|
|
||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology (W.M.K., W.P.T.M.M.) and Julius Health Sciences and Primary Care Center (Y.v.d.G.), University Medical Center Utrecht, Heidelberglaan 100, E.01.132, 3584 CX Utrecht, the Netherlands; Department of Surgery, Slingeland Hospital, Doetinchem, the Netherlands (J.S.); and Department of Surgery, St Antonius Hospital, Nieuwegein, the Netherlands (F.L.M.). Received May 7, 2003; revision requested July 16; final revision received November 5; accepted January 5, 2004. Address correspondence to W.P.T.M.M. (e-mail: w.mali@azu.nl).
| ABSTRACT |
|---|
|
|
|---|
MATERIALS AND METHODS: A total of 279 symptomatic patients with stenosis or short (
5-cm) occlusion of the iliac arteries were randomly assigned to undergo either primary stent placement or primary angioplasty followed by selective stent placement (in case of a residual mean pressure gradient greater than 10 mm Hg at the treated site). Follow-up data for all 279 patients were provided by the general practitioners and referring clinicians. Events of interest were arterial interventions, reinterventions in the iliac arteries, cardiovascular events (myocardial infarction, stroke, or extracranial bleeding), and death. Regression analysis was performed to identify predictors of reintervention and of cardiovascular morbidity and mortality.
RESULTS: The mean follow-up period was 5.6 years ± 1.3 (± standard deviation). There were no significant differences between primary stent placement and primary angioplasty treatment groups in regard to number of reinterventions in the treated iliac arteries (33 [18%] of 187 segments and 33 [20%] of 169 segments, respectively) or in the ipsilateral legs (45 [25%] of 181 legs and 50 [30%] of 164 legs, respectively). The risk of other cardiovascular events in primary stent placement and primary angioplasty groups was 13% (18 of 143) and 11% (15 of 136), and the risk of death was 15% (21 of 143 patients) and 16% (22 of 136 patients), respectively. Sex, presence of critical ischemia, and length of stenosis were predictors of whether a patient would require iliac reintervention. Myocardial infarction, stroke, and vascular death were predicted on the basis of a patients creatinine level and walking distance as tested at the time of inclusion.
CONCLUSION: No difference was found in the number of reinterventions between the two treatment groups 5 years after treatment. Patients with iliac artery disease are at high risk of cardiovascular morbidity and mortality.
© RSNA, 2004
Index terms: Arteries, iliac, 984.721, 986.721 Arteries, stenosis or obstruction, 984.721, 986.721 Arteries, transluminal angioplasty, 984.721, 986.721 Stents and prostheses, 984.721, 986.721
| INTRODUCTION |
|---|
|
|
|---|
Peripheral artery disease (PAD) coexists with other manifestations of symptomatic atherosclerotic disease. Patients with PAD are more likely to have a nonfatal myocardial infarction or stroke than to undergo major amputation for leg ischemia. Also, the risk for cardiovascular death is highthe 5-year mortality rate has been found to be about 30% (1823).
The purpose of our study was to compare long-term cardiovascular morbidity and mortality and their determinants in a population initially treated for stenosis or short occlusion of an iliac artery with one of two endovascular treatment strategies.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Study Population
A total of 279 patients with intermittent claudication or critical ischemia caused by stenosis or short occlusion in the iliac arteries were recruited from the departments of vascular surgery at seven study centers (see Acknowledgments) between November 1993 and November 1996. Inclusion criteria for patients were PAD with pain localized in the buttock, upper leg, or calf; reduced pulsation of the femoral artery and reduced ankle-brachial index; greater than 50% reduction in arterial diameter, evident at angiography; or stenosis of 10 cm or less in length in the common or external iliac artery or occlusion of 5 cm or less that allowed passage of a guidewire. Exclusion criteria were stenosis of more than 10 cm in length; occlusion of more than 5 cm in length, or of 5 cm or less if it did not allow the passage of a guidewire; stenosis involving the distal aorta; or severe comorbidity (eg, severe cardiac or cerebrovascular abnormality, malignant disease).
Eighty-six patients were excluded from the study. Of these, 27 were unwilling to participate. The remaining 59 were excluded due to protocol restrictions, as follows: In 21 patients, stenoses extended into the distal aorta; in 20 patients, the extent of disease assessed at duplex ultrasonography (US) could not be confirmed at angiography; in eight patients, stenosis was longer than 10 cm or occlusion was longer than 5 cm; in seven patients, lesions could not be crossed with a guidewire; and in three patients, there was extensive diffuse atherosclerotic changes of the vessels for which the type of PTA or stent used in this study protocol would have been insufficient. In patients with multiple unilateral or bilateral iliac stenoses, all lesions were assigned to the same treatment regimen. Multiple stenoses localized in one arterial segment (ie, the common iliac artery or external iliac artery) were classified as a single lesion.
