The prognostic value of symptomatic peripheral arterial disease (PAD) in patients with coronary heart disease (CHD) is well documented, but few reports differentiating between symptomatic and asymptomatic forms of PAD are available. We investigated the respective prognostic effect of clinical and subclinical PAD on long-term all-cause mortality in patients with stable CHD. We analyzed 710 patients with stable CHD referred for hospitalization for CHD evaluation and management. As a part of the study, they completed questionnaires on medical history, underwent a standardized clinical examination, including ankle-brachial index (ABI) measurement, and provided a fasting blood sample. Three groups of patients were individualized: no PAD (no history of PAD and ABI >0.9 but ≤1.4); subclinical PAD (no history of PAD but abnormal ABI [i.e., ≤0.9 or >1.4); and clinical PAD (history of claudication, peripheral arterial surgery, or amputation due to PAD). Clinical and subclinical PAD was present in 83 (11.7%) and 181 (25.5%) patients, respectively. After a median follow-up of 7.2 years, 130 patients died. On multivariate analysis adjusted for age, hypertension, diabetes, dyslipidemia, smoking, left ventricular ejection fraction, CHD duration, heart rate, history of stroke or transient ischemic attack, and coronary revascularization, previous clinical PAD (hazard ratio 2.11, 95% confidence interval 1.28 to 3.47) and subclinical PAD (hazard ratio 1.65, 95% confidence interval 1.11 to 2.44) were significantly associated with increased all-cause mortality. In conclusion, our study has demonstrated that the detection of subclinical PAD by ABI in patients with stable CHD provides additional information for long-term mortality risk evaluation.
The ankle-brachial index (ABI) is a marker of subclinical peripheral arterial disease (PAD). Because only a few studies have previously taken into account both clinical PAD and ABI regarding the determination of the long-term prognosis of patients with coronary heart disease (CHD), the respective effect of subclinical PAD and symptomatic forms on mortality needs to be clarified. We hypothesized that in addition to the clinical forms, subclinical PAD would also be a long-term prognostic factor in patients with CHD, although its effect might possibly be less pronounced. The aim of the present study was to determine the risk of mortality according to the presence of clinical or subclinical PAD in a cohort of patients with stable CHD, representative of daily clinical practice and followed up for a median period of 7 years.
Methods
From 2001 to 2004, we prospectively and nonconsecutively recruited 791 male patients aged 45 to 74 years, living in the Toulouse area (Southwestern France), admitted to the cardiology department of the Toulouse University Hospital, and referred for evaluation and management of CHD. The subjects were patients with stable CHD hospitalized for cardiovascular examination. Stable CHD was defined by a history of acute coronary syndrome, a history of coronary revascularization, documented myocardial ischemia, stable angina, or the presence on the coronary angiogram of a coronary stenosis of ≥50%. Patients who had presented with an acute coronary episode during the previous 7 days were not included in the study because they were considered to not have stable CHD. Those with a history of cancer were not included because their life expectancy was potentially reduced owing to their neoplasm. The Genetique et ENvironnement en Europe du Sud (GENES) study was initially a case-control study conducted at the Toulouse University Hospital and designed to assess the role of gene–environment interactions in the occurrence of CHD. In the present analysis, we only studied the case patients (i.e., those with CHD). For these patients, we obtained the vital status after a median delay of 7.2 years after inclusion. The local ethics committee (file no. 1–99–48, 22/2/2000) approved the study protocol, in agreement with the French law on human biomedical research and the Declaration of Helsinki. All participants provided written informed consent attesting they had received all the information they needed about the study and that they agreed to participate.
Vital status (and date of death when applicable) was obtained on December 31, 2009 for each participant through the French national database, which records all deaths occurring in French subjects. Our end point was all-cause mortality. Age, socioeconomic variables, educational level, and information on cardiovascular risk factors were collected through standardized face-to-face interviews performed by a single physician. Smoking status was classified as smokers (current smokers or smokers who had quit for <3 years) and nonsmokers, and tobacco consumption was quantified in pack-years. The medical history was collected through the standardized questionnaire but was also checked in the patients’ medical files for chronic obstructive pulmonary disease, previous CHD, previous percutaneous intervention or coronary artery bypass grafting, duration of CHD in the case of previous CHD, previous PAD, previous stroke, and transient ischemic attack. The medications at discharge were also considered.
