Early complications and long-term outcome after open surgical treatment of popliteal artery aneurysms: Is exclusion with saphenous vein bypass still the gold standard?
Article Outline
Background
Popliteal artery aneurysms (PAAs) are rare, but thromboembolic complications may result in limb loss. To define complications and outcomes after open surgical repairs, we reviewed our experience.
Methods
Clinical data of patients with PAA seen between 1985 and 2004 at Mayo Clinic, Rochester, Minnesota, were reviewed and outcomes in 289 patients with open revascularization were analyzed. Kaplan-Meier method with log-rank tests, χ2, and Wilcoxon rank sum tests were used for analysis.
Results
A total of 358 PAAs were treated in 289 patients, consisting of 281 (97%) men and eight (3%) women. There were 133 (46%) unilateral and 156 (54%) bilateral PAAs with a mean diameter of 2.9 cm (range, 1.5 to 9 cm). Abdominal aortic aneurysm (AAA) was more frequent with bilateral than unilateral PAAs (65% [101/156] vs 42% [56/133] P = .001). There were 144 (40%) asymptomatic limbs (group 1), 140 (39%) had chronic symptoms (group 2), and 74 (21%) had acute ischemia (group 3). Great saphenous vein (GSV) was used in 242 limbs (68%), polytetrafluoroethylene (PTFE) in 94 (26%), and other types of graft in 22 (6%). Early mortality was 1% (3/358), all in group 3 (4% [3/74]). Six of seven patients with perioperative myocardial infarctions belonged to group 3 (8%). The 30-day graft thrombosis rate was 4%, with 1% in group 1 (1/144), 4% in group 2 (5/140), and 9% in group 3 (7/74). All six early amputations (8%) were in group 3, five with failed bypass (4 PTFE, 1 GSV). Mean follow-up was 4.2 years (range, 1 month to 20.7 years). The 5-year primary and secondary patency rates were 76% and 87%, respectively, higher with GSVs (85% and 94%) than PTFE (50% and 63%, P < .05). Seven recurrent PAAs (2%) required reintervention. The 5-year freedom from reintervention was 100% after endoaneurysmorrhaphy vs 97% after ligations (P = .03). Five-year limb salvage rate was 97% (85% in group 3). There was no limb loss in group 1 and none in group 2 with GSV. In group 3, preoperative thrombolysis reduced the amputation rate in class II patients with marginally threatened limbs (96% vs 69%, P = .02).
Conclusion
Acute presentation of PAA continues to carry high mortality and cardiac morbidity; although preoperative thrombolysis appears to improve results, the 8% early and 15% late amputation rates remain ominous. Early elective repair is recommended because these patients had no surgical mortality, a low rate of complications, and asymptomatic patients had no limb loss at 5 years. GSV and endoaneurysmorrhaphy continues to be the gold standard for open repair of PAA.
Popliteal artery aneurysms (PAAs) are rare, with a prevalence of <0.01% reported in hospitalized patients.1 These aneurysms account for up to 70% of all peripheral arterial aneurysms.1 The most frequent and severe complication is thromboembolism, with subsequent acute limb ischemia and major amputation rate of up to 30% in surgical series.2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 The disease is bilateral in approximately half of the cases, and one in two patients with PAAs will also harbor an abdominal aortic aneurysm (AAA).
Open surgical treatment includes in situ arterial replacement or a bypass using autologous or prosthetic graft. To decrease late recurrence of the aneurysm, endoaneurysmorrhaphy has been preferred over simple ligation of the artery proximal and distal to the aneurysm.17 Thrombolysis has been used to improve outflow in patients with acute presentation.9, 18, 19
Several reports, including one prospective randomized study, have advocated endovascular treatment of PAAs over open surgical management.20, 21, 22, 23 To provide standards for comparison, the aim of our study was to define contemporary complications and long-term outcomes of patients treated with conventional open surgical techniques.
Methods
The clinical data of patients with the diagnosis of PAA seen at Mayo Clinic, Rochester, Minnesota, between 1985 and 2004, were reviewed, and outcomes in 289 patients who underwent open revascularization for 358 PAAs were analyzed. The study was approved by the Mayo Foundation Institutional Review Board. Clinical categories for both acute and chronic limb ischemia and criteria for primary and secondary patency were defined as recommended by Rutherford et al.24 Cardiac, pulmonary, and renal risk factors were documented using criteria of the Department of Veterans Affairs National Surgical Quality Improvement Program (NSQIP).25 PAA was defined as focal dilation of the popliteal artery by ≥50% over the expected normal diameter (0.9 ± 0.2 cm), as measured with ultrasonography,26 or a diameter >1.5 cm when measured intraoperatively.27
Patients were divided into three groups: asymptomatic limbs (group 1), symptoms of chronic limb ischemia (group 2), and limbs with acute limb ischemia (group 3). Local compression was defined if symptoms of numbness, tingling, pain, or feeling of fullness in the popliteal fossa or signs of edema or popliteal vein thrombosis were documented in the chart. Indications for operation included acute limb ischemia with a salvageable limb, PAAs with compressive symptoms or disabling claudication, asymptomatic PAA with aneurysm >2 cm in size, or those with mural thrombus or evidence of previous thromboembolism.8
Operative technique
Most revascularizations were performed through a median approach. If the aneurysm involved the proximal popliteal artery, inflow for the bypass was the superficial femoral or common femoral artery. The distal popliteal artery, one of the tibials, or the peroneal artery was the site of the distal anastomosis. Proximal and distal ligation was usually used for small aneurysms, and endoaneurysmorrhaphy was performed at the discretion of the surgeon, usually in patients with a large aneurysm (>2 cm) or if large genicular arteries were noted on preoperative imaging.
The aneurysm was opened, thrombus evacuated, and the lumen oversewn, usually without transection of the medial head of the gastrocnemius muscle. A thigh tourniquet facilitated endoaneurysmorrhaphy in a bloodless field. If the posterior approach was used, the aneurysm was opened and feeding arterial branches were ligated or oversewn. Reconstruction was performed with autologous or prosthetic bypass, or interposition graft. Prosthetic graft was used at the discretion of the surgeon, usually only if a suitable autologous vein was not available.
Indications for thrombolysis included patients with acute limb ischemia in a viable extremity. Decisions to proceed with thrombolysis were made at the discretion of the interventionists. In group 3, 34 patients had intra-arterial thrombolysis (preoperative, 24; intraoperative, 14; both, 4). Of the 24 patients who had preoperative lytic treatment, 23 presented with threatened limb ischemia (class II) and one with irreversible limb ischemia (class III). Urokinase was used at a dosage of 1000 U/min to 4000 U/min in 16 patients. Tissue plasminogen activator (t-PA) was given at 0.5 mg/h in eight patients. Mean duration of thrombolysis was 19 ± 13 hours. For intraoperative thrombolysis, seven patients had urokinase (60,000 U to 250,000 U) and three had t-PA (0.5 mg to 1 mg). Activated partial thromboplastin time and fibrinogen were monitored during thrombolysis.
Information on concomitant aneurysms was obtained from reports of imaging studies and from history of previous aneurysm repairs. Of 289 patients, 271 underwent imaging of the contralateral popliteal fossa or had repair of a contralateral popliteal aneurysm previously; 274 patients underwent imaging of the aortoiliac vessels or had repair of an aortoiliac aneurysm previously. Follow-up information was obtained from the medical records and mailed questionnaires. Mailing was used to establish survival and limb loss. Patients were advised to return every 6 months, and patency rates were based on the last imaging studies.
During the time interval of this study, an additional 10 PAAs were treated by primary major amputation (n = 2), ligation alone (n = 2), endoaneurysmorrhaphy alone (n = 4), lumbar sympathectomy (n = 1), and stent graft (n = 1). One of these patients presented with aneurysm rupture. These patients were excluded from our analysis.
Statistical analysisDescriptive statistics, including means, standard deviations, ranges, and proportions were calculated as appropriate. Rates of patency, reintervention for recurrent PAA, amputation per limb, and survival were estimated using the Kaplan-Meier method with log-rank tests to compare groups. Multivariate analyses were performed using Cox proportional hazards regression. Length of stay in the hospital and the intensive care unit (ICU) were compared between groups using Wilcoxon rank sum tests. The χ2 test or Fisher exact test, as appropriate, were used to assess associations with binary outcomes such as early morbidity. Values of P < .05 were considered statistically significant for all analyses. Analyses were performed using SAS 9 software (SAS Institute, Inc, Cary, NC).
Results
Demographic data
Between 1985 and 2004, 494 patients were diagnosed with 651 PAAs at Mayo Clinic. Open surgical revascularizations were done in 289 patients for 358 PAAs; they were the subjects of this study (Table I). Of these, 281 men (97%) and eight women (3%) had 133 (46%) unilateral and 156 (54%) bilateral PAAs. At least one remote aneurysm was present in 192 patients (66%), more frequently in those with bilateral vs unilateral PAAs (79% [123/156] vs 52% [69/133], P = .001). Of the patients who had both femoral and PAA, 74% also had AAA. AAA was present in 157 patients (54%), more frequently in patients with bilateral vs unilateral PAAs (65% [101/156] vs 42% [56/133], P = .001; Fig 1).
Table I. Demographic data and risk factors of 289 patients with 358 popliteal artery aneurysms
| Data | No (%) |
|---|---|
| Patients | 289 (100) |
| Male | 281(97) |
| Female | 8(3) |
| Age, mean y (range) | 70(17-88) |
| Risk factors⁎ | |
| Cardiovascular | 128(44) |
| Pulmonary | 41(14) |
| Renal | 3(1) |
| Hypertension | 192(66) |
| Diabetes mellitus | 46(16) |
| Current smokers | 81(28) |
⁎As defined by Department of Veterans Affairs National Surgical Quality Improvement Program.25 |
Fig 1.
Prevalence of concomitant aneurysms in 299 patients with popliteal artery aneurysms.
Clinical characteristics
Ultrasound imaging confirmed the diagnosis in 311 limbs (87%), 341 limbs (95%) were studied with contrast arteriography, 12 (3%) had computed tomography (CT) scan or CT angiography (CTA), and 12 (3%) had magnetic resonance (MR) angiography (MRA). The diameter of the aneurysm was measured with ultrasound in 309 limbs, CT scan in one, and MRA in two. In 40 limbs studied with contrast arteriography alone, intraoperative measurements were used when documented.
Thrombus was present in 337 PAAs (94%), and 95 (27%) had a completely thrombosed popliteal artery. In group 3, 66% (49/74) had a thrombosed popliteal artery. Runoff was assessed in 349 (97%) limbs: 341 (98%) were assessed with contrast arteriography and eight (2%) with CTA arteriography. A total of 144 PAAs (40%) were asymptomatic (group 1), 140 (39%) had chronic limb ischemia (group 2), and 74 (21%) had acute limb ischemia (group 3; Table II). Mean aneurysm diameter was 2.9 ± 1.1 cm (range, 1.5 to 9 cm). Current or previous deep vein thrombosis (DVT) was present in 20 limbs (6%), 36 (10 %) had symptoms of local compression by the aneurysm, 28 (8%) limbs had blue toe syndrome, and no aneurysms ruptured.
Table II. Clinical data of 358 limbs with popliteal artery aneurysms
| Data | N limbs with PAA (%) |
|---|---|
| Clinical class | |
| Group 1 (asymptomatic) | 144(40) |
| Group 2 (chronic limb ischemia)⁎ | 140(39) |
| 1 Mild claudication | 32(23) |
| 2 Moderate claudication | 37(26) |
| 3 Severe claudication | 21(15) |
| 4 Ischemic rest pain | 29(21) |
| 5 Minor tissue loss | 21(15) |
| 6 Major tissue loss | 0(0) |
| Group 3 (acute limb ischemia)⁎ | 74(21) |
| I Viable | 28(38) |
| IIa Marginally threatened | 31(42) |
| IIb Immediately threatened | 12(16) |
| III Irreversible | 3(4) |
| Total | 358 |
| Mean (range) | |
| Size of aneurysm (cm) | 2.9 ± 1.1(1.5-9) |
| Group 1 | 2.6 ± 0.8(1.5-5) |
| Group 2 | 3.0 ± 1.3(1.5-9)† |
| Group 3 | 2.9 ± 0.9(1.5-5)† |
†P < .05 compared with group 1. |
Intraoperative data
All 358 PAAs were repaired with bypass or interposition graft: 337 (94%) were performed through a medial approach, and 21 (6%) through a posterior approach (Table III). Pathologic studies of 236 PAAs showed degenerative atherosclerosis in 232 aneurysms (98.3%), fibromuscular dysplasia in three (1.3%), and thromboangiitis obliterans in one (0.4%).
Table III. Surgical data of 358 operations for popliteal artery aneurysms
| Surgical data | Presentation, n (%) | Total | ||
|---|---|---|---|---|
| Group 1 | Group 2 | Group 3 | ||
| Type of surgery | ||||
| Bypass | 140(97) | 134(99) | 74(100) | 348(97) |
| Endoaneurysmorrhaphy | 50(36) | 39(29) | 26(35) | 115(33) |
| Ligation | 87(62) | 86(64) | 43(58) | 216(62) |
| Excision | 3(2) | 7(5) | 5(7) | 15(4) |
| Bypass alone | — | 2(2) | — | 2(1) |
| In situ interposition | 4(3) | 6(1) | — | 10(3) |
| Conduit | ||||
| Autogenous graft | 108(75) | 101(72) | 50(68) | 259(72) |
| GSV | 97 | 95 | 50 | 242 |
| Non-reversed | 36(37) | 39(41) | 18(36) | 93(38) |
| Reversed | 61(63) | 56(59) | 32(64) | 149(62) |
| SSV | 5 | 2 | — | 7(2) |
| Non-reversed | 1 | 1 | — | 2(40) |
| Reversed | 4 | 1 | — | 5(60) |
| Arm vein | 4 | 2 | — | 6(2) |
| Spliced vein | 2 | 2 | — | 4(1) |
| Prosthetic graft | 36(25) | 39(28) | 22(30) | 97(27) |
| PTFE | 34(94) | 38(97) | 22(100) | 94(26) |
| Dacron | 2(6) | 1(3) | — | 3(1) |
| PTFE-vein sequential | — | — | 2 | 2(1) |
| Total grafts | 144 | 140 | 74 | 358 |
Early outcomes
MortalityThe 30-day overall mortality was 1% (3/358). All deaths occurred in group 3, for an early mortality of 4% (3/74). The causes of death were myocardial infarction, cardiac arrhythmia, and adult respiratory distress syndrome, in one patient each. Two of the three deaths were preceded by major amputation.