Patients were randomly assigned to undergo either primary stent placement (group 1) or primary PTA with subsequent selective stent placement (group 2; stent placement performed in case of a residual mean pressure gradient of more than 10 mm Hg). Group 1 consisted of 143 patients (102 men, 71%) with a mean age of 58 years and 187 lesions in 181 legs. Group 2 consisted of 136 patients (99 men, 73%) with a mean age of 58 years and 169 lesions in 164 legs. In group 2, 59 patients (43%; 65 lesions) needed to undergo stent placement.
We used the following criteria for risk factors: For diabetes mellitus, patients were receiving medical treatment for diabetes mellitus; for hypertension, patients were receiving antihypertensive medication; a high creatinine level was that of greater than 133 µmol/L; for smoking, patients had been smokers in the last 10 years. On the angiogram obtained at the intervention, the radiologist noted the angiographic characteristics of the iliac artery lesions (location, grade, morphology, and length of the stenosis; patency of the runoff vessels). The radiologists (see Acknowledgments) had at least 5 years (F.J.A.B.) or 10 years (T.T.C.O., J.H.S., W.P.T.M.M., J.P.v.S., H.P., J.J.K., F.H.B.T., A.V.T.) of interventional experience. Baseline patient and angiographic characteristics were similar in the two treatment groups (13).
Five-year Follow-up
Patients were not assigned to a specific follow-up protocol after the initial treatment. Five years after treatment, we (W.M.K., W.P.T.M.M.) collected information on cardiovascular morbidity and mortality from the patients general practitioners and referring clinicians. Signs and symptoms, hospital discharge records, and additional information (such as images or electrocardiograms) were collected if events of interest occurred (possible end points). Also, if patients died, we searched the medical records for any cardiovascular events or interventions that occurred during the follow-up period.
Definition of End Points
The primary end point was reintervention in the iliac artery segment that was treated in the DIST study. We chose to consider the whole common or the whole external iliac artery that was treated as the site of reintervention, rather than only the part that was treated, because it is difficult to recognize the precise site treated with PTA. With our method, the same borders would apply for patients in both treatment groups. Secondary end points were interventions in the ipsilateral leg (endovascular intervention, amputation of lower extremities, sympathectomy, or vascular surgery), cardiovascular events (nonfatal stroke, nonfatal myocardial infarction, nonfatal rupture of the abdominal aortic aneurysm, end-stage renal failure, or nonfatal extracranial bleeding), cardiovascular interventions (endovascular intervention or vascular surgery), or death from any cause (Table 1).
|
Statistical Analysis
We analyzed data according to the intention-to-treat principle. We tested for statistical significance of differences (P < .05) between the two treatment groups in reinterventions of the treated iliac artery segment (per lesion), interventions in the ipsilateral leg (per leg), and cardiovascular morbidity and mortality (per patient) by using Kaplan-Meier survival analysis and the log-rank test.
Regression analysis was used to determine which variables were associated with iliac reinterventions and which were associated with myocardial infarction, stroke, and vascular mortality, by analyzing only one lesion per patient. If P < .15 at univariate analysis, variables were entered into a multivariable Cox regression model. Variables that were considered to be predictors were (a) the baseline patient characteristics of age, sex, classic cardiovascular risk factors (cholesterol level was used as a continuous variable), PAD symptoms, and exercise data and (b) the baseline angiographic characteristics of location, grade, morphology, and length of the stenosis, and the patency of run-off vessels.
In addition, we ignored randomization and looked at the whole group of 279 patients to compare the reinterventions between patients who underwent stent placement and those who underwent PTA alone.
| RESULTS |
|---|
|
|
|---|
Reintervention in the Iliac Artery Segment
In group 1, 33 (18%) of 187 iliac arteries that had been treated in the DIST study underwent 46 reinterventions (Table 2). Of these 46 reinterventions, 26 (57%) were endovascular, and in four of these cases they included stent placement. There were 20 (43%) surgical reinterventions performed involving the iliac artery segments, as follows: Seven patients underwent placement of eight aortobifemoral prostheses (a 49-year-old woman underwent placement of a second prosthesis after occlusion of the first); one patient underwent two and one patient underwent three sympathectomy sessions, each for lesions on both sides, which therefore equals surgical intervention in 10 lesions; and one patient underwent removal of both his stents, which were removed in two separate sessions.
|
Kaplan-Meier survival analysis showed no significant difference in reinterventions of the iliac arteries between group 1 and group 2 (P = .7) (Fig 1).
|
|
|
|
Most cardiovascular events and interventions involved the heart (myocardial infarction, PTA of the coronary arteries, coronary artery bypass grafting). In group 1, there were 23 cardiovascular events, of which nine were myocardial infarctions and six were strokes. In group 2, there were 17 cardiovascular events, of which nine were myocardial infarctions and five were strokes. In group 1, 11 of the 20 interventions that were not performed in the legs involved the heart. In group 2, 12 of 17 of these interventions involved the heart.
