Author + information
- Received September 9, 1998
- Revision received March 29, 1999
- Accepted May 14, 1999
- Published online September 1, 1999.
- Karen A Burek, RN, MS∗,*,
- Kim Sutton-Tyrrell, DrPH1,†,
- Maria Mori Brooks, PhD†,
- Barbara Naydeck, MPH†,
- Norma Keller, MD‡,
- Mary Ann Sellers, RN, MSN§,
- Gary Roubin, MD, FACC∥,
- Růžena Jandová, MD, CSc¶ and
- Charanjit S Rihal, MD, FACC#
- ↵*Reprint requests and correspondence: Karen A. Burek, University of Pittsburgh, 127 Parran Hall, 130 DeSoto Street, Pittsburgh, Pennsylvania 15261
The purpose of this study was to evaluate the prevalence and prognostic importance of lower extremity arterial disease (LEAD) in patients with multivessel coronary artery disease.
The presence of clinically evident LEAD increases the risk of death in patients with known coronary artery disease. Because studies have lacked noninvasive measures of subclinical LEAD, the true prognostic importance of lower extremity atherosclerosis in this population has probably been underestimated.
Ankle blood pressures were measured in 405 consecutive patients with angiographically documented multivessel coronary disease from seven Bypass Angioplasty Revascularization Investigation (BARI) sites and a parallel study site within 3 years of enrollment. Lower extremity arterial disease was defined as an ankle/arm systolic blood pressure ratio of 0.90 or less.
Among patients studied, 69 (17%) had LEAD. These patients were more likely to be current smokers, treated for diabetes, older and present with unstable angina compared with patients without LEAD. Among patients who underwent coronary arterial bypass grafting, major complications occurred in 2.8% of those without LEAD compared with 20.7% of those with LEAD (p = 0.002). Five-year mortality rates were similar for symptomatic LEAD (14%) and asymptomatic LEAD (14%). Patients without LEAD had a 3% mortality. After adjusting for baseline differences, the relative risk of death was 4.9 times greater for patients with LEAD compared with those without (95% confidence interval [CI]: 1.8, 13.4, p < 0.01).
Patients with LEAD have a significantly higher risk of death than patients without LEAD, regardless of the presence of symptoms. An abnormal ankle/arm index is a strong predictor of mortality and can be used to further stratify risk among patients with multivessel coronary artery disease.
Evidence of peripheral atherosclerosis is strongly associated with long-term mortality in patients with coronary artery disease (1,2). Although these studies have demonstrated the importance of peripheral atherosclerosis in a population diagnosed with coronary disease, both studies lacked standard, noninvasive evidence of subclinical disease. Thus, the prognostic importance of peripheral atherosclerosis may have been underestimated.
It is well documented that the prevalence of coronary artery disease in patients with lower extremity arterial disease (LEAD) is high (3–5)and that the prevalence of LEAD is also increased in patients with acute coronary artery disease (1,6,7). In fact, cardiovascular and cerebrovascular disease and death occur more often in patients with LEAD (8,9). Criqui et al. (10)noted a high prevalence of cardiovascular events among those with peripheral arterial disease, even when analysis was controlled for the presence of risk factors such as smoking, diabetes, age and gender. Furthermore, these authors reported a four-fold excess risk of mortality in those with peripheral artery disease regardless of other cardiovascular risk factors. Similarly, among 1,440 subjects presenting with intermittent claudication studied by Boyd (11), nonfatal myocardial infarction (MI) occurred within five years in 20%.
Follow-up studies of patients with intermittent claudication report a higher mortality rate than that in the general population (8). However, symptoms of intermittent claudication alone have repeatedly been found to underestimate the prevalence of LEAD (5,10,12). Therefore, ankle blood pressure measurement, a simple, inexpensive and highly accurate method for detecting LEAD, has been used in many large epidemiologic studies. Atherosclerotic disease of the large vessels of the leg can be easily detected by obtaining an ankle blood pressure using a standard blood pressure cuff and a Doppler probe (10,13,14). Compared with angiography, an abnormal ankle/arm index (AAI) has been shown to have a sensitivity and specificity of 96% or higher (8).
The purpose of this study was to assess the prevalence and prognostic importance of LEAD in patients with angiographically documented coronary artery disease. Patients evaluated were enrolled in the Bypass Angioplasty Revascularization Investigation (BARI) study, and LEAD was evaluated using the AAI. Risk factors for LEAD are presented, and mortality rates are compared between those with and without LEAD.
The development and implementation of BARI have been fully described elsewhere (15,16). In brief, BARI is a multicenter study composed of both a randomized trial and two registries. The goal is to compare percutaneous transluminal coronary angioplasty (PTCA) to coronary artery bypass grafting (CABG) as treatment for coronary artery disease. Patients eligible for randomization had angiographically documented multivessel coronary disease with clinically severe angina or objective evidence of ischemia and were suitable candidates for CABG and for PTCA. Patients with a prior coronary revascularization procedure, single-vessel coronary disease, primary congenital, valvular or myocardial heart disease and of the ages less than 17 or greater than 80 years were excluded. Eligible patients who did not consent to randomization as well as a random sample of patients who were excluded based on angiographic grounds were followed in one of two registries. Patients in the randomized study were randomly assigned to CABG or PTCA, and patients in the two registries, in conjunction with their physicians, selected either CABG, PTCA or medical therapy. Between 1988 and 1993, 18 participating clinical centers enrolled 4,261 patients, of whom 1,829 were randomized, 2,010 were eligible patients who refused randomization and 422 were angiographically excluded patients. An additional 112 patients were enrolled into a parallel study conducted in Prague, Czech Republic; these patients consisted of 79 randomized, 11 eligible but not randomized and 22 angiographically excluded patients.
Patients enrolled in the LEAD substudy came from all components of BARI including the randomized trial, the eligible not randomized registry, the angiographically excluded registry and the Prague parallel study. The patient population and, therefore, the baseline and procedural data presented in this paper are different from data presented in other BARI papers (2,16).
Ankle blood pressures were measured within three years of enrollment between December 1992 and March 1994 in 405 consecutive patients from seven participating BARI sites and from the Prague parallel study. All subsets of patients have been followed for an average of 5.4 years from study entry. It is important to note that most participants were enrolled into the LEAD substudy approximately three years after the start of the BARI trial. This population consists of BARI trial participants who are three-year survivors. Thus, events begin for patients in the LEAD substudy at approximately year three.
Participating BARI centers followed a standard protocol for obtaining ankle blood pressures. Central training was provided and at least one observer from each participating clinical site was certified. Certified observers provided additional training at their clinical site as necessary.
During follow-up clinic visits, patients were placed in a supine position for at least 5 min. The systolic blood pressure of the right brachial artery and the posterior tibial artery of both ankles were measured using a standard blood pressure cuff and a Parks Model 841-A pocket Doppler probe (Parks Medical Electronics Inc., Aloha, Oregon). Systolic blood pressure was recorded as the cuff pressure when Doppler flow became detectable. Each blood pressure was performed twice and results were averaged. The ratio of ankle to arm systolic blood pressure was calculated for each leg and the lowest ratio (right leg/arm or left leg/arm) was recorded as the AAI. Based on validity studies for using this technique, lower extremity arterial disease was defined as an AAI of 0.09 or below.
Baseline characteristics and follow-up events for patients with LEAD were compared with those for patients without LEAD. Baseline comparisons were made using chi-square statistics for categorical variables and Student ttests or Wilcoxon nonparametric statistics for continuous variables. Due to the small number of in-hospital complications, Fisher’s exact tests were used to compare the in-hospital complication rates for those with and without LEAD. Survival curves were estimated by the Kaplan-Meier method (18)and were compared with the log rank test. A multivariate Cox proportional hazards model (19)was used to test the independent association between low AAI and mortality while adjusting for covariates that are related to AAI and mortality in this population. Variables included in this model were gender, history of smoking and a comorbidity scale accounting for history of MI, hypertension, congestive heart failure and treated diabetes (age was also considered in the model but it had no significant influence on mortality or on the relative risk of LEAD after controlling for the other variables listed). To evaluate the prognostic importance of LEAD with accompanying symptoms of intermittent claudication, a three way comparison involving symptomatic LEAD, asymptomatic LEAD and no LEAD was considered using Kaplan-Meier survival curves and the Cox regression analysis. Symptomatic LEAD was defined as having an AAI of 0.90 or less and reporting accompanying symptoms of intermittent claudication at baseline. Statistical significance was defined as a p value of 0.05 or less.
The population analyzed consisted of 405 patients: 302 (75%) randomized, 57 (14%) eligible but not randomized and 46 (11%) angiographically excluded patients. Three hundred sixteen (78%) patients were from the seven participating BARI sites and 89 (22%) were from the Prague parallel study. The distribution of the AAI measurements is shown in Figure 1. The median AAI was 1.0 with a range of 0.39 to 1.4. Of the 405 patients evaluated, 69 (17%) had an AAI of 0.9 or less.
