Author + information
- Received April 22, 2015
- Revision received August 7, 2015
- Accepted August 13, 2015
- Published online December 21, 2015.
- Hisao Kumakura, MD, PhD∗ (, )
- Hiroyoshi Kanai, MD, PhD,
- Yoshihiro Araki, MD, PhD,
- Yoshiaki Hojo, MD,
- Toshiya Iwasaki, MD and
- Shuichi Ichikawa, MD, PhD
- Department of Vascular Medicine, Vascular Center, Cardiovascular Hospital of Central Japan (Kitakanto Cardiovascular Hospital), Shibukawa, Gunma, Japan
- ↵∗Reprint requests and correspondence:
Dr. Hisao Kumakura, Department of Internal Medicine, Vascular Center, Cardiovascular Hospital of Central Japan (Kitakanto Cardiovascular Hospital), Shimohakoda 740 Hokkitu-machi, Shibukawa-city, Gunma Prefecture 377-0061, Japan.
Objectives The purpose of this study was to evaluate 15-year patency and life expectancy after endovascular treatment (EVT) with primary stenting guided by intravascular ultrasound (IVUS) for iliac artery lesions.
Background Fifteen-year patency, factors causing restenosis, and survival after IVUS-guided EVT are unclear based on the TransAtlantic Inter-Society Consensus II (TASC-II) classification in peripheral arterial disease (PAD).
Methods EVT was performed for 507 lesions in 455 patients with PAD. The 15-year endpoints were primary, primary-assisted, and secondary patency; overall survival; freedom from major adverse cardiovascular events (MACE); and freedom from major adverse cardiovascular and limb events (MACLE).
Results The 5-, 10-, and 15-year primary and secondary patencies were 89%, 83%, and 75%, respectively, and 92%, 91%, and 91%, respectively. There were no significant differences among TASC-II categories.
Conclusions IVUS-guided stenting for the iliac artery had favorable 15-year patency in all TASC categories. Life expectancy after EVT was poor, but stenting is feasible for patients with PAD.
The therapeutic strategy for peripheral arterial disease (PAD) is generally selected based on the lesion type, using the morphological stratification in the Inter-Society Consensus for the Management of Peripheral Arterial Disease (TransAtlantic Inter-Society Consensus [TASC]) classification in the aortoiliac artery (1). Favorable results have been obtained with endovascular treatment (EVT) for iliac artery lesions compared with lesions of the femoral artery (2–4). However, there are few reports on very long-term patency after treatment of TASC type A to D lesions.
Intravascular ultrasound (IVUS) gives a high-resolution cross-sectional image for use during coronary and peripheral intervention, and the additional morphological information may decrease the restenosis rate after stent implantation (5–8). However, factors evaluated by IVUS that cause restenosis in long-term patency after EVT have not been analyzed in detail. PAD is a systemic disease with multiple atherosclerotic risk factors and poor long-term survival (1,9,10), and more information is needed on long-term outcomes after EVT for PAD. Therefore, in this study, we analyzed 15-year patency and life expectancy after IVUS-guided primary stenting for iliac artery lesions categorized using the TASC classification.
Patients with PAD admitted to Kitakanto Cardiovascular Hospital from June 1993 to December 2013 for EVT for subjective symptoms categorized as higher than Fontaine class II and with stenosis of the iliac artery ≥70% in angiography were included in a single-center prospective study. Written informed consent for participation in the study was obtained from all patients, and the protocol was approved by our institutional ethical committee before initiation of the study (4). A total of 874 patients were treated for infrarenal aortoiliac disease. Of these patients, 194 with intermittent claudication were treated with exercise rehabilitation and medication alone (1). Revascularization was performed in 680 patients with critical limb ischemia (CLI) or intermittent claudication whose symptoms were not improved or deteriorated after medication. EVT was performed for all type A and B lesions. Of patients with type C lesions, 4 with iliac lesions and severe calcified stenosis in the common femoral artery were excluded because EVT and endarterectomy were performed concurrently. Of patients with type D lesions, 18 were excluded because of severe calcified stenosis in the common femoral artery, as mentioned in the preceding text (n = 2), abdominal aortic aneurysm requiring stent graft placement (n = 4), severe calcified aortoiliac occlusion (n = 9), and aortoiliac occlusion requiring concomitant femoropopliteal bypass (n = 3). As a result, EVT was performed in 658 patients, of whom 203 were excluded because of treatment with angioplasty alone (n = 169; types A: 98, B: 57, C: 12, D: 2), restenotic lesions (n = 29), and primary stenting without adequate IVUS data (n = 5). Therefore, analysis of endpoints was finally performed in 455 patients who underwent primary stenting for de novo aortoiliac lesions.
All patients received a bolus of 5,000 IU of heparin before treatment. The puncture site for arterial access was in the inguinal region of the femoral arteries, and EVT was performed through the iliac artery via an ipsilateral or contralateral route. All procedures for stent implantation were guided by IVUS. Images were acquired with an IVUS console (Volcano s5 Imaging system, Volcano Corp., San Diego, California, or CVIS, Boston Scientific, San Jose, California) and an IVUS catheter (Visions PV .018, 20 MHz, Volcano Corp., or Ultracross catheter, 30 MHz, Boston Scientific). After guidewire passage through the lesion, IVUS was performed at the lesion segment and at proximal and distal reference sites in the lesion segment with manual pullback. A lesion was classified as calcified on the basis of IVUS identification of an angle of calcification >180° in the plaque in the minimal lesion area. In cases with severely calcified lesions, minimal pre-dilation with balloon angioplasty was performed with Sterling or Wanda (Boston Scientific, Natick, Massachusetts) and Powerflex (Cordis, Fremont, California) balloon catheters. In most other cases, direct stent implantation was performed using Palmaz, SMART (Cordis), Wallstent RP, Express LD (Boston Scientific), and Luminexx (Bard Inc, Murray Hill, New Jersey) stents. For balloon-expandable stents, a stent diameter equivalent to the reference diameter was selected, whereas for self-expandable stents, a diameter 1 mm larger than the reference diameter was chosen. Expansion after stent implantation was performed by adjusting the balloon diameter and inflation pressure to minimize the difference between the arterial wall in the proximal and distal areas of the lesion and stent, guided by IVUS. If vessel dissection was found at the stent edges, another stent was added over the dissection. Minimal in-stent thrombus protrusion was identified in the same way. If a large thrombus protrusion was found in the stent, further ballooning or aspiration was added over the thrombus protrusion.
After EVT, the mean value of the 3 stent areas (proximal, middle, and distal) was calculated to give the mean stent lumen area (MLA). Measurements of the proximal and distal regions were performed at points defined as 10% of the stent length toward the center from the stent ends (7). Cross-sectional areas of the vessels and stents were determined using software on the IVUS display by 2 technologists who were blinded to the EVT procedure.
Administration of oral drugs (cilostazol, beraprost, sarpogrelate) for treatment of PAD was continued and oral aspirin, clopidogrel or ticlopidine was added as antiplatelet treatment for more than 3 days before EVT. Patients with side effects such as bleeding, appetite loss, headache, or palpitation were treated with only 1 agent. Drugs for hypertension and hyperlipidemia that were used before EVT were also administered continuously.
Blood was collected during fasting in the morning to determine levels of total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglyceride, albumin, glucose, glycosylated hemoglobin A1c, creatinine, fibrinogen, D-dimer, thrombin-antithrombin complex, and plasmin-α2 plasmin inhibitor complex. Diabetes mellitus, hypertension, coronary artery disease, and cerebral infarction were examined as risk factors for arteriosclerosis. Lipid abnormalities were diagnosed based on treatment with lipid-lowering agents; or low-density lipoprotein cholesterol ≥140 mg/dl, high-density lipoprotein cholesterol <40 mg/dl, or triglyceride ≥150 mg/dl (1,11).
Data analysis and endpoints
Pre- and post-operative rates of stenosis for the reference diameter were calculated after performing digital subtraction angiography (Philips Medical Systems, Best, the Netherlands) in 2 directions, using software from the manufacturer. Cases with post-operative residual stenosis <30% were classified as an initial success. Patients were followed up at 1, 3, and 6 months after EVT and then evaluated at 4- or 6-month intervals at an office visit (12). Restenosis in the follow-up period was defined as a decrease in the ankle brachial pressure index of ≥0.1 and a peak systolic velocity ratio >2.4 on duplex ultrasonography or ≥50% stenosis on angiography (13). Vital status and status of medications were assessed using hospital records and written questionnaires completed at the Foot Care Club in our hospital or in response to an invitation letter (14). Cases in which expansion could not be performed because a guidewire could not be passed through the lesion were excluded from the analysis of long-term patency. Patency rates are expressed as primary (defined as patency of the target lesion during follow-up), primary-assisted (defined as patency of the target lesion following EVT at the target vessel site in case of symptomatic restenosis, but without occlusion at any time), and secondary (defined as patency of the target lesion after treatment of a (re)occlusion, with patency ending with an untreated or surgically treated occlusion) (15,16). The endpoints of life expectancy were overall survival, freedom from major adverse cardiovascular events (MACE) (all-cause death, myocardial infarction, cardiac revascularization, and stroke), and freedom from major adverse cardiovascular and limb events (MACLE) (MACE, any repeat revascularization for limb, and major amputation). Myocardial infarction was defined as detection of an increase in cardiac troponin and/or creatine phosphokinase during: 1) symptoms of myocardial ischemia; 2) new (or presumably new) significant electrocardiographic ST-segment/T-wave changes or left bundle branch block; 3) development of pathological electrocardiographic Q waves; 4) new loss of viable myocardium or regional wall motion abnormality identified by an imaging procedure; or 5) identification of intracoronary thrombus by angiography or at autopsy (17). Cardiac revascularization was defined as a history of percutaneous coronary intervention or coronary artery bypass graft surgery. Stroke was defined as a hospital or neurologist report of diagnosis of ischemic stroke or transient ischemic attack. Major amputation was defined as above-the-ankle amputation.
