBP, Systolic blood pressure
The calculated measures of accuracy for the ABI (≤ 0.9) compared to the results of coronary angiography revealed sensitivity of 64%, specificity of 99.7%, positive predictive value of 95.8%, and negative predictive value of 49.8%.
The crude OR and 95%CI of the relationship between the studied risk factors and CAD associated with the likelihood ratio are presented in Table 2. Based on the likelihood ratio, we could strongly predict angiographically confirmed CAD with the presence of gender differences, cigarette smoking, age, and low ABI.
Table 2.
Comparison between case and controls based on predictive factors
| OR | 95% CI | Likelihood Ratios | P value | |
|---|---|---|---|---|
| Gender | ||||
| Male | 3.15 | 2.32–4.30 | 54.26 | < 0.001 |
| Female | 0.31 | 0.23–0.43 | 54.26 | < 0.001 |
| Age | 1.04 | 1.02–1.06 | 26.17 | < 0.001 |
| Dyslipidemia | 1.38 | 1.02–1.88 | 4.43 | 0.021 |
| Cigarette smoking | 2.72 | 1.86–3.99 | 28.60 | < 0.001 |
| Family history of CAD | 1.72 | 1.17–2.51 | 8.06 | 0.003 |
| ABI(≤ 0.90) | 19.31 | 3.06–69.76 | 24.07 | < 0.001 |
| FBS (≤ 126 mg/dl) | 1.66 | 1.15–2.40 | 8.84 | 0.003 |
| HDL-C (≤40 mg/dl) | 0.79 | 0.68–0.92 | 9.15 | 0.002 |
CAD, Coronary artery disease; ABI, Ankle brachial index; OR, Odds ratio; CI, Confidence interval, FBS, Fasting blood sugar, HDL-C, High-density lipoprotein
Crude and adjusted Mantel-Haenszel odds ratio and 95%CI [OR (95% CI)] for age- and sex-matched comparisons were as: for age under 60 years to age above 60 years, the comparison crude estimate was 0.07 (0.01–0.45) and the common adjusted parameter estimate was 2.65 (0.97–4.86).
The sex-matched estimate of the parameters demonstrated that the crude odds ratio for the male-to-female comparison was 3.15 (2.32–4.30) and that the common adjusted odds ratio was 3.69 (2.43–5.58).
A low ABI did not significantly predict the unstable angina pectoris in terms of an odds ratio and 95%CI of 2.94 (0.29–29.23) and a likelihood ratio of 0.97 (p value = 0.32).
Acute myocardial infarction could not be predicted by a low ABI with an odds ratio and 95%CI of 1.11 (0.97–1.28) and a likelihood ratio of 0.19 (p value = 0.65).
Non-ST-segment elevation myocardial infarction was of a prognostic value of 1.42 (1.05–1.96) and a likelihood ratio of 1.30 (p value = 0.25) with a low ABI.
A low ABI predicted stable angina pectoris with an odds ratio and 95%CI of 2.43 (2.17–2.70) and a likelihood ratio of 20.65 (p value = 0.001).
Table 3 shows the adjusted odds ratios and 95%CIs for the different variables derived from a logistic regression model that controlled the potential confounders, when the ABI was treated as a categorical variable and was considered abnormal at ≤ 0.9. According to the above multivariate results, an ABI ≤ 0.9 had a strong and significant prognostic value for predicting CAD (odds ratio = 13.86).
Table 3.
Adjusted OR and 95% CI for the relationship between different factors with CAD when ABI is categorical
| OR | 95% CI | P value (Wald test) | |
|---|---|---|---|
| Gender | |||
| Male | 3.69 | 2.43–5.58 | < 0.001 |
| Female | 0.27 | 0.17–0.41 | < 0.001 |
| Age | 1.04 | 1.02–1.06 | < 0.001 |
| Family history of CAD | 2.18 | 1.41–3.37 | < 0.001 |
| Cigarette smoking | 1.69 | 1.08–2.64 | 0.021 |
| ABI (≤ 0.9) | 13.86 | 1.78–17.62 | 0.012 |
| ABI (> 0.9) | 0.07 | 0.05–0.56 | 0.012 |
CAD, Coronary artery disease; ABI, Ankle-brachial index; OR, Odds ratio; CI, Confidence interval
Discussion
The results of our investigation demonstrated that the measures of the accuracy of a low ABI (≤ 0.90) had very high specificity (99.7%) and a positive predictive value of 95.8% but low sensitivity (64%) and a low negative predictive value of 49.8%. Furthermore, the ABI provided strongly significant information on the prediction of CAD [OR (95%CI): 13.86 (1.78–17.62)] after multivariate adjustment for the potential confounding variables.
