Comparison of popliteal artery aneurysm therapies Francisco Javier Serrano Hernando, MD, PhD, Isaac Martínez López, MD, Manuela María Hernández Mateo, MD, Manuel Hernando Rydings, MD, Luis Sánchez Hervás, MD, Rodrigo Rial Horcajo, MD, Guillermo Moñux Ducajú, MD, PhD, and Antonio Martín Conejero, MD, Madrid, Spain Objective: Surgical treatment of popliteal artery aneurysms (PAAs) has advanced over time. Constant development of new endovascular techniques has converted these into the most attractive methods available today. However, results for each endovascular procedure are still limited, and available data have emerged from studies examining only a small number of cases. The present retrospective review was designed to examine early and late results for PAA treatment and to identify possible factors associated with graft patency. Methods: This was a retrospective review of all PAAs treated from January 1993 to December 2013. Symptomatic and asymptomatic PAAs >2 cm treated using open surgery or an endovascular procedure were included. Kaplan-Meier curves and the Breslow test were used to analyze data. Results: A total of 171 aneurysms were treated in 142 men (mean age, 69.3 years); of these, 53.3% were asymptomatic and 18.7% presented as acute ischemia. Saphenous vein was used for bypass in 57.9% of the patients, expanded polytetrafluoroethylene (ePTFE) in 23.4%, and a stent graft in 18.7%. In the open surgical group, a popliteal-popliteal bypass was performed in 37.4% and a distal anastomosis to a tibial vessel was required in 14.4%. Good runoff (two to three vessels) was present in 69%. Perioperative mortality (30 days) was 1.8%. Of eight early occlusions recorded, five (2.9%) underwent reoperations. Major amputations were needed in five patients (all with previous acute ischemia). Median follow-up for the entire cohort was 49 months (range, 1-228 months). Primary and secondary patency rates at 24, 36, and 60 months were 76.3% and 89.5%, 73.4% and 87.4%, and 68.3% and 80.9%, respectively. Popliteal-popliteal bypasses showed better primary patency at 24 months when saphenous vein was used vs ePTFE (94.9% vs 79%; P [ .04); however, similar patency rates were recorded for short ePTFE bypasses and stent grafts (79% vs 79.7%). On multivariate analysis, only poor runoff emerged as an independent factor for worse primary patency (hazard ratio, 3.5; 95% confidence interval, 1.7-7.2; P [ .001). Conclusions: The open repair of PAA offers good long-term results, especially in asymptomatic patients, those undergoing elective surgery, and those showing good runoff. Given the also good midterm outcomes of endovascular treatment, this may be a feasible option in selected patients. (J Vasc Surg 2015;61:655-61.)

Popliteal artery aneurysms (PAAs) represent w70% of peripheral aneurysms. Despite this, their estimated incidence is 65 years and have two characteristic features: they are often bilateral, and in 40% to 60% of cases, aneurysmal disease is also observed at other levels. Most PAAs are asymptomatic, although up to 30% may show complications, especially those related to embolization to infrapopliteal vessels or aneurysm thrombosis. PAA rupture has only rarely been reported.1 Indications for the surgical repair of a PAA have not yet been well defined. However, a diameter >2 cm, especially in the presence of wall thrombus, and the presence of symptoms are those most widely accepted.2 The surgical treatment of PA has advanced over time from the first exclusion and revascularization procedure From the Department of Angiology and Vascular Surgery, Hospital Clínico San Carlos, Universidad Complutense de Madrid. Author conflict of interest: none. Reprint requests: Isaac Martínez López, MD, Hospital Clínico San Carlos, Department of Angiology and Vascular Surgery, Martín Lagos s/n, 28040 Madrid, Spain (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2015 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2014.10.007

performed by Goyanes at the start of the 20th century to the recent introduction of endovascular techniques.3 The technique described by Edwards et al4 in the late 1960s of PAA exclusion by ligation and revascularization by venous bypass is still today’s most widely used procedure, and it was not until 1994, that Marin et al5 reported the first endovascular repair of a PAA using a stent graft. However, results for endovascular procedures are still limited, and the data generated so far have emerged from studies that examined only a small number of patients. The present retrospective study was designed to examine early and late results for PAA treatment and to identify possible factors associated with graft patency. METHODS The protocol for this study was revised and approved by our Hospital Ethics Committee. A patient consent signature was not required before the study. Study design. Data were analyzed for all patients treated at our center from January 1993 to December 2013. Patients were included if they had a noncomplicated PAA >2 cm in diameter or had thrombosis or ischemia due to embolization. The PAA diagnosis was performed by Doppler ultrasound imaging or computed tomography, or both. Preoperative hemodynamic data were available for all patients, including systolic pressures and pulse volume recordings. Angiographies were conducted when there were clinical signs of limb ischemia or an abnormal 655

JOURNAL OF VASCULAR SURGERY March 2015

656 Serrano Hernando et al

Table I. Demographic data Variables a Age, years Male sex Tobacco use Hypertension Diabetes Hyperlipemia Chronic renal failure Coronary artery disease COPD

Entire cohort (N ¼ 171) 69.3 171 99 90 15 65 5 32 27

Open surgical group (n ¼ 139)

(40-87) (100) (57.9) (52.6) (8.8) (38) (2.9) (18.7) (15.9)

68.7 139 82 74 10 54 2 24 23

(40-87) (100) (58.9) (53.2) (7.2) (38.8) (1.4) (17.2) (16.5)

Endovascular group (n ¼ 32) 74.3 32 17 16 5 11 3 8 4

(56-88) (100) (53.1) (50) (15.6) (34.4) (9.3) (25) (12.5)

P .10 e .10 .48 .09 .73 .19 .32 .79

COPD, Chronic obstructive pulmonary disease. a Continuous data are shown as mean (range) and the categoric data are presented as number (%).

ankle-brachial index. Patient data were introduced prospectively in a specific database; thus, although the study was retrospective, data collection was prospective. Patients. A total of 171 PAAs were treated in 142 patients during the 20 years. Data were collected on demographics, cardiovascular risk factors, and associated comorbidities. Open surgery. The involved aneurysm was excluded by ligation, followed by saphenous vein bypass through a medial approach. This was the preferred technique, provided the vein was $3 mm in diameter and there was no prior phlebitis or varicose disease. In patients in whom the vein was unsuitable for bypass, an expanded polytetrafluoroethylene (ePTFE) prosthesis or stent graft was used. When there was no occlusive or aneurysmal disease of the superficial femoral artery (SFA), the proximal anastomosis was placed on the suprageniculate popliteal or the distal SFA (popliteal-popliteal “short” bypass). If this was not the case, the graft was sutured to the common femoral artery or proximal segment of the SFA. The saphenous vein was placed in situ when a common femoral artery anastomosis was needed, whereas reversed saphenous vein was used for the popliteal-popliteal bypasses. In most cases, the distal anastomosis was placed in the infrageniculate popliteal artery, although a tibial bypass due to disease at this level was sometimes required. After the intervention, patients received single-antiplatelet therapy with acetylsalicylic acid (100 mg once daily) during the first half of the study and, more recently, clopidogrel (75 mg once daily). Endovascular treatment. This was used in high-risk surgical patients or in those with no suitable vein for bypass, or both, provided they showed a favorable anatomy, defined as proximal and distal landing zone >10 mm, a caliber difference between the two segments of #2 mm, and a possible distal landing zone of the stent graft in the infrageniculate popliteal artery. The Viabahn or Hemobahn stent graft was used (W. L. Gore & Associates, Flagstaff, Ariz) and was implanted by surgical access through the ipsilateral femoral artery because of the largecaliber sheaths needed (8F-9F). Technical success was defined as complete exclusion of the aneurysm without leaks and good filling of infrapopliteal vessels observed in the postprocedural arteriography. After the procedure,

patients received simple antiplatelet therapy with ASA (100 mg once daily) until 2009, when our standard medical care changed to dual-antiplatelet therapy for 6 weeks consisting of ASA (100 mg) and clopidogrel (75 mg) and then clopidogrel (75 mg) indefinitely thereafter. Thus, aneurysm exclusion, followed by saphenous vein bypass, was the preferred technique, provided the vein could be used and the patient had reasonable surgical risk. When the vein was not available or in high-risk surgical patients, prosthetic ePTFE bypass was the alternative before the availability of the endovascular devices. However, after the introduction of the endovascular treatment, this became the preferred alternative as long as a favorable PAA anatomy was present. Follow-up. This consisted of clinical and hemodynamic (systolic pressures and pulse volume recordings) postintervention assessments at 1, 3, 6, and 12 months and yearly thereafter. In patients with a venous bypass or stent graft, the follow-up examination included Doppler ultrasound imaging. Clinical failure was defined as worsening of at least one stage of the Rutherford classification. Hemodynamic failure was defined as a drop in ankle-brachial index $0.15 and Doppler ultrasound detection of a velocity ratio $3. All patients with patent hemodynamic failure underwent reintervention. The chosen therapy was individualized for each patient, although endovascular reintervention was primarily considered in those with stent grafts. Statistical analysis. Data were compared using the c2 and Student t-tests, depending on the variable examined. Survival and patency were determined by Kaplan-Meier curves, and differences were assessed using the Breslow test. A P value of .05

53 118 91 32 48

(31) (69) (53.3) (18.7) (28)

47 92 68 32 18

(33.8) (66.2) (48.9) (23) (12.9)

6 (18.7) 26 (81.2) 23 (71.9) 0 3 (9.4)

4 17 23 4

(2.3) (9.9) (13.4) (2.3)

2 16 23 4

(1.4) (11.5) (13.4) (2.3)

2 (6.2) 1 (3.1) 0 0

.07 .012 .003 >.05

a

Continuous data are shown as the mean (range) and categoric data as number (%).

Fig 1. Kaplan-Meier curves show (A) primary patency (P1) and secondary patency (P2) rates and (B) limb salvage for the entire cohort of patients. N, number at risk; SD, Standard deviation.

Surgical indications and the anatomic characteristics of the femoropopliteal and distal sectors are described in Table II. The PAAs in 53.3% of patients were asymptomatic. In 18.7% of the patients, the indication for surgery was acute ischemia, usually due to thrombosis of the PAA or to distal embolization, and all were treated by open surgery. Different stages of chronic limb ischemia were seen in 28% of patients. Good runoff (two to three vessels) was present in 69% of patients. The aneurysm was excluded by venous bypass in 99 patients (57.9%), by prosthetic bypass in 40 (23.4%), and by a stent graft in 32 (18.7%). Within the open surgical group, 52 patients (37.4%) received a popliteal-popliteal short bypass, and 20 patients (14.4%) required an infrapopliteal bypass, all of them using the saphenous vein. Early results. The 30-day mortality was 1.8% (two patients died of acute myocardial infarction and one died of respiratory complications). Eight early occlusions were recorded (five prosthetic and three venous bypasses), five of which were reoperated on, three of them successfully. Five major amputations were performed in the first 30 days, all in patients treated for acute ischemia due to a thrombosed PAA. Primary and secondary patency and limb salvage rates

at 30 days were 94.1%, 95.4%, and 96.7%, respectively, for the entire cohort. The 30-day primary patency was 97.7% for venous bypass, 91.9% for prosthetic bypass, and 100% for stent grafts (P ¼ .27). Follow-up. All patients who came to follow-up visits received clinical and hemodynamic evaluations. Median follow-up was 49 months (range, 1-228 months) for the entire cohort, 49 months for the open surgery group, and 22 months for the endovascular group. Although no patients were lost to follow-up during the first postoperative 30 days, 26 patients were lost to follow-up in the first year. No patients in the endovascular group were lost to followup in the same period. Patient survival rates were 87.8% at 36 months and 82.6% at 60 months. During follow-up, 27 occlusions (14 prosthetic bypasses, 7 stent grafts, and 6 venous bypasses) occurred, and 12 patients underwent reintervention. A further 12 patients underwent reintervention due to hemodynamic failure caused by graft stenosis. Kaplan-Meier curves were constructed for primary and secondary patency rates (5 years) for the entire population (Fig 1, A). At 60 months, the affected limb was preserved in 94.5% of patients (two major amputations; Fig 1, B). No

JOURNAL OF VASCULAR SURGERY March 2015

658 Serrano Hernando et al

Fig 2. Kaplan-Meier curve shows (left) primary patency and (right) secondary patency rates for venous and prosthetic bypass.

Fig 3. Kaplan-Meier curve shows (left) primary patency and (right) secondary patency rates for venous and prosthetic popliteal-popliteal bypass vs stent graft.

significant effects on primary or secondary patency rates were observed between conventional bypass (P ¼ .16) vs endoluminal therapy (P ¼ .12), symptomatic (P ¼ .14) vs asymptomatic (P ¼ .53) PAA, or distal anastomosis to a tibial vessel (P ¼ .27) vs to the popliteal artery (P ¼ .32). Poor runoff (zero to one vessel) was linked to worse primary and secondary patencies, and a saphenous vein or popliteal-popliteal bypass graft was also correlated with an improved primary patency over a prosthetic or longer graft, respectively. A trend toward worse results was also obtained with emergency surgery. Within the group of open surgery patients, the 54-month primary and secondary patency rates were greater for venous than for prosthetic bypass (primary: 78% vs 54.9%; P ¼ .026; secondary: 90.2% vs 61.4%; P < .001, respectively; Fig 2). On multivariate analysis, only poor runoff emerged as an independent factor for worse primary patency (hazard ratio, 3.5, 95% confidence interval, 1.7-7.2; P ¼ .001). A comparison between popliteal-popliteal short bypasses (venous and prosthetic) and endovascular treatment

found no differences in primary 24-month patency among the three groups. However, a couple comparison showed 24-month better primary and secondary patency rates for venous bypass compared with prosthetic bypass (primary: 94.9% vs 79%; P ¼ .04; secondary: 100% vs 80.3%; P ¼ .008) and compared with stent graft (primary: 94.9% vs 79.7%; P ¼ .045; secondary: 100% vs 88.3%; P ¼ .008). Nevertheless, similar 2-year primary patency rates were recorded for prosthetic grafts compared with endoprosthesis (79% and 79.7%; P > .05; Fig 3). DISCUSSION The main objective of treating a PAA is to prevent limb loss due to thrombosis or distal embolization of the aneurysm. Nowadays, the two most widely accepted therapeutic options for the treatment of a PAA are exclusion, followed by venous or prosthetic bypass and the use of a stent graft. Since the introduction of endoluminal procedures, several studies have highlighted the benefits of a stent graft over the conventional open surgery approach. However, several