Follow-up times are expressed as medians. Continuous variables are expressed as mean ± SD, and were compared by Student t test. Proportions were compared by chi-square test with Yates’ correction. The Kaplan-Meier method was used to determine patencies and survival in the follow-up period, and a log-rank test was performed for comparison of lesion types. The hazard ratio and confidence intervals were calculated for individual factors in a Cox univariate analysis. Factors with p < 0.05 in this analysis were used in a multivariate Cox regression model with a Wald test to determine predictors for endpoints. IBM SPSS Statistics version 22.0 (IBM Corp., Armonk, New York) was used for all calculations. Individual differences were considered to be significant for p < 0.05.
The subjects were 455 PAD patients of average age 71.8 ± 8.5 years. The patients had a total of 507 lesions. The characteristics of all cases and those of cases with type A/B and type C/D lesions are shown in Table 1. The rates of CLI and chronic total occlusion (CTO) were higher for type C/D lesions than for type A/B lesions. The initial success rates were 97.2% (493 of 507) in all subjects, and 99.6%, 97.0%, 98.1%, and 91.1% in lesion types A, B, C, and D, respectively. The initial success rate for type C/D lesions was significantly lower than that for type A/B lesions (Table 1) (p = 0.002). For 8 of the 14 lesions with an unsuccessful initial outcome, the guidewire could not be passed through the lesion. All such lesions were calcified CTO lesions. Other reasons for unsuccessful results included in-stent thrombus protrusion or a heavily calcified lesion, and there were 3 lesions in which residual stenosis <30% was not achieved. In addition, wire perforation occurred in 2 subjects, and major dissection due to sheath insertion occurred in 1 subject. In subjects with successful EVT, improvement of symptoms was observed in all cases with intermittent claudication, but minor amputation was performed for 2 subjects and major amputation was required for 1 subject with CLI. Of the patients with unsuccessful EVT, 12 underwent bypass surgery. Other patients were treated using exercise rehabilitation and medication.
Complications were observed in 18 cases (4.0%): wire perforation in 2, major dissection upon insertion of the sheath in 1, distal emboli in 5, hematoma that prolonged the hospital stay in 5, and pseudoaneurysm in 5. There were no deaths. For the 2 subjects who developed distal emboli, surgical resection of the thrombus was performed using a Fogarty embolectomy catheter, and 3 patients were treated with a thrombus aspiration catheter. With regard to post-operative death, in-hospital mortality was 0.2% (1 of 455), and 30-day mortality was 0.2% (1 of 455). The cause of death was acute myocarditis.
The median observation period was 63 months (range 1 to 246 months). At 5, 10, and 15 years, primary patencies were 87%, 83%, and 75% (Figure 1), primary-assisted patencies were 89%, 85%, and 82%, and secondary patencies were 92%, 91%, and 91%, respectively (Figure 1). The 5-, 10-, and 15-year primary patencies were 90%, 87%, and 79% in type A, 84%, 77%, and 69% in type B, 85%, 78%, and 78% in type C, and 87%, 87%, and 70% in type D lesions, respectively (Figure 2A), with no significant difference among TASC categories (p = 0.149). The 5-, 10-, and 15-year primary patencies were 88%, 83%, and 76% in type A/B, and 83%, 83%, and 70% in type C/D lesions, respectively, with no significant difference between these groups (Figure 2B) (p = 0.406).
