| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Depts of Radiology (E.T., V.L.O., B.T.B., P.P., G.S.), Internal Medicine (J.R.C., R.W., P.V.N.), and Pathology (L.L.), Centre Hospitalier de l’Université de Montréal (CHUM), 3840 St Urbain St, Montreal, Quebec, Canada H2W 1T8, and the Dept of Radiology, Hôpital du Saint-Sacrement, Quebec City, Quebec, Canada (G.C.). Received Nov 30, 1999; revision requested Jan 14, 2000; final revision received Nov 28; accepted Dec 19. Address correspondence to E.T. (e-mail: eric.therasse.chum@ssss.gouv.qc.ca).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Fifty-three patients who underwent technically successful PTAA were compared with 24 patients who underwent aortic stent placement because of PTAA failure (19 patients) or ulcerated lesions (five patients) that otherwise would have been treated surgically because of the embolization hazard associated with PTAA alone. Clinical patency was defined as the absence or improvement of symptoms after the intervention. Hemodynamic patency was defined as a normal Doppler waveform in the common femoral arteries, an ankle-brachial index greater than 0.95, or the absence of a thigh-brachial pressure gradient.
RESULTS: Three-year clinical and hemodynamic patency rates, respectively, were 85% and 79% for PTAA and 69% and 43% for aortic stent placement. No morbidity was encountered. With use of the Cox proportional hazards model, two significant risk factors were retained for restenosis: unchanged smoking habit (P = .04) and small dilatation diameter (P = .001). Aortic stent placement, performed in patients with a smaller aortic diameter (10.3 vs 12.7 mm for PTAA), appeared to be a predictive factor for restenosis by using univariate analysis. By using the Cox proportional hazards model, however, the restenosis rates after PTAA and aortic stent placement were not significantly different.
CONCLUSION: When aortic diameter is taken into consideration, there is no evidence that clinical outcome after secondary aortic stent placement would be poorer than technically successful PTAA.
Index terms: Aorta, arteriosclerosis, 981.721 • Aorta, grafts and prostheses, 981.1268 • Aorta, interventional procedures, 981.1282 • Aorta, stenosis or obstruction, 981.721
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Percutaneous transluminal aortic angioplasty (PTAA) failures are uncommon but do occur because of elastic recoil, obstructive intimal dissection, or late restenosis (6). Since percutaneous stent placement provides better angiographic and hemodynamic results than percutaneous transluminal angioplasty (7), aortic stent placement is expected to have a better clinical outcome than PTAA. Unfortunately, despite good technical results and acceptable complication rates, aortic stent placement has not met these expectations in the few studies (6,8,9) published in the literature. However, the experience so far with aortic stent placement is limited, and long-term clinical follow-up data are scarce. Given the good outcomes provided by technically successful PTAA, the anticipated clinical benefits may be less marked after primary rather than after secondary aortic stent placement (to treat PTAA failure).
The aim of this study was to evaluate the long-term clinical and hemodynamic effectiveness of aortic stent placement in patients in whom intended infrarenal PTAA failed by comparing the outcome in these patients with that of patients who underwent technically successful PTAA.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PTAA was the intended procedure for all these patients (ie, PTAA was the treatment that the radiologist would have first envisioned to treat them). Percutaneous revascularizations were performed for accepted clinical indications, and written informed consent was obtained from all patients after the nature of the procedures had been explained. Given that six patients underwent stent placement for late PTAA restenosis (defined as the recurrence of stenosis greater than 50% after treatment), a total of 100 procedures were evaluated.
For the purpose of comparison, the percutaneous revascularizations were divided into two categories: (a) technically successful PTAA and (b) aortic stent placement as treatment of failure of intended PTAA. Figure 1 shows a synopsis of all percutaneous procedures and of the patients included and excluded in both the PTAA and stent groups.
|
Technical failures were defined as a systolic pressure gradient of at least 10 mm Hg without administration of vasodilators after PTAA or residual stenosis of greater than 30% when the pressure gradient was not available. The remaining 53 patients (44 women, nine men; mean age, 51 years; age range, 30–84 years) were included in the technically successful PTAA group. Among these 53 patients, 50 presented with intermittent claudication, two presented with rest pain, and one had gangrene.