In group 1, 21 (15%) patients died during the follow-up period; of these, nine (43%) patients died of a vascular cause. In group 2, 22 (16%) patients died during the follow-up period; of these, 11 (50%) patients died of a vascular cause (Table 4). The exact 5-year mortality rate among all patients was 12% (34 of 279 patients). The exact 5-year mortality rate among patients with Fontaine stage I or II disease was 11%, and the rate among patients with Fontaine stage III or IV disease was 22%.
Predictive Variables
As we found no significant differences in end points between the two treatment groups, randomization was not included in the regression analysis. Results of univariate Cox regression analyses of baseline patient and angiographic characteristics for patients who did and those who did not undergo iliac reintervention are summarized in Tables 5 and 6. Variables that were predictors for reintervention according to the multivariate Cox model, adjusted for age, were as follows: female sex (hazard ratio, 2.1; 95% CI: 1.2, 3.7), Fontaine stage III or IV disease (hazard ratio, 2.0; 95% CI: 0.9, 4.7), or iliac stenosis greater than 2 cm in length (hazard ratio, 1.6; 95% CI: 0.9, 2.7).
|
|
Stent Placement versus PTA
When we compared the 252 lesions treated with stent placement (among both groups) and the 104 lesions treated with PTA alone, we found that there were no significant differences in regard to numbers of reinterventions in the treated iliac arteries, interventions in the ipsilateral leg, or cardiovascular events and interventions, or in regard to mortality rate.
| DISCUSSION |
|---|
|
|
|---|
After a mean follow-up period of 5.6 years, 15% of the patients had died, and half of these deaths were related to a vascular cause. Another 9% of patients had a nonfatal myocardial infarction or stroke. When we considered the iliac artery segment that was treated in the DIST study, 24% of patients needed to undergo one or more reinterventions in this segment, and more than half of these reinterventions were endovascular procedures. In 34% of patients, an additional intervention was needed in the ipsilateral leg, and in 4% of patients, a minor or major amputation was performed. Thus, less than half of the patients initially treated for iliac artery disease were completely without any event, intervention, or reintervention during the 5 years of follow-up.
These results show that PTA with selective stent placement in case of a pressure gradient of more than 10 mm Hg (after PTA) is an effective strategy. In the DIST study, this strategy led to stent placement in 43% of patients; if we had used the criterion of more than 5 mm Hg, which is recommended by the U.S. Food and Drug Administration (24), 77% of our patients would have undergone stent placement. Yet there were no further reinterventions needed in either treatment group of the DIST study. Results of other studies have shown that selective use of stents also yields good clinical results and that stent placement does not always seem to be necessary (8,9,25,26). Also, in our study, when we ignored randomization and instead compared all patients who underwent stent placement with the patients in whom PTA alone was sufficient, we found that even these apparently selected groups did not differ in regard to the number of iliac reinterventions performed. The results of our study show that stent placement does not lead to further reintervention at the site of implantation. This means that when stent placement, which is frequently considered easier and faster to perform than PTA followed by pressure measurement (which still leads to stent placement in 43% of patients, as found in the DIST study [13]), is used more liberally it does not seem to harm the patient either.
The number of iliac reinterventions performed in this study after more than 5 years (18% in group 1 and 20% in group 2) seems to be well in line with the results of other studies (5,8,9,11,25,2731). Schurmann et al (30) retrospectively investigated 110 patients who underwent iliac stent placement and found a 28% rate of reintervention, but this was over a 10-year period. Hassen-Khodja et al (8) compared PTA alone with PTA followed by selective stent placement (in case of unsatisfactory PTA results [dissection, >30% residual stenosis] or iliac occlusion) and found that reinterventions were needed in 20% of all cases after a mean follow-up period of 3 years. Not all technical failures needed reintervention; in some situations, patient and/or physician decided not to intervene (8,9,30). In our study, we can assume that several stenotic or occlusive lesions did not undergo reintervention because of limited clinical implications or because the physician did not consider reintervention as an option. It therefore seems clinically more relevant to study reintervention, rather than local patency, as an end point. A drawback of our study was that we could not identify whether a reintervention was at the exact location of the original treatment (ie, restenosis) or elsewhere within that same common or external iliac artery segment (ie, progression of the disease). However, we had no indications that there would be any difference in reinterventions between the two treatment groups.