A physical examination was performed by a single physician using standard measurement methods. The anthropometric measurements, including height and body weight, were performed according to standardized procedures. The body mass index was determined as the weight divided by the height squared (kg/m 2 ). Blood pressure and heart rate were measured with an automatic sphygmomanometer (OMRON 705 CP, Omron Corporation, Kyoto, Japan) using a cuff fitted to the upper arm perimeter. The right arm blood pressure was measured twice and the average value recorded for the analysis. Measurements were performed after a minimum of 5 minutes of rest. Elevated blood pressure was defined as a history of hypertension or blood pressure of ≥160/95 mm Hg during the physical examination.
A fasting blood sample was taken for the measurement of plasma lipids, glucose, and creatinine. Low-density lipoprotein cholesterol was determined by the Friedewald formula when triglycerides were below 4.6 mmol/L (400 mg/dl). Patients were considered to have diabetes if they required oral antidiabetic drugs and/or insulin or if their blood glucose level was ≥7 mmol/L (≥126 mg/dl). They were considered to have dyslipidemia if they required lipid-lowering treatment or if the low-density lipoprotein cholesterol was ≥4.1 mmol/L (≥160 mg/dl). Renal function was assessed using the Cockcroft and Gault formula. Severe chronic renal failure was defined by a glomerular filtration rate <30 ml/min.
The left ventricular ejection fraction (LVEF) was assessed by ventriculography, using an isotopic method, or by echocardiography when ventriculographic measurements were lacking. Coronary stenoses of ≥50% were considered significant. Diffusion of CHD lesions was assessed by calculating the Gensini score.
The lower limb blood pressure was determined with the patient in the supine position from the right and left posterior tibial arteries. When the posterior tibial artery blood pressure was not measurable, the dorsalis pedis artery was used. All measurements were performed using an appropriately sized cuff. The systolic blood pressure was detected with a hand-held Doppler probe. The ABI was calculated for each lower limb by dividing the ankle systolic blood pressure by the average of the 2 systolic blood pressure measurements performed on the arm. For each patient, the lowest ABI recorded in the 2 ankles was kept for additional analysis. An ABI ≤0.9 or >1.4 was considered abnormal (given that it has been demonstrated that a high ABI has the same poor prognostic significance as a low ABI ).
For the present analysis, 3 groups of patients were individualized: no PAD, subclinical PAD, and clinical PAD. Clinical PAD was defined as a documented medical history of symptomatic PAD in the medical file of the patient (history of claudication, peripheral arterial surgery, or amputation due to PAD). Subclinical PAD was defined by a lack of a history of PAD but an abnormal ABI (ABI ≤0.9 or >1.4). Patients with no PAD were those without any history of PAD and a normal ABI (ABI >0.9 but ≤1.4).
Statistical analysis was performed using Stata statistical software, release 11.1 (StataCorp, College Station, Texas). Continuous variables were summarized as the mean ± standard deviation for normal distributions and as the median and interquartile range when distributions were not normally distributed. Categorical variables are presented as proportions. On univariate analysis, categorical variables were compared using the chi-square test (or Fischer’s exact test when necessary). Student’s t test or analysis of variance was used to compare the distribution of continuous normally distributed data according to categorical variables. The Mann-Whitney U test and Kruskal-Wallis test were used to compare the ranges of continuous non-normally distributed variables according to the categorical variables. A p value <0.05 was considered statistically significant.