Aortic Stent Placement as Treatment of Failure of Intended PTAA: Stent Group
Aortic stent placement was performed in 30 patients, and technical results and clinical follow-up were available for all of them. We excluded six patients because they underwent stent placement for late abdominal aortic restenosis after PTAA (secondary patency was not considered in this study). The remaining 24 patients (21 women, three men; mean age, 52 years; age range, 32–76 years) were included in the stent group. Among these 24 patients, 23 presented with intermittent claudication, and one presented with rest pain. Two patients who had claudication also showed signs of blue toe syndrome.
Stent placement was performed for the following indications: (a) PTAA was an immediate technical failure in 19 patients. These technical failures were due to residual pressure gradients of 10–60 mm Hg (mean, 21.7 mm Hg) in 16 patients, large dissections after PTAA in two patients, and 90% residual stenosis in one patient. (b) Markedly ulcerated aortic lesions were present in five patients. In these patients with ulcerated aortic lesions, PTAA was considered a failure because they would have been treated with surgery by their referring surgeons, who were concerned about the embolization hazard associated with PTAA alone. We included these patients because PTAA was the intended treatment (the treatment the radiologist would have first envisioned to treat the patient).
All aortic interventions were performed with the femoral approach and local anesthesia. Dilatation diameter was chosen on the basis of the diameter of normal aortic segments in both the PTAA and stent groups. Because most diagnostic arteriograms were obtained with digital equipment, the exact length of the stenosis could not be evaluated precisely. However, all stenoses could be dilated with balloon catheters of 3–4 cm in length without having to reposition the catheter, and, in fact, most lesions were roughly estimated to be 2 cm or shorter. Intraarterial boluses of heparin (Heparin Leo; Leo Pharma, Ajax, Ontario, Canada; 3,000–5,000 IU) were given during the procedure in both groups. Pressure gradients were measured across the lesion before and after the interventions, respectively, in 53 (69%) and 61 (79%) of 77 procedures (53 PTAAs; 24 stent placements). Pressure gradients were not measured in all cases because the angiographic image alone was sometimes considered enough to motivate percutaneous revascularization or to aid in evaluation of the result after the procedure.
In neither the PTAA nor the stent group was systemic anticoagulation recommended, but aspirin was given routinely after the procedure unless contraindicated. In the PTAA group, two balloons were inflated simultaneously (kissing balloon technique) in 31 (58%) patients, and a single balloon was used in the remaining 22 (42%) patients. In the stent group, all but one of the aortic lesions were treated with one stent each, and a total of 25 endoprostheses (21 Palmaz P 308 and four Palmaz P 394; Johnson & Johnson Interventional Systems, Warren, NJ) were implanted. Stents were always delivered through a long 8–10-F vascular sheath by using a single angioplasty balloon catheter.
The patients were followed up on an outpatient basis by their referring physicians, who were mostly vascular surgeons. In the spring of 1998, the patients were asked to return for Doppler ultrasonography and clinical evaluation by an internist (including J.R.C., R.W., P.V.N.); the evaluation included segmental pressure measurements of the lower extremities with the ankle-brachial index (ABI), clinical examination of distal pulses, and completion of a questionnaire identifying claudication symptoms. However, some patients did not return for that reevaluation either because they were lost to follow-up (two of these patients were dead) or because they did not want to come back for reevaluation.
We defined clinical patency as either the absence or improvement of symptoms after PTAA or stent placement. This definition corresponds to improvements of +2 and +3, according to the Joint Council of the Society for Vascular Surgery and the International Society for Cardiovascular Surgery (SVS/ISCVS) classification system (10). We chose not to include +1 improvement because patients with a slight ABI increase but without symptom improvement could have been included.
Hemodynamic patency was defined as a normal triphasic Doppler waveform in the common femoral artery, a resting ABI greater than 0.95, or the absence of a thigh-brachial pressure gradient in either limb at rest. When follow-up arteriography was available, hemodynamic patency was defined as aortic stenosis less than 50% in diameter. Follow-up angiography was performed only in patients with recurrent symptoms who were considered for reintervention.
All patients were advised to walk regularly and to stop smoking, but they did not undergo a supervised walking program after PTAA or stent insertion, and we did not evaluate the effect of walking on long-term success. Only primary patency after either the first PTAA (in the PTAA group) or stent placement (in the stent group) was considered. The patency duration for patients who initially underwent technically successful PTAA and who subsequently underwent stent insertion for restenosis was calculated from the date of the PTAA to the date of symptom recurrence.