Baseline characteristics for patients with and without LEAD are shown in Table 1. Patients with LEAD tended to be older, were more likely to have a history of treated diabetes, were more often current smokers and more often presented with symptoms of unstable angina. The proportion of patients with a history of hypertension and hypercholesterolemia was similar between those with LEAD and those without LEAD. The baseline angiographic characteristics between those with and without LEAD were quite similar with respect to the number of diseased vessels, the number of significant lesions and mean ejection fraction. Vascular disease other than coronary artery disease in the LEAD substudy patients at baseline is shown in Table 2. In general, other vascular disease was more prevalent among those with LEAD compared with those without LEAD. Thirty-one percent of the LEAD patients reported symptoms of intermittent claudication. Those with LEAD were more likely to have a history of stroke/TIA and evidence of carotid artery disease. However, a history of MI was more prevalent among those without LEAD. Finally, in this study, the proportion of patients with LEAD was similar within each of the three BARI study groups, within the U.S. and Prague and among the patients who received PTCA, CABG and medical therapy (p > 0.30 for all three comparisons).
One hundred ninety-eight patients in the LEAD study underwent angioplasty. Patients with LEAD had fewer lesions intended as well as fewer lesions attempted (Table 3). Among attempted lesions, there was no difference in angioplasty success rates between the two groups. There were no in-hospital deaths following the initial PTCA procedure in the LEAD study. Patients with and without LEAD had similar rates of major in-hospital complications (a combined end point of MI, stroke and coma) as well as similar rates of in-hospital emergency revascularization.
Among the 171 patients who underwent bypass surgery, major lesions were bypassed equivalently in both LEAD and non-LEAD groups (Table 3). It is interesting to note that patients with LEAD were more likely to have a calcified aorta (p = 0.017). Following the initial bypass surgery procedure, patients with LEAD had significantly higher rates of in-hospital major complications compared with those without LEAD (MI, stroke or coma: 20.7% vs. 2.8%; p = 0.0019); in the group of patients with LEAD, three patients had an MI, three had a stroke and none had a coma. There were no in-hospital deaths or emergency revascularizations following the initial CABG procedure in either group. Results were consistent across all strata when the analysis of complications was stratified by study group (randomized vs. registry) and by country (U.S. vs. Czech Republic).
For PTCA and CABG procedures combined, subsequent revascularization rates were 29% for both those with and without LEAD. When focusing on the PTCA group alone, repeat procedure rates were the same for those with LEAD versus those without LEAD (29%), but those with LEAD were slightly more likely to have subsequent CABG (22% LEAD, 19% no LEAD) and less likely to have subsequent PTCA (19% LEAD and 22% no LEAD).
During follow-up procedures, patients with LEAD had significantly lower survival rates compared with patients without LEAD (Figure 2). Five-year cumulative mortality rates were 14% for patients with LEAD and 3% for patients without LEAD (p < 0.001). The unadjusted relative risk of death for patients with LEAD compared with those without LEAD was 5.2 (95% confidence interval [CI] = 1.9, 13.8). After adjusting for baseline differences between the two groups, the relative risk of death was 4.9 times greater for patients with LEAD compared with those without (95% CI = 1.8, 13.4, p = 0.0022). Evaluating the symptomatic and asymptomatic LEAD groups separately, five-year mortality rates were 14% among those with symptomatic LEAD (n = 21) and 14% for those with asymptomatic LEAD (n = 48). In multivariate analysis, there was no significant difference between the symptomatic and asymptomatic LEAD groups regarding risk of death (p = 0.88). Compared with patients without LEAD, the adjusted relative risk of death was 4.4 (95% CI: 0.92, 21.4) for patients with symptomatic LEAD and 5.1 (95% CI: 1.7, 15.6) for patients with asymptomatic LEAD.
When using an ankle/arm systolic blood pressure ratio of 0.90 or less to define LEAD in patients with coronary disease, we found a relative risk of mortality 4.9 times higher for patients with LEAD compared with those without LEAD. This risk of mortality is substantially higher than that found in previous studies of patients with coronary artery disease (1,2). The greater relative risk found here is likely due to a more accurate measure of subclinical peripheral vascular disease with the use of the AAI. The striking association between the presence of LEAD and an increased incidence of major complications following bypass surgery was also of clinical significance. This association was also noted in the full BARI randomized and registry populations when peripheral atherosclerosis was identified by questionnaire (20).
This study supports the use of the AAI measurement for detection of asymptomatic LEAD even in patients with advanced atherosclerosis. Patients with LEAD had higher mortality rates regardless of the presence or absence of symptoms. Consequently, many patients who are at increased risk could easily be identified if ankle blood pressures were a routine part of the physical exam in patients being evaluated for coronary disease. A lack of symptoms in many of these patients is likely due to the inherent low physical activity of patients with existing coronary atherosclerosis and low activity level of older adults in general.
