| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Departments of Angiology (M.S., W.M., M.H., S.S., R.A., E.M.) and Laboratory Medicine (M.E., O.W.), University of Vienna, Medical School, Waehringer Guertel 18–20, A-1090 Vienna, Austria. Received July 10, 2003; revision requested September 29; revision received October 1; accepted November 6. Address correspondence to M.S. (e-mail: martin.schillinger@akh-wien.ac.at).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: A total of 650 patients (median age, 70 years; 389 men) with peripheral artery disease who underwent iliac artery PTA (n = 95), iliac artery stent placement (n = 83), femoropopliteal PTA (n = 406), or femoropopliteal stent placement (n = 66) were selected from a prospective database. Patients were categorized according to their preintervention smoking habits as nonsmokers (n = 352), light smokers (one to nine cigarettes daily) (n = 54), habitual smokers (10–20 cigarettes daily) (n = 82), or heavy smokers (>20 cigarettes daily) (n = 162). Multivariate Cox proportional hazards analysis was used to determine whether there was an association between smoking habits and restenosis (
50%) in the treated vessel segment within 1 year after treatment.
RESULTS: Cumulative restenosis rates at 6 and 12 months according to patients’ smoking habits were 99 and 190 nonsmokers, 18 and 22 light smokers, 16 and 29 habitual smokers, and 26 and 47 heavy smokers, respectively (P < .001). Adjusted hazard ratios for restenosis in smokers compared with nonsmokers were 1.51 (95% CI: 0.92, 2.50) for light smokers, 0.49 (95% CI: 0.28, 0.87) for habitual smokers, and 0.46 (95% CI: 0.30, 0.71) for heavy smokers, indicating a reduced restenosis risk in patients who smoked 10 or more cigarettes daily. These patients had reduced restenosis rates after either iliac (P = .011) or femoropopliteal intervention (P = .009). However, endovascular treatment at a younger age, coronary artery disease, and history of myocardial or cerebrovascular infarction were more frequently found in smokers.
CONCLUSION: Smoking 10 or more cigarettes daily is associated with a reduced rate of intermediate-term restenosis after lower-limb endovascular interventions.
© RSNA, 2004
Index terms: Arteries, extremities, 92.128, 92.12983 • Arteries, restenosis, 92.454, 92.458 • Arteries, stenosis or obstruction, 92.721 • Arteries, transluminal angioplasty, 92.1282 • Smoking
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Smokers typically have elevated levels of carboxyhemoglobin and higher blood concentrations of carbon monoxide (11,12). Carbon monoxide has potent antiinflammatory and antiproliferative capacity, and it inhibits vascular smooth muscle cell proliferation after vascular injury due to balloon dilation (13–16). Furthermore, cigarette smoke extract induces necrosis in proliferating vascular smooth muscle cells (17). Localized proliferationof vascular smooth muscle cells is a key factor in luminal diameter decrease after endovascular treatment (18). On the basis of these findings, we hypothesized that smoking may protect vessels from restenosis after lower-limb endovascular interventions, presumably through the antiproliferative effect of increased carbon monoxide levels. The aim of the present study, therefore, was to investigate whether smoking has an effect on recurrent luminal narrowing after PTA or stent placement in lower-limb arteries.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
During the observation period, 712 patients with peripheral artery disease underwent iliac or femoropopliteal interventions at our institution. Of these, 62 patients (9%) were excluded from further analysis because of incomplete baseline and follow-up data. The excluded patient group closely resembled the group of patients for whom we had complete follow-up data with respect to demographics, baseline clinical and angiographic findings, and smoking habits, and there were no significant differences between the two groups. No other patients, apart from these 62, were excluded from the final analysis, and we are confident that no patients were missed during data collection for the study. Thus, 650 patients with a median age of 70 years (interquartile range, 60–76 years) were included in the final analysis, 389 (60%) of whom were men (median age, 69 years; interquartile range, 61–75 years) and 261 (40%) of whom were women (median age, 71 years; interquartile range, 60–77 years; Mann-Whitney test, P = .19).
Patient Data
At admission, two independent observers recorded the patient’s medical history and data from physical examination by using a standard questionnaire (M.S., W.M.). Data were checked for interobserver agreement on the day of the patient’s discharge; if discrepancies were found, the patient was reevaluated by both investigators in consensus. At the time of hospital admission, patients were categorized as nonsmokers, light smokers (one to nine cigarettes daily), habitual smokers (10–20 cigarettes daily), or heavy smokers (>20 cigarettes daily). Further analysis was based on this classification of smoking habits at the time of hospital admission because it was assumed that any potentially causal effect of smoking on restenosis would become manifest mainly within the 1st month after intervention and that few of these elderly patients would abruptly change their smoking habits.
Color-coded Duplex Ultrasonography
The entire arterial segment from the iliac bifurcation to the origin of the anterior tibial artery was evaluated at baseline and follow-up examinations with color-coded duplex ultrasonography (US). The extent and location of the treated lesion were documented during the baseline evaluation as follows: The distance from the aortic bifurcation (in patients with iliac artery stenosis) or from the femoral bifurcation (in patients with femoropopliteal stenosis) to the proximal segment with reduced luminal diameter, as well as the length of the target lesion, was determined and displayed in a graphic illustration. The peak systolic velocity in the diseased segment was then measured and compared with the peak systolic velocity in the preceding normal segment. A focal increase of at least 140% in peak systolic velocity in the diseased segment (ie, peak systolic velocity at least 2.4 times the normal level) was considered to indicate a stenosis of more than 50% (22). All duplex US investigations were performed by one of four medical technicians with at least 3 years of experience in vascular US, under the supervision of one of the authors (E.M., with more than 15 years of experience in vascular US).
Interventions
Angiography and PTA with elective stent implantation were performed in accordance with a previously published standard protocol (20) that included transfemoral access and administration of 5,000 IU of heparin intraarterially. All interventions were performed by one of two experienced interventionalists (R.A. or E.M., each with more than 15 years of experience with endovascular therapy) using a nonionic low-osmolality contrast agent (Optiray 320; Mallinckrodt Medical, St Louis, Mo). PTA was performed with the balloon diameter corresponding to the proximal nondiseased vessel diameter. Pre- and postintervention biplane angiograms were used to determine the severity of stenosis, the technical success of the intervention, and the degree of residual stenosis. In addition, all lesions were classified according to criteria developed by the TransAtlantic Inter-Society Consensus (TASC) (19). The criterion for elective stent implantation was suboptimal technical success of the intervention, defined as (a) a residual stenosis of more than 30% after initial PTA, (b) elastic recoil in the dilated segment after repeated PTA, with a residual stenosis of more than 30% in the dilated segment, or (c) arterial dissection after repeated PTA, with a reduction of more than 30% in luminal diameter. Stent implantation was performed with different commercially available stents in iliac arteries (Symphony, Boston Scientific, Natick, Mass; Dynalink, Guidant, Santa Clara, Calif) and in femoropopliteal vessels (Easy Wallstent; Boston Scientific). In cases of simultaneous interventions in these two different vessel areas, only the area (either iliac or femoropopliteal) that contained the most extensive lesions and highest degree of stenosis was considered for the analysis of restenosis. The vessel area for the analysis of restenosis was specified immediately after the intervention, to avoid any bias.
The occurrence of periintervention and postintervention complications at the site of arterial puncture and in the dilated vessel segment was documented within 48 hours after the intervention by an observer (W.M.) other than the interventionalist. All patients received antithrombotic medication (100 mg acetylsalicylic acid daily), starting at least 1–4 days before the scheduled intervention. Patients who underwent stent implantation also received clopidogrel in an initial loading dose of 300 mg immediately after stent implantation and a daily dose of 75 mg for 1 month thereafter.
Follow-up Imaging
Color-coded duplex US and ankle-brachial index measurements were performed 24 hours after PTA or stent implantation, to exclude early thrombotic reocclusion. Patients were then followed up at 3, 6, and 12 months in the outpatient clinic with further measurements of ankle-brachial index, evaluations of symptoms and smoking habits, and physical examinations to identify any recurrence of stenosis. However, patients also were advised to visit the outpatient clinic at any new onset of claudication or increase in symptoms. Patients with new onset of claudication, increase in symptoms, or substantial reduction in ankle-brachial index (ie, decrease by at least 0.15 from the maximum postintervention level) were further evaluated with mandatory color-coded duplex US (7,8,23). A focal increase in peak systolic velocity by at least 2.4 times the normal level at duplex US was considered to indicate restenosis. In patients without decrease in the ankle-brachial index and without new onset of symptoms, restenosis was excluded without morphologic follow-up. In these patients, diabetic medial sclerosis and, thus, incompressibility of the arteries were excluded by measuring the waveforms of pulse-volume recordings (oscillometry). Patients in whom restenosis was suspected because of US findings or in whom findings were inconclusive (eg, decrease in the ankle-brachial index but no evidence of restenosis at duplex US) underwent follow-up angiography. Patients who developed de novo stenosis in the treated limb at a location 50 mm or farther from the treated site were not classified as having restenosis. Follow-up data were evaluated by two independent observers (M.S., S.S.) who were blinded to patients’ preintervention data and laboratory findings.
Definitions
The diagnosis of peripheral artery disease according to the Fontaine classification system was based on clinical evaluation, ankle-brachial index measurements, and duplex US and was confirmed with lower-limb angiography in all patients. We defined primary technical success as residual stenosis of 30% or less in the treated segment, as measured on the final angiogram. Primary technical success was further classified either as successful intervention without residual stenosis (in the presence of less than 10% remaining luminal diameter reduction in the treated segment) or as successful intervention with residual stenosis (in the presence of 10%–30% luminal diameter reduction in the treated segment). Poor runoff was defined as either occlusion or significant stenosis of the femoral or popliteal artery distal to the treated segment and/or occlusion of at least two tibioperoneal arteries.
We defined restenosis as recurrent reduction of at least 50% in the diameter of the previously treated vessel segment. Color-coded duplex US images obtained with a 5-MHz linear array color probe (XP 10; Acuson, Mountain View, Calif) and, if available, follow-up angiograms, were used for categorization of restenosis according to the protocol described earlier. We did not segregate early reocclusion from restenosis but included both events as failures in the final analysis. Stratification of iliac and femoropopliteal stenoses was performed according to recommendations 31 and 34 in the TASC criteria (19). Briefly described, type A lesions included single stenoses of less than 3 cm in length; type B lesions included single stenoses and occlusions of 3–9 cm and multiple stenoses and occlusions of less than 10 cm in the iliac arteries, as well as single stenoses and occlusions of 3–4 cm in the femoropopliteal arteries; type C lesions included single iliac stenoses and unilateral iliac occlusions of 10 cm or more in length, as well as multiple femoropopliteal lesions of less than 5 cm; and type D lesions included multiple iliac stenoses and bilateral iliac occlusions of 10 cm or more, as well as multiple femoropopliteal stenoses and occlusions of 5 cm or more.
Statistical Analysis
To estimate the sample size needed to detect significant differences in restenosis rates between different groups of smokers, we tested the following two assumptions in a power analysis (
= .05; 1 – ß = .80): First, we hypothesized that combined restenosis rates of nonsmokers and occasional smokers versus habitual smokers and heavy smokers would differ by at least 15% (assuming 1-year restenosis rates of 50%–55% versus 40%–45% for these two combined groups, respectively). A sample size of 500–550 patients therefore would be necessary to detect significant differences. Second, we assumed that the four groups of patients according to smoking habits were cohorts with restenosis risks that would differ by an odds ratio of more than 1.8. Thus, a sample size of 150 patients in each group (ie, a total study population of 600 patients) was determined to be necessary for the detection of significant differences.
Continuous variables were measured as median and interquartile range (range from the 25th to the 75th percentile). Discrete variables were measured as numbers and percentages. The main outcome measure (restenosis) was dichotomized according to the absence or presence of a recurrent stenosis of 50% or more in the treated segment at follow-up. We used the 
2 test to compare proportions and the Mann-Whitney test for univariate comparison of continuous data. Freedom from restenosis according to patients’ smoking habits was calculated with the Kaplan-Meier method, and the results were compared by means of the log-rank test. The Kaplan-Meier curve was plotted to include patients with primary PTA failure (ie, loss of patency at 0 months) and, thus, indicates the restenosis rate according to the intention to treat. Multivariate Cox proportional hazards analysis was applied to assess the effect of postintervention smoking habits on restenosis rates. To adjust for confounding effects, baseline variables were incorporated into the model as possible predictor variables if they were imbalanced between smokers and nonsmokers, as indicated by a P value of less than .2; were imbalanced between patients with restenosis and those without restenosis, as indicated by a P value of less than .2; or were established risk factors for restenosis. We tested for interactions between baseline variables by using stratification and multiplicative interaction terms and log likelihood 2 tests. Results of Cox proportional hazards modeling were presented as the hazard ratio and 95% CI. A two-sided P value of less than .05 was considered to indicate a statistically significant difference. Calculations were performed with statistical software (SPSS version 10.0, SPSS, Chicago, Ill) by one of the authors (M.S.).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Smoking Habits
Among the 650 patients studied, 352 (54%) were nonsmokers, 54 (8%) were light smokers (one to nine cigarettes daily), 82 (13%) were habitual smokers (10–20 cigarettes daily), and 162 (25%) were heavy smokers (>20 cigarettes daily). Discrepancies in the smoking histories of 71 (11%) of the patients were found by the two observers. Demographic data and clinical characteristics of the study patients grouped according to smoking history at the time of hospital admission are shown in Table 1. Smokers were younger than nonsmokers at the time of intervention, and they were more frequently men. Furthermore, coronary artery disease was more frequently found in smokers than in nonsmokers, as was a higher frequency of prior myocardial infarction and cerebral infarction. Comparing the affected vessel areas, smokers more frequently had iliac artery occlusive disease, and nonsmokers more frequently had femoropopliteal lesions. TASC type B and D lesions were more frequently found in smokers, whereas nonsmokers more frequently had TASC type A and C lesions. Fewer smokers than nonsmokers had arterial hypertension and diabetes mellitus (Table 1).
|
Follow-up for Restenosis
At follow-up, 521 (80%) of the 650 patients were examined with US, 381 (59%) were examined with angiography, and 129 (20%) were excluded from morphologic imaging because they had no deterioration in the ankle-brachial index and no new onset or exacerbation of symptoms of stenosis. Restenosis was found in 278 (43%) of the 650 patients during the follow-up period. Cumulative restenosis rates at 6 and 12 months according to the type of intervention were 50 (28%) and 69 (39%) of 178 iliac artery interventions and 170 (36%) and 241 (51%) of 472 femoropopliteal interventions, respectively. Demographic data and clinical characteristics of the 650 patients, according to the presence or absence of restenosis, are given in Table 2. Patients with critical limb ischemia (Fontaine stage III or IV peripheral artery disease), recurrent stenosis, longer lesions, residual stenosis after intervention, or poor runoff had an increased restenosis rate after lower-limb endovascular treatment. Patients treated for TASC B or D lesions had higher restenosis rates, whereas those with TASC A lesions had higher patency rates. The results of a comparison of iliac lesions with femoropopliteal lesions showed that restenosis occurred less frequently after iliac artery interventions (Table 2). Furthermore, stent implantation was associated with a slightly reduced restenosis rate compared with that for PTA.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Smoking is a major risk factor for the development of peripheral artery disease (1,2). Consistently, in the present study, smokers required endovascular treatment at a younger age compared with nonsmokers. Furthermore, smokers had more frequent complications of atherosclerotic disease in other vessel areas (eg, coronary artery disease, cerebrovascular infarction). The predominant allocation of peripheral atherosclerosis to the iliac arteries of smokers confirms prior observations (24,25); however, the causes of the preferential involvement of the iliac arteries over that of the femoropopliteal segment in smokers remain unclear.
Although smoking was associated with a reduced frequency of repeated interventions after initial PTA of the coronary vessels, the rate of angiographic restenosis after initial intervention was similar in smokers and nonsmokers (9,10). So far, smoking therefore has been suggested to be neither a risk factor for restenosis nor protective against recurrent lumen narrowing in the coronary vessel area. To our knowledge, comparable data on an association between smoking and restenosis after peripheral vascular interventions have not been published previously. In contrast to the results observed in the coronary vessel area, our data suggest that smoking 10 or more cigarettes daily is associated with a reduced rate of restenosis after PTA of the iliac or femoropopliteal arteries. Consistency between findings after iliac and femoropopliteal artery PTA and stent implantation supports the view that smoking indeed influences vascular remodeling and neointimal hyperplasia after endovascular treatment. Most likely, this protective effect is mediated by an interaction with wound healing and vascular smooth muscle cell proliferation. Two (mutually nonexclusive) mechanisms may be discussed, although we may only speculate about the underlying pathophysiology of this unexpected paradox: Elevated concentrations of carboxyhemoglobin and increased carbon monoxide blood levels are found in smokers (11,12). Carbon monoxide causes dilatation of blood vessels similar to that with nitric oxide and has been shown to possess antiinflammatory properties (26). Furthermore, carbon monoxide is a potent inhibitor of vascular smooth muscle cell proliferation after arterial injury (13–16), and vascular smooth muscle cell proliferation contributes to restenosis (18). Increased blood levels of carbon monoxide therefore may lessen vascular inflammation and inhibit vascular smooth muscle cell proliferation in the treated segment. Another potentially protective mechanism may be the toxic effect of various chemicals in cigarette smoke (nicotine, benzopyrene, acrolein, acetaldehyde) on vascular smooth muscle cells. These substances have been shown to cause necrosis in vascular smooth muscle cells (17) and, thus, may reduce hypertrophic neointima formation after balloon injury.
Our findings, however, certainly do not suggest that we should recommend smoking to patients or rush to treat patients with carbon monoxide inhalation therapy after PTA. The dangers of carbon monoxide inhalation may outweigh the benefits. Carbon monoxide toxicity is a consequence of tissue hypoxia created by the displacement of oxygen from hemoglobin and the subsequent impairment of oxygen release to the tissues. Because patients with peripheral artery disease frequently suffer from coronary artery disease and cerebrovascular atherosclerosis, even mild hypoxia during carbon monoxide inhalation therapy may be dangerous for these patients. Nevertheless, preliminary data from animal experiments have indicated the efficacy of carbon monoxide inhalation therapy for prevention of restenosis (27), and dose studies in humans have been initiated. Local delivery of carbon monoxide at the site of intervention has not yet been studied, but might hold promise. Other studies have focused on the potential role of physiologic carbon monoxide donors in vascular disease. In this context, heme oxygenase 1 seems to be particularly relevant in restenosis after PTA (7). Heme oxygenase 1–derived carbon monoxide (28,29) has been shown to inhibit vascular smooth muscle cell proliferation and recurrent lumen narrowing after balloon angioplasty (30,31). Therapeutic administration of carbon monoxide donors or upregulation of enzymes such as heme oxygenase 1 may be worth examining experimentally to reduce the high rates of restenosis after endovascular treatment of lower-limb peripheral artery disease.