There is no doubt that a low ABI (≤ 0.90) increases the risk of overall mortality, proportional cardiovascular mortality, and coronary events by about twofold.1 Cardiovascular risk assessment based on traditional CAD risk factors is deemed difficult, and the introduction of new noninvasive, feasible, and valid prognostic factors seems to be necessary. Therefore, in accordance with the guidelines published by the American Heart Association/ American College of Cardiology, the present study showed that the ABI had the potential to be implemented for the risk assessment of CAD.1, 10 Future cardiovascular events in people with traditional and novel risk factors, as well as in individuals with asymptomatic atherosclerosis, may be easily predicted by the ABI, which is feasible, noninvasive, and accurate. The ABI has been traditionally applied for the screening of peripheral arterial disease, Nowadays given that a resting attenuated ABI of ≤ 0.90 is associated with an increased risk of overall mortality of about 2–7 fold and cardiovascular mortality of 2–4 fold,6 it can be considered as a feasible, noninvasive, and accurate tool for the screening of cardiovascular events for generalized atherosclerosis. There is also reported evidence that supports the role of a low ABI in increasing the risk of cerebrovascular events, especially in the elderly. 7, 11, 12 Our results revealed that a resting attenuated ABI had low sensitivity (64%) but enormous specificity (99.7%) and positive predictive value (95.8%). This measure of high specificity implies that a low ABI, as a valid and trustworthy sensitive screening tool, might rule out the presence of CAD. Sensitivity of 64% demonstrates that a low ABI at rest, when considering the higher systolic blood pressure of the limbs, should improve. Considering other major CAD risk factors combined with a low ABI is suggested to have the potential to improve the sensitivity of this index. The second definition and calculation of the ABI based on the measurement of the lower systolic blood pressures of both dorsalis pedis and brachial arteries may also augment the ability of a low ABI to predict the existence of CAD.
The difference between our findings and those of a study previously published in 2008 could be explained by the discrepancy in the applied gold standard test for comparison. In our research, we considered angiographic findings as a gold standard in comparison with the ABI, whereas the older study3 compared the results of the ABI with single-photon emission computed tomography and electrocardiography.3
A low ABI has been found to be associated with other CAD risk factors, and it seems that the modification of modifiable CAD risk factors and declining ABI may cause a statistically significant decrease in the lower extremity arterial diseases.8 In order to evaluate the risk factors of a declining ABI, the Cardiovascular Health Study Cohort was conducted. The frequency of a decrease in the ABI over a follow-up period of 6.5 years was estimated at 9.5%. Predictive factors for a declining ABI included age, cigarette smoking, hypertension, diabetes, and high density lipoprotein cholesterol (HDL-C) level. We found out more features of prognostic factors. In the present study beyond the result of the previous research,8 we showed that gender, age, cigarette smoking, family history of CAD, and ABI ≤ 0.9, played a statistically significant role in the prediction of future CAD.
A low ABI accurately predicts the presence of peripheral arterial disease in both symptomatic and asymptomatic individuals and is also associated with the severity13, 14 and progression15 of peripheral arterial disease. These findings demonstrate that ABI has the potential to detect the peripheral arterial disease at an early stage.
Since a low ABI shows a strong positive correlation with peripheral arterial disease, it can predict cardiovascular and cerebrovascular mortality.16–19 Also, in concord with our findings, a low ABI could be considered as a predictive factor for such events. It also can be pivotal in the risk assessment for afflicted or healthy individuals. A valid and reliable model of risk factor assessment guides the management strategies, in both primary and secondary prevention.20
Conclusion
A low ABI has almost perfect specificity (99.7%) and positive predictive value of 95.8% but low sensitivity (64%). Therefore, in order for a low ABI to be considered a firstline screening test for the early detection of CAD, we need to factor in other CAD risk factors associated with the ABI. Our multivariate adjusted relationships provided strongly accurate information regarding the prediction of CAD with an odds ratio of 13.86 with a low ABI (≤ 0.90).