The AAI is simple, inexpensive and associated with no risk to patients. Thus, the use of this measure with other clinical cardiovascular risk assessment strategies can enhance primary and secondary risk management. This noninvasive diagnostic tool has been underutilized as a risk-stratification tool in routine practice. A compelling argument for the routine use of the AAI is the fact that among the 69 patients with LEAD, 21 were symptomatic and 48 were asymptomatic at baseline. Therefore, if the ankle blood pressures had not been done, 48 participants (70%) would not have had their lower extremity disease identified. Our data indicate that the risk associated with LEAD is similar whether or not symptoms of intermittent claudication are present. Of the 69 patients with LEAD, 35 (51%) had no other evidence of noncoronary atherosclerosis (based on the variables presented in Table 2). Thus, there are a relatively large number of patients in whom an abnormal ankle blood pressure would provide the only evidence of peripheral atherosclerosis.
At baseline, patients with LEAD were more likely to be current smokers, treated diabetic patients, older and presenting with unstable angina compared with patients without LEAD. The prevalence of LEAD among older persons with multivessel coronary artery disease is in accordance with previous studies (21,22). From a public health perspective, the high prevalence of LEAD in the older population is important. As the percentage of older people in the population increases, disease burden will increase the cost of medical care. In addition, coronary atherosclerosis is correlated with LEAD, and LEAD has been demonstrated to correlate with acute coronary artery disease (10). It is not surprising that this study, like other atherosclerotic studies, suggests that LEAD is strongly associated with cardiovascular risks, diabetes mellitus and current smoking (1,3,5,10). It is important for clinicians to provide treatment to slow the atherosclerotic process by eliminating or reducing modifiable risk factors. This should include a concerted effort to minimize disease progression with smoking cessation, regular exercise and control of glucose and lipid abnormalities.
Angiographic baseline comparisons were similar between those with and without LEAD. However, other vascular disease was associated with LEAD. A history of stroke or TIA and carotid vascular disease was present significantly more often among those with LEAD. These findings support the suggestion that peripheral atherosclerosis mirrors the general atherosclerotic process ongoing in the coronary and intracerebral vessels (23).
In summary, patients with coronary artery disease and associated LEAD are at significantly higher risk of death than patients without LEAD. Risk is significantly elevated, even when LEAD is asymptomatic. The simple nature of the AAI, a noninvasive diagnostic test, provides an inexpensive way to provide the clinician with additional information for cardiovascular risk stratification, including risk of complications during bypass surgery (17).
The authors are grateful to Dr. Lewis Kuller at the University of Pittsburgh, who provided funding for the Doppler equipment and the following study coordinators who collected ankle blood pressure measurements: Beth Hankin, RN and Mary Mazur, RN, Boston University; Mary Grogan, RN, Brown University; Mary Ann Sellers, RN, MSN and Laura Drew, RN, Duke University; Sylvia Matheson, RN, Mayo Clinic; Karen Burek, RN, MS, University of Michigan; Johanne Trudel, RN and Claudette Faille, Montreal Heart Institute; Linda Lazzam, RN, Toronto Hospital; and Ruzena Jandova, MD, CSc, Institute of Clinical and Experimental Medicine, Prague, Czech Republic.
↵1 The work of Dr. Sutton-Tyrrell was done during the tenure of an Established Investigatorship from the American Heart Association, Dallas, Texas.
☆ This study was supported by National Institutes of Health grants (HL38493, HL38504, HL38509, HL38512, HL38514-6, HL38518, HL38524-5, HL38529, HL38532, HL38556, HL38610, HL38642, and HL42145) from the National Heart, Lung and Blood Institute, Bethesda, Maryland.
- ankle/arm index
- Bypass Angioplasty Revascularization Investigation
- coronary arterial bypass grafting
- confidence interval
- lower extremity arterial disease
- myocardial infarction
- percutaneous transluminal coronary angioplasty
- Received September 9, 1998.
- Revision received March 29, 1999.
- Accepted May 14, 1999.
- American College of Cardiology
- Coronary Artery Surgery Study (CASS) Investigators,
- Eagle K.A,
- Rihal C.S,
- Foster E.D,
- et al.
- Fowkes F.G.R,
- Housley E,
- Cawood E.H.H,
- et al.
- Widmer L.K,
- Greensher A,
- Kannel W.B
- Cooke J.P,
- Ma A.O
- Boyd A.M
- Marinelli M.R,
- Beach K.W,
- Glass M.J,
- et al.
- Mantek N,
- Haenszel W
- Cox D.R
- Rihal C.S,
- Sutton-Tyrrell K,
- Schaff H.V,
- et al.
- Sutton-Tyrrell K,
- Alcorn H.G,
- Herzog H,
- et al.
- ↵Criqui MH, Coughlin SS, Fronek A. Noninvasively diagnosed peripheral arterial disease as a predictor of mortality: results from a prospective study. Circulation 1985;72:768–3.