Comparisons of demographic, morphologic, and hemodynamic factors between patients in the PTAA group and those in the stent group were analyzed with 
2 testing. The persistence of clinical and hemodynamic patency during follow-up was reported by means of Kaplan-Meier life-table analysis. The factors potentially associated with symptom-free survival of all patients (PTAA and stent groups) were studied by using the Cox proportional hazards regression analysis, with a hierarchical (block) introduction of factors and a forward stepwise regression method to test their statistical significance (11). For each covariate, graphic checks of proportionality of hazards were first performed. Formal checks were derived from models, including time-dependent covariates. The preliminary selection of variables was based on univariate Cox models with an 
value of .15.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
The stent placement procedure had immediate technical success in all but one patient (in whom a residual 15 mm Hg systolic gradient pressure after stent placement was considered to be due to small aortoiliac vessels). This patient had no angiographic evidence of residual stenosis, but the iliac vessels were small. Since pressure measurements were obtained from the distal external iliac artery sheath and from a 5-F catheter within the abdominal aorta above the stenosis, the gradient may have been related to the presence of the catheter within a small iliac artery. There was no residual angiographic stenosis after stent placement in any patient. Five patients in the stent group had minor femoral hematomas, but none required surgical drainage or blood transfusion. One (3%) of 30 patients who underwent aortic stent placement had small external iliac arteries and had iliac artery thrombosis secondary to introducer sheath placement. This was treated successfully with thrombolysis and percutaneous transluminal angioplasty and was the only procedure-related complication in the stent group.
Clinical follow-up information was available in 47 of the 53 patients in the PTAA group, and hemodynamic follow-up data were available in 45 of these patients. Both patients in whom only clinical information (no hemodynamic follow-up data) was available were free of symptoms at their last follow-up. Consequently, their level of improvement was considered to be at least SVS/ISCVS grade +2. The outcomes of six patients without any follow-up information were considered as missing data. Clinical and hemodynamic follow-up data were available for all patients in the stent group.
In the PTAA group, follow-up ranged from 3 to 114 months (mean, 50 months); two patients died of unrelated causes 6 and 7 years after PTAA. At last follow-up, 25 (53%) of 47 patients reported no symptoms of claudication, 11 (23%) were improved, and 11 (23%) were considered to have clinical treatment failures. Therefore, clinical patency was 77% (36 patients) at the end of the follow-up period. In seven of the 11 patients with clinical failures, the failures were due to aortic restenosis, and in four, they were caused by progression of peripheral disease. Four patients with aortic restenosis were treated with aortic stent placement. Two additional patients with aortic restenosis were treated with repeated PTAA, and another patient underwent no additional intervention. In patients with progression of peripheral disease, two were treated with femoral percutaneous transluminal angioplasty, one with iliac percutaneous transluminal angioplasty, and in one, no reintervention was performed. Figure 2 shows the clinical evolution and status of the 47 patients in the PTAA group at last follow-up.
|
In eight of the nine patients, clinical treatment failures were due to aortic restenosis, while one was caused by progression of peripheral disease (an iliac stenosis was treated with iliac percutaneous transluminal angioplasty and stent placement). Both patients who initially had blue toe syndrome showed no evidence of new atheroembolic events during follow-up.
One of the patients with aortic restenosis was treated with intrastent PTAA 10 months after the initial procedure, and this patient was free of symptoms at 46 months of follow-up after the last intervention. Another patient with aortic restenosis was treated with repeat percutaneous transluminal angioplasty and stent placement in the same lesion 6 months after the first aortic stent placement procedure. However, this patient had symptomatic restenosis one more time 3 months later, and she underwent surgical aorto-aortic bypass. Three other patients presented with symptoms of recurrence due to aortic thrombosis (one patient) or aortic stenosis (two patients) 11–26 months (mean, 16 months) after stent placement, and all of them underwent bypass graft surgery and surgical resection of the aorta in which stents were placed. The remaining three patients with aortic restenosis underwent no additional intervention. Figure 3 shows the clinical evolution and status of patients at last follow-up in the stent group.
|
|
|
|
At last follow-up in the stent group, mean resting ABI improved from 0.73 to 0.88, and hemodynamic patency was achieved in 12 (50%) of 24 patients. In the 12 patients with hemodynamic failures, eight also had clinical treatment failures, two remained clinically improved, and two had no symptoms (despite ABIs of 0.85 and 0.88). On the other hand, one of the 12 patients with hemodynamic patency was considered to have a clinical treatment failure because of progression of peripheral disease.