The cumulative survival of patients with no, subclinical, or clinical PAD was determined using the Kaplan-Meier method and compared using the log-rank test. Univariate and multivariate Cox regression models were used to investigate the association between variables and mortality during the follow-up and to determine the hazard ratios for mortality and 95% confidence intervals. Regressions with polynomial terms (quadratic and cubic) were performed to examine for possible nonlinear relationships among heart rate, smoking (quantified in pack-years), CHD duration, and mortality risk. Finally, these variables were analyzed as continuous variables because the second- and third-order polynomial terms did not bring significant information.
All variables associated with a p value of <0.20 on univariate analysis were introduced in a multivariate Cox model. A backward procedure was applied to assess the variables that were significantly and independently associated with mortality (p <0.05). However, cardiovascular risk factors (e.g., high blood pressure, diabetes, tobacco consumption, and dyslipidemia) were kept in the multivariate Cox analysis even if they were not significantly associated with mortality, given that these variables are well-described prognostic factors in the published data. The proportional-hazard assumption was tested for each covariate using the “log-log” method, plotting (− ln [− ln (survival)]) for each category of a nominal covariate versus ln (analysis time). None of the assumptions could be rejected. The performance of the Cox model was determined by calculating the Harrell’s c-index to assess discrimination and the net reclassification index and the integrated discrimination improvement to assess reclassification after including clinical and subclinical PAD as explanatory variables. Three categories of predicted risk of mortality were considered: 0% to 5%; 5% to 15%; and ≥15%. Reclassification to a higher risk group was considered improvement in reclassification for patients who died during follow-up and reclassification downward was considered failure. In contrast, for patients who did not die, reclassification upward was considered disadvantageous and reclassification downward, advantageous.
Results
We prospectively included 791 men aged 45 to 74 years presenting with CHD. Eight subjects were excluded because they had a history of cancer. Of the 783 remaining patients, 73 were excluded because of missing data on the main interest variables. The excluded patients’ characteristics did not differ significantly from those of the patients remaining in the study. Vital status could be obtained on December 31, 2009 for all patients. During the follow-up period, 130 patients died. Follow-up ranged from 5.7 to 8.6 years among those living on December 31, 2009. The median follow-up was 7.2 years. Of the 710 patients remaining in the final analysis, 446 (62.8%) did not have PAD and 264 (37.2%) had subclinical (181 [25.5%]) or clinical PAD (83 [11.7%]). More than 1 (57.8%) in 2 patients presented with a medical history of CHD diagnosed before the index hospitalization. The baseline patient characteristics according to their peripheral arterial status are listed in Table 1 .
| Variable | All (n = 710) | PAD | p Value ⁎ | ||
|---|---|---|---|---|---|
| None (n = 446) | Subclinical (n = 181) | Clinical (n = 83) | |||
| Age (years) | NS | ||||
| Median | 60.9 | 60.2 | 61.2 | 63.1 | |
| Interquartile range | 53.3–67.5 | 52.9–67 | 54–67.7 | 55.5–68.6 | |
| High blood pressure | 383 (53.4%) | 230 (51.6%) | 96 (53%) | 57 (69%) | 0.01 † |
| Systolic blood pressure (mm Hg) | 139 ± 21 | 137 ± 20 | 139 ± 21 | 144 ± 22 | 0.03 † |
| Diastolic blood pressure (mm Hg) | 84 ± 11 | 85 ± 11 | 83 ± 11 | 85 ± 11 | NS |
| Dyslipidemia † | 515 (72.5%) | 316 (70.9%) | 136 (75.1%) | 63 (76%) | NS |
| Diabetes mellitus | 185 (26.1%) | 108 (24.2%) | 51 (28.2%) | 26 (31%) | NS |
| Current smoker or quit <3 years | 248 (34.9%) | 136 (30.5%) | 76 (42%) | 36 (43%) | 0.005 † |
| Body mass index (kg/m 2 ) | 27.4 ± 4 | 27.7 ± 4 | 27.2 ± 4.1 | 26.8 ± 3.8 | NS |
| Heart rate (beats/min) | 64 ± 12 | 63 ± 11 | 65 ± 13 | 68 ± 15 | <0.001 † |
| Left ventricular ejection fraction | NS | ||||
| >50% | 405 (57%) | 269 (60.3%) | 91 (50.3%) | 45 (54%) | |
| >35% but ≤50% | 200 (28.2%) | 120 (26.9%) | 55 (30.4%) | 25 (30%) | |
| ≤35% | 105 (14.8%) | 57 (12.8%) | 35 (19.3%) | 13 (16%) | |
| Gensini score | NS | ||||
| Median | 48 | 45 | 49 | 51 | |
| Interquartile range | 31.5–71 | 31–69.5 | 33.5–73.5 | 30.5–71.5 | |
| Coronary arteries with stenosis ≥50% ‡ | 0.07 | ||||
| 0 | 52 (7.3%) | 37 (8.3%) | 7 (3.8%) | 8 (10%) | |
| 1 | 230 (32.4%) | 160 (35.9%) | 47 (26%) | 23 (28%) | |
| 2 | 213 (30%) | 126 (28.3%) | 61 (33.7%) | 26 (31%) | |
| 3 | 191 (26.9%) | 111 (24.9%) | 58 (32.0%) | 22 (27%) | |
| Left main coronary artery with stenosis ≥50% | 60 (8.5%) | 38 (8.8%) | 14 (8.1%) | 8 (10%) | NS |
| Coronary heart disease duration (years) | NS | ||||
| <5 | 522 (73.5%) | 333 (74.7%) | 132 (72.9%) | 57 (69%) | |
| 5–10 | 98 (13.8%) | 63 (14.1%) | 23 (12.7%) | 12 (15%) | |
| >10 | 90 (12.7%) | 50 (11.2%) | 26 (14.4%) | 14 (17%) | |
| Coronary revascularization | |||||
| All types | 487 (68.6%) | 307 (68.8%) | 128 (70.7%) | 52 (63%) | NS |
| Coronary arteries bypass grafting | 86 (12.1%) | 49 (11%) | 26 (14.4%) | 11 (13%) | NS |
| Percutaneous coronary intervention | 407 (57.3%) | 261 (58.5%) | 104 (57.5%) | 42 (51%) | NS |
| Drug prescription at discharge | |||||
| Aspirin | 605 (85.2%) | 388 (87%) | 151 (83.4%) | 66 (80%) | NS |
| Thienopyridine | 425 (59.9%) | 277 (62.1%) | 102 (56.4%) | 46 (55%) | NS |
| Aspirin or thienopyridine | 629 (88.6%) | 400 (89.7%) | 156 (86.2%) | 73 (88%) | NS |
| β Blockers | 543 (76.5%) | 358 (80.3%) | 141 (77.9%) | 44 (53%) | <0.001 † |
| Statins | 459 (64.7%) | 286 (64.1%) | 117 (64.7%) | 56 (68%) | NS |
| Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker | 399 (56.2%) | 246 (55.2%) | 107 (59.1%) | 46 (55%) | NS |
| Previous percutaneous coronary intervention | 272 (38.3%) | 171 (38.4%) | 69 (38.1%) | 32 (39%) | NS |
| Previous coronary artery bypass grafting | 62 (8.7%) | 38 (8.5%) | 16 (8.8%) | 8 (10%) | NS |
| Previous coronary heart disease treated with medical therapy without myocardial revascularization | 82 (11.6%) | 47 (10.5%) | 25 (13.8%) | 10 (12%) | NS |
| Previous stroke or transient ischemic attack | 27 (3.8%) | 18 (4%) | 6 (3.3%) | 3 (4%) | NS |
| Severe chronic renal failure | 11 (1.6%) | 3 (0.7%) | 4 (2.1%) | 4 (5%) | 0.01 † |
| Chronic obstructive pulmonary disease | 25 (3.5%) | 13 (2.9%) | 6 (3.3%) | 6 (7%) | NS |
| Death | 130 (18.3%) | 61 (13.7%) | 43 (23.8%) | 26 (31%) | <0.001 † |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