Limb lossThe 30-day amputation rate was 2% (6/358). All amputations, one below the knee and five above the knee, were performed in group 3 patients (8% [6/74]). One patient had a patent graft. Five of the group 3 amputations were performed after failed bypass grafts consisting of polytetrafluoroethylene (PTFE) in four, and great saphenous vein (GSV) in one.
Morbidity and graft failureSystemic complications occurred in 27 patients (8%), most frequently in group 3 (19% [14/74]). Seven patients (2%) had perioperative myocardial infarctions (group 1, 0%; group 2, 1% [1/140]; group 3, 8% [6/74]; Table IV). By 30 days, 13 grafts (4%) had thrombosed (group 1, 1 [1%]; group 2, 5 [4%]; group 3, 7 [9%]; Table IV).
Table IV. Early complications
| Complication | Number of patients (%) | Total (n = 358) | ||
|---|---|---|---|---|
| Group 1 (n = 144) | Group 2 (n = 140) | Group 3 (n = 74) | ||
| Systemic morbidity | 3(2) | 10(7)⁎ | 14(19)⁎† | 27(8) |
| Cardiac | 1(1) | 3(2) | 8(10)⁎† | 12(3) |
| Cardiac arrest | 1(1) | 1(1) | 2(3) | 4(1) |
| Myocardial infarction | 0 | 1(1) | 6(8)⁎† | 7(2) |
| Arrhythmia | 0 | 1(1) | 0 | 1(1) |
| Pulmonary | 0 | 2(1) | 4(5) | 6(2) |
| Acute renal failure | 0 | 2(1) | 5(7)⁎† | 7(2) |
| Deep vein thrombosis | 2(1) | 6(4) | 2(3) | 10(3) |
| Wound | 7(5) | 9(6) | 12(16)⁎† | 28(8) |
| Wound infection | 4(3) | 6(4) | 3(4) | 13(4) |
| Wound dehiscence | 1(1) | 0 | 2(3) | 3(1) |
| Hematoma | 3(2) | 1(1) | 8(11)⁎† | 12(3) |
| Lymph leak | 0 | 2(1) | 3(4) | 5(1) |
| Graft failure | 1(1) | 5(4) | 7(9)⁎ | 13(4) |
⁎P < .05 compared with group 1. |
†P < .05 compared with group 2. |
Median ICU and hospital length of stays were 1 and 6 days in group 1, and 1 and 8 days in group 2, respectively, both of which were shorter than the respective 2 and 12 days in group 3 (P < .0001).
ThrombolysisThe 30-day primary patency rate was 94% ± 4% for 34 patients with thrombolysis vs 87% ± 5% for 39 patients without thrombolysis (P = NS). The limb salvage rate was 91% ± 5% vs 92% ± 4% (P = NS). The primary patency and limb salvage rates were better in limbs with class II ischemia (threatened limb ischemia) with preoperative thrombolysis than without thrombolysis (96% ± 4% vs 80% ± 9%, P = .02; 96% ± 4% vs 85% ± 8%, P = .02, respectively).
The rate of bleeding complications was 20.5% (7/34). Two patients had hematuria, one of them also had hemoptysis, and hematomas developed in the surgical wounds of five patients. There was no stroke or death due to thrombolysis.
Late results
PatencyMean follow-up was 4.2 years (range, 1 month to 20.7 years). Five-year primary and secondary patency rates were 76% ± 3% and 87% ± 2%, respectively; they were higher in GSV grafts (85% ± 3% and 94% ± 2%) than in PTFE grafts (50% ± 7% and 63% ± 7%, P < .05; Fig 2). GSV grafts had superior patency rates in all three groups. In group 3, secondary patency rates at 1 year were higher in patients with GSV grafts than in those who had PTFE (96% ± 3% vs 67% ± 12%, P < .001; Fig 3). In group 1, the 5-year secondary patency rate of GSV grafts was 98% ± 2% compared with 87% ± 8% for PTFE grafts (P < .001).
Fig 2.
Cumulative patency of 336 grafts after popliteal artery aneurysm repair with great saphenous vein (GSV) or polytetrafluoroethylene (PTFE) grafts.
Fig 3.
Cumulative patency of 72 grafts in group 3 (acute presentation) after popliteal artery aneurysm repair with great saphenous vein (GSV) or polytetrafluoroethylene (PTFE) grafts.
In limbs with class II acute ischemia (threatened limb ischemia),24 22 of 43 patients were available for follow-up. Preoperative thrombolysis appeared to improve primary graft patency at 1 year (84% ± 9% vs 62% ± 12%, P = .02).
Univariate analysis revealed multiple factors that were more frequent in patients who developed graft occlusion (Appendix, online only).28 Symptomatic patients, poor runoff, distal anastomosis with pedal arteries, and PTFE grafts were significant predictors for loss of primary patency in a Cox proportional hazards model (Table V).