Figure 5 demonstrates the results of Kaplan-Meier life-table analysis of primary hemodynamic patency. The hemodynamic patency rates after 1 and 3 years were, respectively, 95% and 79% for the PTAA group and 71% and 43% for the stent group.
|
Table 4 presents the estimates of risks of clinical recurrence according to demographic and technical factors. Dilatation diameter, use of a stent, and changes in tobacco consumption were predictive of symptom recurrence by using univariate analysis. However, from these three variables used in the Cox proportional hazards model analysis, only dilatation diameter and changes in tobacco consumption were retained as significant correlates of survival symptom improvement, the risk of recurrence being reduced as dilatation diameter increased and tobacco consumption decreased. The other parameter (use of a stent) lost all predictive value as soon as dilatation diameter was introduced in the model, suggesting that smaller dilatation diameter was associated with stent implantation (10.4 mm for the stent group vs 12.9 mm for the PTAA group, P < .001).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The effectiveness of secondary aortic stent placement is a different issue. The benefit of placing a stent after failed percutaneous transluminal angioplasty has been well established for iliac lesions (for which stent placement is now an accepted procedure) (17), but to our knowledge it has never been demonstrated in abdominal aortic lesions. Although many patients are reported (18) to be free of symptoms, despite residual pressure gradients greater than 10 mm Hg after PTAA, the lowest residual gradient that could be left after a revascularization procedure is unknown.
However, secondary stent insertion may be indicated in cases of marked residual pressure gradients after PTAA. The threshold for a marked pressure gradient after PTAA was empirically established as a systolic gradient of 10 mm Hg in this series. Those who performed the percutaneous aortic procedures in our study generally consider this pressure gradient to be important in iliac arteries. Although a randomized clinical trial in which aortic stent placement after a failed PTAA is compared with no additional treatment after failed PTAA would be a better way to measure the clinical benefit of stent placement in cases with poor PTAA results, such a trial would be currently unethical, given the established role of stent placement after failed percutaneous transluminal angioplasty of iliac stenosis.
The experience reported (3,8,9,19–22) so far with infrarenal aortic stent placement is summarized in Table 5. The series (3,8,9,17,19–22) contain a small number of patients with short follow-ups, reported mostly as mean duration rather than as values derived through life-table analysis. The indications for treatment, technical details, and type of endovascular devices, as well as the results, vary considerably.
|
In the only other study (9) in which the results of PTAA were compared with those of aortic stent placement, mean follow-up of the stent group was 9 months, and the number of patients was too small to compare both groups with life-table analysis. In addition, the investigators did not specify whether aortic stent placement was a primary intervention or a bailout procedure to treat failed PTAA, an important point that may be associated with a different long-term clinical outcome, as was the case in our investigation. Finally, the reported absence of a significant difference of outcome between PTAA and aortic stent placement in that study may have merely resulted from a lack of statistical power due to the small number of patients included.
As in other series (8,19,20,22), we obtained an excellent immediate technical success rate after aortic stent placement by using Palmaz stents, and the morbidity with stent insertion was low and not notably different from that associated with PTAA alone. In our series, however, the rate of late symptomatic restenosis was surprisingly high after aortic stent placement despite excellent initial angiographic and hemodynamic results. The disappointingly low 3-year patency rate of 69% after stent placement (85% after PTAA) may be difficult to explain at first and could even discourage aortic stent placement in patients with suboptimal PTAA results.
Although the majority of patients who underwent stent placement in our study had poor results after PTAA, the final morphologic and hemodynamic results after stent placement were nearly perfect, and they were even better than after technically successful PTAA. Therefore, it is surprising that nearly complete abolition of residual stenosis and intimal dissection did not result in a lower stenosis recurrence rate after stent insertion, compared with PTAA. Although tobacco consumption was a positive predictive factor for symptom recurrence, there was no statistically significant difference in smoking habit changes after the intervention between patients in the PTAA group and those in the stent group.