We are aware of several limitations of the present study: We could not demonstrate a pathophysiologic mechanism of the protective effect of smoking, because intravascular levels of carbon monoxide or carboxyhemoglobin could not be measured in this setting. In addition, although we consistently recorded patients’ smoking habits at the time of the intervention, we were able to obtain similar data at follow-up only for approximately half of the patients, and therefore we could derive only a crude estimation of the change in smoking habits. However, given the age of our patients, one may assume that only a minor proportion would abruptly change their smoking habits within this period. Moreover, the protective effects of smoking probably would be most relevant during the early phases of vascular wound healing, within the first month after PTA, and a moderate change in smoking habits during later follow-up, therefore, may have only marginally influenced our findings. On the other hand, the inclusion of patients’ self-reported smoking status at the time of follow-up certainly would have introduced a substantial reporting bias into our analysis, because heavy smokers with restenosis probably would have underreported their daily cigarette consumption, whereas those without restenosis would likely be more honest about their smoking habits. Therefore, recording patients’ smoking status at the time of the intervention seems an acceptable strategy for excluding a systematic reporting bias with respect to restenosis.
A certain bias may have been introduced by patients’ underreporting of the amount of their daily cigarette consumption. Nevertheless, the observed effect was quantitatively impressive and consistently reproducible in all subgroups, and the presumption of underlying pathophysiology therefore seems reasonable.
In conclusion, smoking 10 or more cigarettes daily is associated with a reduced rate of intermediate-term restenosis after lower-limb endovascular interventions.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, M.S., E.M.; study concepts, M.S., M.E., W.M., O.W., E.M.; study design, M.S., M.H., S.S., R.A., E.M.; literature research, M.S.; clinical studies, R.A., E.M.; data acquisition, M.S., W.M., S.S., M.H.; data analysis/interpretation, all authors; statistical analysis, M.S.; manuscript preparation, M.S., E.M.; manuscript definition of intellectual content, editing, revision/review, and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-H. Liu, H.-H. Lin, Y.-F. Yang, Y.-L. Liu, H.-L. Kuo, I-K. Wang, C.-Y. Chou, and C.-C. Huang SUBCLINICAL PERIPHERAL ARTERY DISEASE IN PATIENTS UNDERGOING PERITONEAL DIALYSIS: RISK FACTORS AND OUTCOME Perit. Dial. Int., January 1, 2009; 29(1): 64 - 71. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-H. Liu, P.-W. Lin, Y.-L. Liu, H.-H. Lin, and C.-C. Huang Comparison of classical and non-classical cardiovascular risk factors influencing the patency of native arteriovenous fistulas after percutaneous transluminal angioplasty therapy among haemodialysis patients Postgrad. Med. J., August 1, 2007; 83(982): 547 - 551. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Gulesserian, C. Wenzel, G. Endler, R. Sunder-Plassmann, C. Marsik, C. Mannhalter, N. Iordanova, M. Gyongyosi, J. Wojta, S. Mustafa, et al. Clinical Restenosis after Coronary Stent Implantation Is Associated with the Heme Oxygenase-1 Gene Promoter Polymorphism and the Heme Oxygenase-1 +99G/C Variant Clin. Chem., September 1, 2005; 51(9): 1661 - 1665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. P. Prabhu, M. Schillinger, and E. Minar Chemical Composition of Cigarettes Could Be a Confounding Factor * Drs Schillinger and Minar respond: Radiology, January 1, 2005; 234(1): 310 - 310. [Full Text] [PDF]
|
|
|
|
|
|
S. Corrao, M. Schillinger, and E. Minar Protective Effect of Smoking: Misleading Use of Statistics * Drs Schillinger and Minar respond: Radiology, December 1, 2004; 233(3): 934 - 934. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