The results of Cox univariate analysis of individual factors and calculated hazard ratios for restenosis are shown in Table 2. Lesion length, pre- and post-procedural angiographic stenosis rates, pre- and post-procedural MLA, calcified lesions, in-stent thrombosis, and discontinuation of antiplatelet therapy were found to be significant factors associated with restenosis. In Cox multivariate analysis, post-procedural MLA, in-stent thrombosis, antiplatelet therapy discontinuation, and calcified lesions were independent predictors of primary patency. Overall, 7 patients discontinued antiplatelet agents with medical advice for major surgery or bleeding, and 5 patients self-discontinued. Restenosis occurred in 4 of the self-discontinued patients and in 1 case discontinued for major bleeding.
There were 158 (34.7%) deaths during follow-up due to cardiac or major vascular disease (n = 50, 31.6%), cerebrovascular disease (n = 28, 17.7%), malignancy (n = 36, 22.8%), pneumonia (n = 31, 19.6%), renal failure (n = 5, 3.2%), and other causes (8 patients, 5.1%). At 5, 10, and 15 years, the overall survival rates were 82%, 56%, and 40% (Figure 3), freedom from MACE was 71%, 44%, and 25%, and freedom from MACLE was 57%, 29%, and 14%, respectively (p < 0.001) (Figure 3). The 5-, 10-, and 15-year overall survival rates were 83%, 58%, and 42% in type A/B, and 76%, 47%, and 23% in type C/D lesions, respectively, with a tendency for lower overall survival in cases with type C/D lesions (p = 0.088). The 5-, 10-, and 15-year freedom from MACE rates were 72%, 48%, and 28% in type A/B, and 68%, 31%, and 21% in type C/D lesions, respectively, with no significant difference between the groups (p = 0.144). The 5-, 10-, and 15-year freedom from MACLE rates were 58%, 30%, and 14% in type A/B, and 54%, 25%, and 13% in type C/D lesions, respectively, also with no significant difference between the groups (p = 0.409).
The results of Cox univariate analysis of individual factors are shown in Table 3. In multivariate analysis using significant risk factors in univariate analysis, age, CLI, diabetes mellitus, hemodialysis, and higher D-dimer level were found to be independent predictors of all-cause mortality (Table 3).
There has been no previous comparison of 15-year patency and life expectancy among TASC categories after iliac stenting. In primary or selective stenting for type C/D iliac artery lesions, recent meta-analyses have found an initial success rate of 86% to 100% (2,3,18). The initial success rate in the current study was similarly favorable, except for type D lesions. This may be because there was a high incidence of severe calcified CTO lesions in type D cases. In primary stenting, there was no correlation between outcome and TASC category, and 15-year patency was approximately 75% for all lesions. For comparison, a meta-analysis found 5-year primary patencies of 67.1% and 63.0% for primary and selective stenting for type C/D lesions, respectively (3). Thus, our 15-year patency was more favorable than data from the meta-analysis for 5-year primary patency.
In Cox multivariate analysis, post-procedural MLA, in-stent thrombosis, discontinuation of antiplatelet therapy, and calcified lesions were independent predictors of primary patency. The high-resolution cross-sectional IVUS image clearly identifies in-stent thrombosis and calcification. Platelets play a central role in the pathogenesis of atherothrombosis (19), and early and late stent thromboses are causes of stent restenosis in the coronary artery (20). Residual in-stent thrombosis in iliac stenting may also be a cause of restenosis, and calcified lesions are a refractory risk with a high rate of restenosis following coronary stenting (21). Recent experimental studies have shown that chronic inflammation is closely associated with development of coronary calcification and possible development of neointimal hyperplasia (22). Local hemodynamic factors, particularly low endothelial shear stress, are causes of in-stent restenosis and stent thrombosis (23), and calcified lesions may modify the stent figure and derange the shear stress to increase neointimal proliferation.
Discontinuation of antiplatelet therapy was another cause of restenosis in long-term patency. Self-discontinuation is a serious problem in more than 5-year follow-up after EVT. The reasons for self-discontinuation include adverse effects, cost, and perception that medication is not needed (24). Many chronically used drugs are prescribed prophylactically, in the absence of symptoms, and most patients may decide to use their medication only for a limited period of time, based on a lack of information (25). Older patients have been found to be less likely to persist with drugs after an acute coronary event at 3-month follow-up (24). Because patients with PAD are older than those with coronary disease, maintenance of drug compliance may be particularly important after EVT.
The 15-year survival rate was higher than the reported life expectancy in patients with PAD (1,12). This may be because there were relatively few patients with CLI in the current study. In multivariate Cox analysis for life expectancy, age, CLI, diabetes mellitus, hemodialysis, and higher D-dimer were independent predictors for all-cause mortality. Mortality is strongly correlated with CLI and ankle brachial pressure index (as a measure of the severity of PAD), and diabetes mellitus also increases the risks of PAD and cardiovascular mortality (1,26,27). These consistent results suggest that severe PAD and complication with diabetes mellitus involve extensive and severe degrees of systemic atherosclerosis that are responsible for the mortality caused by cardiovascular disease.
Hemodialysis was also found to be an independent predictor of mortality. Several studies have suggested that chronic kidney disease is a prognostic indicator of cardiovascular disease (28,29) and O'Hare et al. (30) found that renal dysfunction was a strong independent predictor of mortality in CLI patients. We also found an association between D-dimer and all-cause mortality. D-dimer is an end product of fibrinolysis that promotes the inflammatory cascade by activating neutrophils and monocytes, inducing secretion of inflammatory cytokines, and promoting hepatic synthesis of acute-phase proteins (31). Plasma D-dimer levels are also independently associated with the presence of coronary artery disease (32). Thus, a higher D-dimer level may be a cause of increased mortality after EVT in patients with PAD.
The limitations of this study include the relatively small sample size and performance of the study at a single facility. The serum level of high-sensitivity C-reactive protein is also an independent risk factor for cardiovascular disease and restenosis (33,34), but only the serum level of non–high-sensitivity C-reactive protein was measured in cases before 2004. Therefore, further studies of long-term patency and life expectancy are required in patients with PAD.
The 15-year patency was favorable after EVT in patients with PAD. Post-procedural MLA, in-stent thrombosis, discontinuation of antiplatelet therapy, and calcified lesions were independent predictors of primary patency. The 15-year survival rate was more favorable than the reported life expectancy in patients with PAD. Age, CLI, diabetes mellitus, hemodialysis, and higher D-dimer levels were independent predictors of all-cause mortality after EVT.
WHAT IS KNOWN? Fifteen-year patency, factors causing restenosis, and survival after endovascular treatment with primary stenting guided by IVUS for iliac artery lesions are unclear based on the TASC-II classification in peripheral arterial disease.
WHAT IS NEW? The 15-year primary patency showed no significant difference among TASC-II categories. Post-procedural mean lumen area, in-stent thrombosis, antiplatelet therapy discontinuation, and calcified lesions were predictors of primary patency. Age, critical limb ischemia, diabetes, hemodialysis, and D-dimer level were predictors of overall survival.
WHAT IS NEXT? Further studies of long-term patency and life expectancy are required for IVUS-guided iliac stenting. New approaches are needed to improve drug compliance after EVT in patients with PAD.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- critical limb ischemia
- chronic total occlusion
- endovascular treatment
- intravascular ultrasound
- major adverse cardiovascular event(s)
- major adverse cardiovascular and limb event(s)
- mean stent lumen area
- peripheral arterial disease
- TransAtlantic Inter-Society Consensus
- Received April 22, 2015.
- Revision received August 7, 2015.
- Accepted August 13, 2015.
- American College of Cardiology Foundation
- Fitzgerald P.J.,
- Oshima A.,
- Hayase M.,
- et al.
- Yilmaz S.,
- Sindel T.,
- Golbasi I.,
- Turkay C.,
- Mete A.,
- Luleci E.
- Thygesen K.,
- Alpert J.S.,
- Jaffe A.S.,
- et al.
- Fuster V.,
- Moreno P.R.,
- Fayad Z.A.,
- Corti R.,
- Badimon J.J.
- Philip F.,
- Agarwal S.,
- Bunte M.C.,
- et al.
- Wexler L.,
- Brundage B.,
- Crouse J.,
- et al.
- Higgins C.L.,
- Marvel S.A.,
- Morrisett J.D.
- Koskinas K.C.,
- Chatzizisis Y.S.,
- Antoniadis A.P.,
- Giannoglou G.D.
- Criqui M.H.,
- Ninomiya J.K.,
- Wingard D.L.,
- Ji M.,
- Fronek A.
- Sarnak M.J.,
- Levey A.S.,
- Schoolwerth A.C.,
- et al.
- O'Hare A.M.,
- Bertenthal D.,
- Shlipak M.G.,
- Sen S.,
- Chren M.M.
- McDermott M.M.,
- Ferrucci L.,
- Guralnik J.M.,
- et al.
- Koenig W.,
- Rothenbacher D.,
- Hoffmeister A.,
- Griesshammer M.,
- Brenner H.
- Koenig W.,
- Sund M.,
- Frohlich M.,
- et al.