Table V. Multivariate analysis of risk factors predicting loss of primary patency
| Variable | HR (95% CI) | P |
|---|---|---|
| Chronic symptoms | 2.16(1.20-3.88) | .01 |
| Acute presentation | 2.57(1.24-5.30) | .01 |
| Poor runoff ⁎ | 2.44(1.38-4.31) | .002 |
| Tibial bypass | 1.66(0.94-2.91) | .08 |
| Pedal bypass | 5.44(1.56-18.96) | .008 |
| PTFE | 12.03(6.47-22.36) | <.0001 |
| GSV, nonreversed vs reversed | 2.53(1.28-5.00) | .008 |
⁎No tibial or peroneal artery visualized at the ankle level. |
To salvage a failing (n = 21) or failed (n = 46) graft, 67 reinterventions were performed on 41 patients during follow-up. Reinterventions for recurrent PAA were performed on seven (2%) of 358 limbs at a mean of 35 months (range, 2 to 79 months). Four had symptoms of local compression, two had acute ischemia, and one had chronic ischemia with rest pain. One of seven grafts was occluded. All limbs had previous proximal and distal ligation of the popliteal artery. The sac was opened through the posterior approach in each patient, the hematoma was evacuated, and feeding branches were ligated or oversewn. One patient underwent repeat grafting with PTFE. The 5-year freedom from reintervention owing to PAA recurrence was 98% ± 1%, and 100% in those who underwent endoaneurysmorrhaphy during the first operation vs 97% ± 2% in those with arterial ligation during revascularization of the limb (P = .03).
Limb salvageOverall limb loss was 3% (12/358) and was 14% (10/74) in group 3. The 5-year cumulative limb salvage rate was 97% ± 1%: 100% for group 1, 99% ± 1% for group 2, and 85% ± 4% for group 3. Patients with GSV grafts had better limb salvage at 5 years than those with PTFE grafts (99% ± 1% vs 90% ± 3%, P < .001). In group 3, the 5-year limb salvage with GSV graft was 93% ± 4% vs 66% ± 11% with PTFE grafts (P = .003; Fig 4). There was no limb loss in group 1 and none in group 2 if GSV was used for bypass. In group 3, preoperative thrombolysis in 26 lower limbs with class II acute ischemia (threatened limb) appeared to reduce the amputation rate (96% ± 4% vs 69% ± 11%, P = .02).
Fig 4.
Cumulative limb salvage of patients (group 3, acute presentations) after popliteal artery aneurysm repair with great saphenous vein (GSV) or polytetrafluoroethylene (PTFE) grafts.
Of the 107 patients who died during follow-up, cardiac disease was the most frequent cause of death (32%, 34/107). Five-year survival rate was 75% ± 3%.
Discussion
Aneurysm of the popliteal artery is a disease of men. It is frequently bilateral and often associated with AAA. In a review of the English literature that included 1673 patients with PAAs,29 97% were men, a rate exactly the same as our observation. Dawson et al29 found 50% had bilateral PAAs and 36% had AAA. In our series, 54% of PAAs were bilateral and 54% had AAA. We observed an increased prevalence of AAAs in patients with bilateral (65%) vs unilateral (42%) PAAs.
Thromboembolism is the most frequent complication of PAAs; rupture is rare. Local compression increases the risk for popliteal vein thrombosis, 6% of the patients in our study had current or previous DVT. Duplex scan is excellent for screening, to confirm mural thrombus, and for follow-up. Contrast arteriography was frequently used in this retrospective study, but CT scan and three-dimensional CTA has become the most important preprocedure imaging study. It defines inflow and runoff and identifies concomitant aneurysms elsewhere. In patients with renal insufficiency, MRA is an excellent alternative.
Indications for interventions are well defined for patients who present with acute ischemia and for most of those who have chronic ischemic symptoms. Patients with severe claudication, rest pain, and tissue loss are considered for elective repair, and patients who have symptoms of local compression are also candidates for surgical repair.
Indications for asymptomatic patients and for those with mild-to-moderate claudication are less well defined. Factors affecting surgical decision include the rate of development of symptoms, aneurysm size, mural thrombus, evidence of previous thromboembolism, risk of surgical repair, and long-term surgical outcome.