JOURNAL OF VASCULAR SURGERY Volume 61, Number 3

reports published in the past 30 years have revealed excellent outcomes of bypass surgery.6-11 These studies have described 40% to 45% asymptomatic patients and an w20% rate of emergency surgery due to acute ischemia, along with the use of saphenous vein in >65% of cases in most series. Cited primary patency is >65% at 5 years, and secondary patency and limb salvage rates are >85% in the long-term. All these rates are similar to those of the present patient series. In a prior study, we reported a primary patency rate for venous bypass of 75% at 10 years.12 A retrospective study examining 123 Veterans Affairs Medical Centers in the United States included 583 surgically treated PAAs in 537 patients from 1994 to 2005.13 This is the largest study to emerge from the United States. Perioperative mortality was 1.4%, and the reintervention rate was low (6.3%). Limb salvage rates were 99% at 30 days and 96.2% at 24 months. In 2007, the Swedish Vascular Registry presented its outcomes of the treatment of 717 PAAs with a mean follow-up of 7.2 years.14 A stent graft was used in 3.6% of cases. Excellent results were obtained in primary patency, with rates >80% at 12 months. There is therefore sufficient evidence that the surgical treatment of a PAA offers excellent limb salvage and survival outcomes in the midterm and long-term, even in high-surgical risk patients. The first reports of the endovascular treatment of PAAs were somewhat disappointing. Gerasimidis et al15 reported primary patencies of 47% at 1 year. In another series of 23 PAAs treated using the Hemobahn device, the 4-month occlusion rate was 22% and 1-year primary patency was 71%.16 Notwithstanding, more recent studies have improved on the results of endovascular therapy. In a study describing 23 cases conducted in 2006,17 1-year primary patency rates were 93% and secondary patency rates were 100%. In 2007, Tielliu et al18 reported the longer-term results of a larger trial of 73 PAAs, indicating a 5-year primary patency of 70% and secondary patencies of 76%. These improved results of endovascular therapy may be attributed to improvements in experience, devices, and postoperative medical treatment. There are few comparisons in the literature of the open and endovascular repair of a PAA. Curi et al19 reported 56 PAAs treated from 2000 to 2006, with a mean follow-up of 16.5 months. A stent graft was deployed in 26% of their cases. Although these authors observed no significant differences in patency and limb salvage between their two patient groups, it should be underscored that a prosthetic graft was used in 25% of cases in the open surgery group, that one-third of this group had one distal outflow vessel, and that the two patient groups also varied in the proportion of symptomatic patients (54% vs 13%; P < .05) and of acute disease onset (12% vs 0%; P < .05). In a prospective randomized trial comparing the open repair and endovascular treatment of asymptomatic PAAs, 15 PAAs were treated using the Viabahn stent graft, and

Serrano Hernando et al 659

a further 15 were surgically treated. Mean follow-up was 46 months.20 Results revealed no significant differences between the two groups: primary patency at 12 months was 100% for the open repair group and 87% for the endovascular repair group. Later in 2007,21 these same authors reported similar midterm results for 27 patients in their open surgery group and 21 in the endovascular treatment group, including no difference in 6-year patency. A meta-analysis of studies comparing the open and endovascular repair of PAAs was published in 2008.22 Only three studies were included, comprising 141 PAAs, of which only 26% were treated endovascularly. In the open repair group, an ePTFE graft was used in 21% of the cases. No significant differences were detected between the two treatments in the long-term, although the endovascular therapy group showed a greater risk of thrombosis and early reintervention. The authors thus concluded that to date, it was difficult to argue for endovascular treatment as the gold standard. Also recently published has been a review23 that included 5459 treated PAAs during a 25-year period. Of these, 41% of the patients were asymptomatic, 25% underwent intervention due to acute limb ischemia, and 69% had poor runoff. Only 6% of the patients received endovascular therapy for the PAA. The 3-year reported assisted primary and secondary rates were similar between both groups of treatment, and the 3-year amputation rate was

Comparison of popliteal artery aneurysm therapies.

Surgical treatment of popliteal artery aneurysms (PAAs) has advanced over time. Constant development of new endovascular techniques has converted thes...
1MB Sizes 2 Downloads 14 Views