In our study, the diameter of aortic dilatation and diabetes mellitus were the only significantly different factors between both groups. Cox regression analysis did not demonstrate diabetes mellitus to be a predictor of aortic restenosis. However, smaller-diameter aortic dilatation was a statistically significant independent risk factor for symptomatic restenosis (P < .001). Inclusion of patients with PTAA failures into the stent group, therefore, introduced a selection bias in our series, because these patients had a smaller abdominal aorta than those who underwent a technically successful PTAA. Stent insertion as a variable is, therefore, confounded due to the higher number of patients in whom stents were placed who had the small aortic diameter variable, which explains why it appeared to be a statistically significant risk factor (P = .04) for symptomatic restenosis with univariate analysis. With the Cox proportional hazards model, when dilatation diameter was considered, stent insertion was not a risk factor for restenosis.
The relatively high restenosis rate in smaller aortas is difficult to explain. In our study, the restenosis rate in the stent group was not better than that after coronary artery stent placement, and this is surprising considering the greater diameter of the abdominal aorta compared with that of coronary arteries (23). Aortic restenosis after either aortic stent placement or PTAA is often assumed to be due to intimal hyperplasia, although there are few pathologic data in the literature to substantiate this presumption. Given our histopathologic findings, the pathophysiologic findings of infrarenal aortic restenosis might be different from those of the usual intimal hyperplasia observed in the more peripheral vascular system. Complications of progressive atherosclerosis could be more determinant than neointimal hyperplasia for aortic restenosis, and prevention of restenosis could possibly be more efficiently achieved with more aggressive control of atherosclerosis risk factors. In fact, a smaller abdominal aorta could merely be an indicator of an accelerated atherosclerotic process that is markedly enhanced by dyslipidemia and tobacco use in these relatively young women.
Our study was not a randomized trial and, as such, suffered from the defects of retrospective studies in general. There was a referral bias in that only patients with the most symptomatic aortic stenoses were referred for PTAA; the other patients underwent conservative treatment. There was also confounding with indication bias (stents were inserted because of technically failed PTAA) in the design of this study, but it should have disadvantaged the stent group. The fact that patients with the worst technical results underwent stent placement should also have led to a selection bias unfavorable to the stent group in our study. This is important, especially given that survival analyses did not show significant differences between the technically successful PTAA group (PTAA group) and the failed PTAA plus stent placement group (stent group).
Biases that may result from not having performed a final follow-up examination in all patients may have had slightly unfavorable effects on the survival curves of both groups but should not have affected one group more than the other in our series. The power of a study to demonstrate a difference is related to the size of the study. Given the relatively limited number of patients included in our study, a small difference in outcome between both groups may have been missed, and the results of our analyses cannot support a claim of equivalence.
Despite a relatively high restenosis rate, aortic stent placement was clinically effective in the majority of our patients. Considering that aortic stent placement was performed in cases of PTAA failure, PTAA in many of these patients would probably have been considered as a clinical failure. Although the advantage of stent placement over PTAA in treating patients with ulcerated lesions is speculative, our patients who had such lesions probably benefited from stent placement because they would have undergone surgery performed by their referring surgeons who were concerned about the embolization hazard associated with PTAA alone.
In summary, infrarenal aortic stent placement is safe, has a high technical success rate, and is minimally invasive. Small dilatation diameter and smoking are predictive of bad outcomes. Although our findings could not be used to support a claim of equivalence, there was no evidence that clinical outcome after aortic stent insertion is poorer compared with successful PTAA when the size of the abdominal aorta is considered. Therefore, despite a relatively high restenosis rate, we believe that aortic stent insertion is a safe and effective way to treat failed PTAA.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, E.T., G.C., V.L.O., G.S.; study concepts, E.T., G.C., G.S.; study design, E.T., G.C., V.L.O., G.S.; definition of intellectual content, E.T., G.C., V.L.O., G.S.; literature research, E.T., G.S., J.R.C.; clinical studies, J.R.C., R.W., P.V.N.; experimental studies, E.T., G.C., V.L.O., G.S.; data acquisition, J.R.C., R.W., P.V.N.; data analysis, P.P., B.T.B., L.L., E.T.; statistical analysis, E.T., G.S., J.R.C.; manuscript preparation, E.T., G.S., V.L.O.; manuscript editing, B.T.B., P.P., L.L.; manuscript review, G.S., V.L.O., J.R.C., L.L.; manuscript final version approval, G.C., V.L.O., J.R.C., P.P., G.S., E.T.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