Michaels and Galland6 used Markov decision analysis to determine optimum treatment of PAAs. To justify a conservative approach, mortality and limb loss after repair of symptomatic PAAs have to be close to those observed after elective repair; alternatively, an asymptomatic patient group with an event rate of <10% per year has to be identified.6 A review of 12 publications revealed that development of symptoms averaged 14% per year (range, 5% to 24%).6 Lowell et al8 observed outcome in 67 patients with PAAs initially managed conservatively, and symptoms developed in 18% during a mean follow-up of 17 months. Factors predicting the development of symptoms included size >2 cm, poor runoff, and mural thrombus (P < .05).8
Data to correlate aneurysm size with risk of thromboembolism have been sparse. Whitehouse et al4 observed that asymptomatic patients and those with claudication have smaller aneurysms than those with acute ischemia or local compressive symptoms. However, Ascher et al30 found that in a group of patients with 34 popliteal aneurysms, those with smaller aneurysms developed more thrombotic complications and a higher incidence of thrombosis and clinical symptoms than those with larger aneurysms. Mural thrombus in small aneurysms (< 2 cm) was as frequent (64%) as in large ones (70%).30
We found a correlation between aneurysm size and clinical presentation: symptomatic patients presented with larger aneurysms. Increasing size in addition to development of symptoms were indications to intervene in 30% of 58 asymptomatic patients in one prospective multicenter study.9 Galland et al31 found that the mean expansion rate of PAAs was 1.5 mm/y for aneurysms <20 mm, 3.0 mm/y for size 20 to 30 mm, and 3.7 mm/y for size >30 mm. Distortion of the shape of the aneurysm in one study was also more frequent in symptomatic patients.32
Mortality and complications after elective vs emergency repair are major factors affecting the decision to treat asymptomatic PAAs early. In a review of 13 surgical series, Michaels and Galland6 found that asymptomatic patients had 0.4% mortality and a 0.8% early rate of limb loss, whereas symptomatic patients had 4.7% mortality and an 18.2% rate of early amputations.6 A sensitivity analysis revealed that with a 1% elective mortality, it takes 34 months before surgical repair will be the preferred treatment of asymptomatic patients. Fortunately, mortality for repair in asymptomatic patients is <1% in most contemporary surgical series,13, 23, 33 and we observed no elective mortality and a low rate of complications. Several authors advocate early repair of even small asymptomatic PAAs in good-risk patients.3, 4, 6, 34, 35 High-risk patients with an increased chance of surgical mortality (2% to 3%) are frequently observed; those with suitable anatomy and high risk for thromboembolic complications are now considered for endovascular treatment.20, 21, 22, 23
In patients with acute ischemia, we observed elevated morbidity and mortality, and prolonged ICU and hospital stays. An increased rate of complications in this group was also reported by others,13, 15, 23, 33 although one publication on 51 patients, including 14 with emergency repair, reported no early mortality or cardiac complications.14 Acute ischemia in our series carried an 8% early risk and 15% late risk of limb loss.
The GSV, implanted through a median approach, has been our preferred graft. Advantages include easy harvesting of the vein and easy access to anastomotic sites to assure optimal inflow and outflow. The proximal anastomosis was the femoral artery in two thirds of the cases, and an infrapopliteal artery was used distally in one fourth of the operations. The benefits of using the tibial arteries for the distal anastomosis, if needed, were emphasized by Aulivola at al.14
Endoaneurysmorrhaphy should be performed for all but the small, thrombosed aneurysms to prevent late recurrence.17 Depending on the location of the PAA, the sac is approached either from below or from above the knee incision by retracting the medial head of the gastrocnemius muscle. A bloodless field can be assured using a thigh tourniquet. The aneurysm is incised, and careful evacuation of the thrombus is followed by oversewing the genicular branches and suturing the posterior to the anterior wall. Gelfoam or surgical fibrillar can be placed in the sac to aid thrombosis. All reinterventions in this study were performed in patients who underwent revascularization with proximal and distal arterial ligation, without endoaneurysmorrhaphy.
The posterior approach, used only in a small group of our patients, was advocated by Beseth and Moore.16 This is an effective approach to avoid recurrence of the aneurysm because the sac is opened and all branches are ligated or oversewn under direct vision. Interposition grafting is usually done with a prosthetic graft, most frequently with PTFE. Access to the distal superficial femoral artery or to the tibial arteries through this approach is limited. In the experience of Beseth and Moore, 13 patients required a medial approach because of proximal (n = 11) or distal (n = 2) extension of the disease; so 70% of the cases were suitable for posterior repair. Pulli et al15 recently reported that 37% of 156 patients with PAA were treated through a posterior approach and 61% through a medial approach.
Selection of graft material has become somewhat controversial because a prosthetic graft is a good size match to the enlarged popliteal artery and a sizable number of patients with PAA have good-to-excellent runoff assuring high flow and prolonged patency of a prosthesis. Pulli et al15 found no difference in patency at 60 months between 118 PTFE grafts (71.5%) and 34 vein grafts (79.9%), although the number of limbs at risk during this time interval was not reported. Beseth and Moore16 implanted 25 PTFE and five Dacron grafts through the posterior approach in patients with PAA, two with acute symptoms, with a median follow-up of 21.5 months. Twenty-five patients had two-vessel or three-vessel runoff. The 2-year primary and secondary patencies were 92.2% and 95.8%, with a limb salvage of 100%.
Although no controlled studies are available, most data from the literature indicate superior long-term patency of vein grafts vs prosthetic grafts. In a systemic review of literature that included 2445 PAAs in 1673 patients, Dawson et al29 showed patency at 5 years of 77% to 100% for vein grafts vs 29% to 74% for prosthetic grafts. The type of conduit, the presence of symptoms, and the quality of the runoff correlated with operative results. In our experience, GSVs performed best: 5-year primary and secondary patency rates were 85% and 94% for GSV, and 50% and 63% for PTFE grafts. When GSV was used, no limb loss was observed in patients who underwent elective repair for asymptomatic PAA or for chronic ischemic symptoms.
To decrease the high rate of amputations, intra-arterial thrombolysis has been used successfully in patients with acute thromboembolism.9, 18, 19 Carpenter et al7 reported a 100% limb salvage rate for combined preoperative thrombolysis and bypass grafting compared with 57% for surgery alone; however, the benefits of preoperative thrombolysis have not always been confirmed by other investigators.10, 11, 13, 15, 36 In our experience, 1-year graft patency and limb salvage rates appeared to improve with preoperative thrombolysis. We continue to use thrombolysis in patients who present with mild-to-moderate acute limb ischemia (class I, IIa).
As with all retrospective studies, our review also has certain limitations. Although general guidelines in management as outlined in our report were followed, the selection of graft material, surgical technique, and indications for elective surgery were at the discretion of the vascular surgeon who performed the operation. There was no randomization to evaluate different techniques.
Conclusion
Our study confirmed that surgical treatment of acute presentation of PAA continues to carry elevated mortality and cardiac morbidity. Although preoperative thrombolysis appears to improve results, the 8% early and 15% late amputation rates with acute presentation remains ominous. Asymptomatic PAAs with thrombus or those with evidence of distal embolization should be considered for repair. PAAs ≥2 cm should be repaired even without mural thrombus. Early elective repair is recommended to good-risk patients with PAAs because operation in this group can be performed with no surgical mortality, a low rate of complications, and no limb loss for asymptomatic patients at 5 years. In our experience GSV, and endoaneurysmorrhaphy continues to be the gold standard for open repair of PAA.
Author contributions
Appendix
Additional material for this article may be found online at www.jvascsurg.org.
Appendix (online only). Univariate analysis of risk factors predicting primary graft patency in 358 limbs with popliteal artery aneurysm
| n | 5-year patency | HR | P | |
|---|---|---|---|---|
| High surgical risk⁎ | 358 | .07 | ||
| No | 308 | 77.5 | 1.000 | |
| Yes | 50 | 68.8 | 1.695 | |
| Groups | 358 | .003 | ||
| Group 1 | 144 | 88.0 | 1.000 | |
| Group 2 | 140 | 71.5 | 2.087 | |
| Group 3 | 74 | 63.1 | 2.469 | |
| Number of runoff vessels | 349 | <.001 | ||
| 0 | 57 | 48.8 | 3.031 | |
| 1 | 147 | 77.3 | 1.320 | |
| 2 | 127 | 83.3 | 0.682 | |
| 3 | 18 | 93.3 | 1.000 | |
| Pedal arch | 230 | .003 | ||
| Present or seen | 83 | 86.1 | 1.000 | |
| Not present or not seen | 147 | 62.5 | 3.482 | |
| Urgency of surgery | 358 | <.001 | ||
| Elective | 325 | 79.6 | 1.000 | |
| Emergency | 33 | 44.9 | 2.906 | |
| Surgical approach | 358 | .68 | ||
| Posterior | 21 | 76.9 | 1.000 | |
| Medial | 337 | 76.3 | 1.206 | |
| Type of exclusion | 344 | .55 | ||
| Endoaneurysmorrhaphy | 115 | 79.3 | 1.000 | |
| Ligation | 216 | 75.2 | 1.240 | |
| PAA thrombosis | 358 | .009 | ||
| Patent | 260 | 80.1 | 1.000 | |
| Completely thrombosed | 98 | 66.1 | 1.760 | |
| Graft† | 336 | <.001 | ||
| GSV | 242 | 85.4 | 1.000 | |
| PTFE | 94 | 50.3 | 5.623 | |
| GSV | 242 | .002 | ||
| Reversed | 149 | 91.9 | 1.000 | |
| Nonreversed | 93 | 72.2 | 2.648 | |
| Proximal anastomosis | 358 | <.001 | ||
| Distal SFA | 136 | 86.4 | 1.000 | |
| AK PA | 102 | 81.9 | 1.121 | |
| Proximal and mid-SFA | 36 | 67.5 | 2.324 | |
| Other | 50 | 64.2 | 2.922 | |
| CFA | 34 | 26.7 | 7.406 | |
| Distal anastomosis | 358 | .0001 | ||
| BK popliteal bypass | 268 | 78.0 | 1.000 | |
| Tibial bypass | 86 | 74.3 | 1.455 | |
| Pedal bypass | 4 | — | 8.082 | |
| Concomitant surgery | 358 | .02 | ||
| No | 225 | 80.9 | 1.000 | |
| Other aneurysm repair | 59 | 68.7 | 1.806 | |
| Others interventions | 74 | 65.5 | 1.839 |
⁎Defined as age >80 years, myocardial infarction ≤6 months before surgery, severe class IV angina or severe valvular heart disease, left ventricular ejection fraction <30%, cardiac stress test positive for ischemia, severe chronic obstructive or restrictive pulmonary disease (forced expiratory volume in 1 second <1.0), and serum creatinine >1.8 mg/dL.28 |
†Only GSV grafts and PTFE were compared, other autologous grafts and Dacron grafts were excluded. |
References