On the other hand, findings of this study show that not only do patients with iliac artery disease often require reinterventions of the iliac artery segment and interventions of the ipsilateral leg, but they also have a high risk of other types of cardiovascular morbidity and, consequently, a high risk of mortality. This has also been found in several noninterventional studies of patients with PAD (18,20). We showed that patients treated for iliac artery disease have a 19% (53 of 279 patients) chance of a cardiovascular event or vascular mortality and that they have a 56% (156 of 279 patients) chance of any end point (event, intervention, mortality) after a mean follow-up period of 5.6 years. This again emphasizes that the real danger for patients with PAD is not amputation but rather cardiovascular complications and death. PAD should be viewed as a marker of potentially diffuse and dangerous atherosclerotic disease elsewhere in the body. This needs to be investigated and treated accordingly in every patient with PAD.
Schurmann et al (30) studied a population similar to that in our study, consisting mainly of patients with intermittent claudication, and found a 5-year mortality of 17%, compared with 12% in our study. Half of the deaths were due to a cardiovascular cause, which is similar to the findings in our study. Both of these 5-year mortality rates are considerably lower than the approximately 30% mortality in patients with intermittent claudication that was reported in a review by Dormandy et al (22), as well as by the TransAtlantic Inter-Society Consensus, or TASC (23). However, the studies referred to by Dormandy et al and by the TASC concerned patients who were treated with reconstructive surgery. The inclusion criteria of the DIST study and other studies of endovascular treatment have probably attracted patients with less advanced PAD. We found that the mortality rate for patients with rest pain was double that for patients with intermittent claudication (22% vs 11%, respectively). This also shows that the severity of PAD is an important predictor for death. In addition, the lower mortality rate found in recent studies can probably also be partly attributed to better treatment of atherosclerotic risk factors and early prevention of progression.
Prognostic factors for reintervention have been investigated frequently, and we were able to confirm some of the findings of these investigations. The presence of more severe PAD symptoms, such as critical ischemia, is a well-known prognostic variable for iliac reintervention (32,33). Authors of several studies have found the length and degree of stenosis to be predictive of a successful outcome (4,9,26,29,3234). We also found that longer stenoses were more at risk for reintervention than shorter stenoses. Another prognostic factor we found was sexwomen were more likely to undergo iliac reintervention. We do not have a good explanation for this, but this prognostic factor was also found in one other study (35).
A high creatinine level and a shorter walking distance were predictive for myocardial infarction, stroke, or vascular death. Abnormal kidney function is a well-known predictor of cardiovascular disease (3638). In this study, age was not found to be prognostic for reintervention or for myocardial infarction, stroke, or vascular death.
A drawback of the present study may be that information was collected retrospectively. This could have led to loss of information on events and interventions of interest. Although we were able to obtain information on all patients, this information was not always complete. Information on vascular interventions, especially, was sometimes found to be ambiguous and inconsistent over the consecutive patient reports. Also, the cause of death was sometimes not recorded or was unclearly recorded. However, we collected as much relevant information as possible, and a review board gave the final diagnosis and classification.
We conclude that the two treatment strategies, primary stent placement and PTA with selective stent placement, provide equal long-term clinical results. Patients with iliac artery disease are at high risk for cardiovascular events, interventions, and mortality.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, W.P.T.M.M.; study concepts, W.P.T.M.M.; study design, W.P.T.M.M., Y.v.d.G.; literature research, W.M.K., W.P.T.M.M.; clinical studies, W.M.K.; data acquisition, W.M.K., W.P.T.M.M.; data analysis/interpretation, W.M.K., Y.v.d.G., W.P.T.M.M.; statistical analysis, W.P.T.M.M., Y.v.d.G.; manuscript definition of intellectual content, editing, and revision/review, W.M.K., Y.v.d.G., W.P.T.M.M.; manuscript preparation and final version approval, all authors
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Dinh, S. Scovell, and A. Veves Review: Peripheral Arterial Disease and Diabetes: A Clinical Update International Journal of Lower Extremity Wounds, June 1, 2009; 8(2): 75 - 81. [Abstract] [PDF] |
||||
![]() |
C. J. White and W. A. Gray Endovascular Therapies for Peripheral Arterial Disease: An Evidence-Based Review Circulation, November 6, 2007; 116(19): 2203 - 2215. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. M. Klein, Y. v. d. Graaf, J. Seegers, J. H. Spithoven, E. Buskens, J. G. v. Baal, J. Buth, F. L. Moll, T. T. C. Overtoom, M. R. H. M. v. Sambeek, et al. Dutch Iliac Stent Trial: Long-term Results in Patients Randomized for Primary or Selective Stent Placement Radiology, December 21, 2005; (2005) 2382041053. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |