Surgical technique and long-term results after popliteal artery aneurysm repair: Results from 717 legs
Article Outline
- Abstract
- Patients and methods
- Results
- Discussion
- Conclusions
- Swedvasc participants
- Author contributions
- References
- Copyright
Background
This study investigated the importance of surgical technique on long-term outcome after treatment of popliteal artery aneurysms (PAAs).
Methods
Records from 571 patients (717 legs) primarily operated on for PAAs were identified in the Swedish Vascular Registry. Surgical approach, type of graft, and anastomotic and ligation techniques were studied. After mean 7.2 years (range, 2 to 18 years) information on amputation was obtained for all patients, and 190 patients were re-examined with ultrasound imaging.
Results
The approaches used were medial (medial approach group, MAG) in 87%, posterior (PAG) in 8.4%, endovascular in 3.6%, and other in 1.4%. Primary patency at 1 year with venous and prosthetic grafts was 85% vs 81% in the PAG (P = .719) and 90% vs 72% in the MAG (P < .001). Sixty-three legs (8.8%) were amputated ≤1 year, and 80 (11%) had been amputated at re-examination or by the end of follow-up. The median time from operation to amputation of 17 legs amputated after 1 year was 3.1 years (range, 1.1 to 9.8 years). The frequency of late amputation was 3.7% (2/54) in the PAG and 2.6% (15/571) in the MAG. In a Cox regression model, age (odds ratio [OR] 1.06/year, P < .001), emergency procedure (OR 2.67, P < .001), and prosthetic graft (OR 2.02, P = .008) were independently associated with long-term amputation rate. The risk of expansion of the excluded PAA at re-examination was 33% in the MAG and 8.3% in the PAG (P = .014). It was not affected by the ligation technique used.
Conclusions
The risk of late amputation was higher with prosthetic grafts. Operation with a posterior approach decreased the risk of expansion.
The primary aim of the management of popliteal artery aneurysm (PAA) is to prevent thromboembolism and amputation.1 The secondary objective is to prevent aneurysm expansion and rupture.2 With the recognition of the high complication rate of untreated PAA in late 1940s and start of the 1950s, most surgeons have been aggressive in the approach to this problem.3, 4
The most common treatment in that time was resection of the PAA with interposition of an autogenous vein using a posterior approach to the popliteal space, which was often combined with resection of the aneurysm.2 In 1969, Edwards described a technique of bypassing the aneurysm and ligating the popliteal artery immediately proximal and distal to the aneurysm through two small medial incisions above and below the knee joint.5 Resection of the aneurysm was reserved for patients with local symptoms in the popliteal fossa. Many surgeons have adopted this technique through the years. From 1994 and onwards, percutaneous endovascular treatments of PAA were reported.6, 7, 8, 9, 10
PAA is the most commonly treated aneurysm outside of the aortoiliac and the intracranial arteries.11 PAA is more prevalent among men and is often associated with abdominal aortic aneurysm (AAA) as well as with other peripheral aneurysms.1 Even if PAA is fairly common, no single surgeon or institution has enough patients to study this disease with appropriate scientific methods, and consequently, the management remains controversial.12
The Swedish Vascular Registry (Swedvasc)13 was created in 1987. Since 1994, >90% of vascular surgical procedures performed in Sweden have been registered prospectively,14 offering a unique possibility to study large groups of unselected patients treated at multiple institutions in routine clinical practice. In the Swedvasc protocol, a large number of variables are registered prospectively, such as preoperative risk factors, comorbidities, indications for the procedure, anatomic inflow and outflow, graft used, and type of operation. Survival, complications, patency, redo operations, and amputations are registered at 30 days and at 1 year.
The general epidemiologic findings of this study, including symptoms, presentation, the prevalence of synchronous aneurysms, risk factors for amputation at 1 year, time trends, and survival have previously been reported.15 A subgroup of 229 patients presenting with acute ischemia, and in particular the effect of preoperative thrombolysis, was described in another article.16
The aim of this study was to analyze, in the largest group of patients ever studied to our knowledge, the importance of surgical approach, type of graft, anastomotic and ligation used, on medium and long-term outcome.