- . Multiple arteriosclerotic arterial aneurysms. Arch Surg. 1972;105:338–344
- . Progression of popliteal aneurysmal disease following popliteal aneurysm resection with graft: a twenty year experience. Surgery. 1976;4:426–432
- . A review of one hundred and forty-seven popliteal aneurysms with long-term follow-up. Surgery. 1981;90:1009–1014
- Limb-threatening potential of arteriosclerotic popliteal artery aneurysms. Surgery. 1983;93:694–699
- . Surgical management of popliteal aneurysms: trends in presentation, treatment, and results from 1952 to 1984. J Vasc Surg. 1986;3:125–134
- . Management of asymptomatic popliteal aneurysms: the use of a Markov decision tree to determine the criteria for a conservative approach. Eur J Vasc Surg. 1993;7:136–143
- . Popliteal artery aneurysms: current management and outcome. J Vasc Surg. 1994;19:65–73
- Popliteal artery aneurysms: the risk of nonoperative management. Ann Vasc Surg. 1994;8:14–23
- . A multicenter study of popliteal aneurysms. J Vasc Surg. 1994;20:171–177
- Preoperative intraarterial thrombolysis before surgical revascularization for popliteal artery aneurysm with acute ischemia. Ann Vasc Surg. 2000;14:360–364
- Acute leg ischemia from thrombosed popliteal artery aneurysms: role of preoperative thrombolysis. Eur J Vasc Surg. 2002;23:251–254
- . Management of popliteal aneurysm. Br J Surg. 2002;89:1382–1385
- . Surgery of popliteal artery aneurysms: a 12-year experience. J Vasc Surg. 2003;37:586–593
- Popliteal artery aneurysm: a comparison of outcomes in elective versus emergent repair. J Vasc Surg. 2004;39:1171–1177
- Surgical management of popliteal artery aneurysms: which factors affect outcomes?. J Vasc Surg. 2006;43:481–487
- . The posterior approach for repair of popliteal artery aneurysms. J Vasc Surg. 2006;43:940–945
- . Fate of excluded popliteal artery aneurysms. J Vasc Surg. 2003;37:954–959
- . Infra-inguinal aneurysms: outcome for patient and limb. Br J Surg. 1991;78:996–998
- Acute limb deterioration during intra-arterial thrombolysis. Br J Surg. 1993;80:1118–1120
- Percutaneous endovascular treatment of peripheral aneurysms. J Cardiovasc Surg (Torino). 2000;41:871–883
- Wallgraft endoprosthesis for the percutaneous treatment of femoral and popliteal artery aneurysms. J Endovasc Ther. 2002;9:76–81
- . Endovascular treatment of popliteal artery aneurysms: results of a prospective cohort study. J Vasc Surg. 2005;41:561–567
- Open repair versus endovascular treatment for asymptomatic popliteal artery aneurysm: results of a prospective randomized study. J Vasc Surg. 2005;42:185–193
- Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg. 1997;26:517–538
- The Department of Veterans Affair’s NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care (National VA Surgical Quality Improvement Program). Ann Surg. 1998;228:491–507
- . Suggested standards for reporting on arterial aneurysms. J Vasc Surg. 1991;13:444–450
- . Lower extremity aneurysms. In: Rutherford RB editors. Vascular surgery. 6th ed.. Philadelphia, PA: Elservier; 2005;p. 1534–1551
- Perioperative complications and early outcome after endovascular and open surgical repair of abdominal aortic aneurysms. J Vasc Surg. 2004;39:497–505
- . Atherosclerotic popliteal aneurysm. Br J Surg. 1997;84:293–299
- . Small popliteal artery aneurysms: are they clinically significant?. J Vasc Surg. 2003;37:755–760
- . Expansion rates of asymptomatic popliteal artery aneurysms. Eur J Vasc Surg. 2004;27:382–384
- . Popliteal aneurysms: distortion and size related to symptoms. Eur J Vasc Surg. 2005;30:534–538
- The value of duplex surveillance after open and endovascular popliteal aneurysm repair. J Vasc Surg. 2005;41:936–941
- . Aggressive surgical management of popliteal artery aneurysms. Am J Surg 1983;145:498-502. asymptomatic popliteal aneurysm. Br J Surg. 1990;77:1132–1135
- Limb salvage and popliteal aneurysms: advantages of preventive surgery. Eur J Vasc Endovasc Surg. 2000;19:496–500
- Success of thrombolysis as a predictor of outcome in acute thrombosis of popliteal aneurysms. J Vasc Surg. 2002;35:487–493
Competition of interest: none.
Additional material for this article may be found online at www.jvascsurg.org.
CME article
PII: S0741-5214(06)02235-X
doi:10.1016/j.jvs.2006.12.011
© 2007 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.