Acknowledgements
This research was supported by a grant from the Research Deputy Directorship of Tehran University of Medical Sciences.
References
- 1. Ankle Brachial Index Collaboration. Fowkes FG, Murray GD, Butcher I, Heald CL, Lee RJ, Chambless LE, Folsom AR, Hirsch AT, Dramaix M, deBacker G, Wautrecht JC, Kornitzer M, Newman AB, Cushman M, Sutton-Tyrrell K, Fowkes FG, Lee AJ, Price JF, d'Agostino RB, Murabito JM, Norman PE, Jamrozik K, Curb JD, Masaki KH, Rodríguez BL, Dekker JM, Bouter LM, Heine RJ, Nijpels G, Stehouwer CD, Ferrucci L, McDermott MM, Stoffers HE, Hooi JD, Knottnerus JA, Ogren M, Hedblad B, Witteman JC, Breteler MM, Hunink MG, Hofman A, Criqui MH, Langer RD, Fronek A, Hiatt WR, Hamman R, Resnick HE, Guralnik J, McDermott MM. Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality: a metaanalysis. JAMA 2008; 300: 197–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Espinola-Klein C, Rupprecht HJ, Bickel C, Lackner K, Savvidis S, Messow CM, Munzel T, Blankenberg S, AtheroGene Investigators Different calculations of ankle-brachial index and their impact on cardiovascular risk prediction. Circulation 2008; 118: 961–967. [DOI] [PubMed] [Google Scholar]
- 3. Enoch A, Ijeoma A. The role of ankle-brachial index as a screening test for coronary artery disease in the Hispanic population. South Med J 2008; 101: 1117–1120. [DOI] [PubMed] [Google Scholar]
- 4. Feringa HH, Bax JJ, Hoeks S, van Waning VH, Elhendy A, Karagiannis S, Vidakovic R, Schouten O, Boersma E, Poldermans D. A prognostic risk index for long-term mortality in patients with peripheral arterial disease. Arch Intern Med 2007; 167: 2482–2489. [DOI] [PubMed] [Google Scholar]
- 5. Aquarius AE, Denollet J, Hamming JF, Van Berge Henegouwen DP, De Vries J. Type-D personality and ankle brachial index as predictors of impaired quality of life and depressive symptoms in peripheral arterial disease. Arch Surg 2007; 142: 662–667. [DOI] [PubMed] [Google Scholar]
- 6. Feringa HH, Bax JJ, van Waning VH, Boersma E, Elhendy A, Schouten O, Tangelder MJ, van Sambeek MH, van den Meiracker AH, Poldermans D. The long-term prognostic value of the resting and postexercise ankle-brachial index. Arch Intern Med 2006; 166: 529–535. [DOI] [PubMed] [Google Scholar]
- 7. Murabito JM, Evans JC, Larson MG, Nieto K, Levy D, Wilson PW, Framingham Study The ankle-brachial index in the elderly and risk of stroke, coronary disease, and death: the Framingham Study. Arch Intern Med 2003; 163: 1939–1942. [DOI] [PubMed] [Google Scholar]
- 8. Kennedy M, Solomon C, Manolio TA, Criqui MH, Newman AB, Polak JF, Burke GL, Enright P, Cushman M. Risk factors for declining ankle-brachial index in men and women 65 years or older: the Cardiovascular Health Study. Arch Intern Med 2005; 165: 1896–1902. [DOI] [PubMed] [Google Scholar]
- 9. Patel MR, Peterson ED, Dai D, Brennan JM, Redberg RF, Anderson HV, Brindis RG, Douglas PS. Low diagnostic yield of elective coronary angiography. N Engl J Med 2010; 362: 886–895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hirsch AT, Haskal ZJ, Hertzer NR, Bakal CW, Creager MA, Halperin JL, Hiratzka LF, Murphy WR, Olin JW, Puschett JB, Rosenfield KA, Sacks D, Stanley JC, Taylor LM, Jr, White CJ, White J, White RA, Antman EM, Smith SC, Jr, Adams CD, Anderson JL, Faxon DP, Fuster V, Gibbons RJ, Hunt SA, Jacobs AK, Nishimura R, Ornato JP, Page RL, Riegel B, American Association for Vascular Surgery; Society for Vascular Surgery; Society for Cardiovascular Angiography and Interventions; Society for Vascular Medicine and Biology; Society of Interventional Radiology; ACC/AHA Task Force on Practice Guidelines Writing Committee to Develop Guidelines for the Management of Patients With Peripheral Arterial Disease; American Association of Cardiovascular and Pulmonary Rehabilitation; National Heart, Lung, and Blood Institute; Society for Vascular Nursing; TransAtlantic Inter-Society Consensus; Vascular Disease Foundation ACC/AHA 2005 Practice Guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic): a collaborative report from the American Association for Vascular Surgery/Society for Vascular Surgery, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, Society of Interventional Radiology, and the ACC/AHA Task Force on Practice Guidelines (Writing Committee to Develop Guidelines for the Management of Patients With Peripheral Arterial Disease): endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation; National Heart, Lung, and Blood Institute; Society for Vascular Nursing; TransAtlantic Inter- Society Consensus; and Vascular Disease Foundation. Circulation 2006; 113: e463–654. [DOI] [PubMed] [Google Scholar]
- 11. Zheng ZJ, Sharrett AR, Chambless LE, Rosamond WD, Nieto FJ, Sheps DS, Dobs A, Evans GW, Heiss G. Associations of anklebrachial index with clinical coronary heart disease, stroke and preclinical carotid and popliteal atherosclerosis: the Atherosclerosis Risk in Communities (ARIC) Study. Atherosclerosis 1997; 131: 115–125. [DOI] [PubMed] [Google Scholar]
- 12. Abbott RD, Rodriguez BL, Petrovitch H, Yano K, Schatz IJ, Popper JS, Masaki KH, Ross GW, Curb JD. Ankle-brachial blood pressure in elderly men and the risk of stroke: the Honolulu Heart Program. J Clin Epidemiol 2001; 54: 973–978. [DOI] [PubMed] [Google Scholar]
- 13. McDermott MM, Liu K, Greenland P, Guralnik JM, Criqui MH, Chan C, Pearce WH, Schneider JR, Ferrucci L, Celic L, Taylor LM, Vonesh E, Martin GJ, Clark E. Functional decline in peripheral arterial disease: associations with the ankle brachial index and leg symptoms. JAMA 2004; 292: 453–461. [DOI] [PubMed] [Google Scholar]
- 14. McDermott MM, Greenland P, Liu K, Guralnik JM, Celic L, Criqui MH, Chan C, Martin GJ, Schneider J, Pearce WH, Taylor LM, Clark E. The ankle brachial index is associated with leg function and physical activity: the Walking and Leg Circulation Study. Ann Intern Med 2002; 136: 873–883. [DOI] [PubMed] [Google Scholar]
- 15. McLafferty RB, Moneta GL, Taylor LM, Jr, Porter JM. Ability of ankle-brachial index to detect lower-extremity atherosclerotic disease progression. Arch Surg 1997; 132: 836–840. [DOI] [PubMed] [Google Scholar]
- 16. Mohler ER., 3rd Peripheral arterial disease: identification and implications. Arch Intern Med 2003; 163: 2306–2314. [DOI] [PubMed] [Google Scholar]
- 17. 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. [DOI] [PubMed] [Google Scholar]
- 18. Hertzer NR, Beven EG, Young JR, O'Hara PJ, Ruschhaupt WF, 3rd, Graor RA, Dewolfe VG, Maljovec LC. Coronary artery disease in peripheral vascular patients. A classification of 1000 coronary angiograms and results of surgical management. Ann Surg 1984; 199: 223–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Ness J, Aronow WS. Prevalence of coexistence of coronary artery disease, ischemic stroke, and peripheral arterial disease in older persons, mean age 80 years, in an academic hospital-based geriatrics practice. J Am Geriatr Soc 1999; 47: 1255–1256. [DOI] [PubMed] [Google Scholar]
- 20. Pearson TA. New tools for coronary risk assessment: what are their advantages and limitations? Circulation 2002; 105: 886–892. [DOI] [PubMed] [Google Scholar]