Patients and methods
In November 2002, data for 717 primary operative treatments for PAA among 571 patients were retrieved from the Swedvasc (Fig 1). The diagnosis of PAA was determined by clinical examination and radiologic examination or ultrasound imaging in 680 legs (95%) and on clinical examination only in 37. Date of amputation and complications were prospectively registered in the Swedvasc during the first year of follow-up, and these data were crosschecked against the complete patient records. After 1 year, data were available from case records, telephone interviews, and re-examinations, as will be described.
Fig 1.
Flow chart shows the study design and breakdown of data on popliteal artery aneurysm (PAA) retrieved from the Swedish Vascular Registry. *Information on amputation on all legs, regardless of re-examination.
To validate the completeness of data, the numbers registered in the Swedvasc were compared with those of the Swedish Hospital Discharge Register (SHDR), used for reimbursement, through procedure codes. Furthermore, internal validity was examined by analyzing the frequency of registration among those identified as having been operated on for bilateral disease.
The 337 patients alive in January 2005 were invited to participate in a telephone interview, and 240 (71%) accepted and were asked if they had undergone amputation, among other questions. Every Swedish citizen has a unique identity code, which made it possible to crosscheck survival against the national population registry with 100% accurate survival data. The surgeons responsible for the Swedvasc at the local hospitals checked the remaining 331 patients for information on amputation in the orthopedic case records. The patients were also offered a re-examination, and 190 patients (239 legs) accepted. Most of those who declined to participate in the interview or re-examination were severely aged, diseased, or had dementia.
The re-examination occurred after a mean 7.2 years (range, 2 to 18 years) at 38 local hospitals. The principal investigator and the same experienced ultrasound technician from the vascular laboratory at Uppsala University Hospital examined 159 patients by visiting the local hospitals, and another 31 patients were examined by local vascular surgeons and ultrasound technicians. Two patients were re-examined with computed tomography (CT). The bypass was examined for flow, aneurysm formation, and stenosis, and the operated on PAA was evaluated for size and flow in the aneurysm sac.
Statistical analysis
Differences in proportions were evaluated using the χ2 or Fisher exact test (when frequencies were <5) for nominal variables and the Kendall τ-b test for ordinal variables. The Student t test was used to evaluate differences in continuous variables. Kaplan-Meier plots of cumulative limb salvage were calculated, with censoring for death and end of follow-up. Comparisons of limb salvage were made by log-rank test (χ2) and Cox regression models adjusted for gender, age, surgical approach, type of graft used, distal anastomotic technique, and whether the operation was elective or acute.
Ethical approval
The study was approved by all nine Regional Ethics Committees in Sweden. According to the administrative rules of the Swedvasc, each patient gives informed consent before registration. Patients who die during treatment are registered without consent, in accordance with Swedish law.
Results
Validity of data
Swedvasc was compared with the SHDR for 1998, 2000, and 2001. More operations for PAA were registered in the Swedvasc (38%, 15%, and 30%, respectively). Among the 146 patients with bilateral disease, who according to patient files had been operated on bilaterally, 141 (97%) of the contralateral operations had been reported to the registry.
Surgical technique
Among the 717 legs, 38 (5.3%) legs belonged to female patients. The surgical approach most commonly used was the medial (medial approach group, MAG, 87%), followed by the posterior (PAG) in 8.4%, endovascular (EAG) in 3.6%, and other miscellaneous approaches in 1.4% (Table I). The miscellaneous group consisted of exploration, thromboembolectomy, and a single extra-anatomic bypass to the anterior tibial artery through a lateral approach and will not be discussed further because this group was small (10 legs) and heterogeneous. The treatment groups had similar follow-up times. Preoperative thrombolysis transformed definite treatment of 59 legs with acute ischemia from an acute to an elective situation. There were no differences in maximum diameter of the PAA among the three groups.
Table I. Surgical and anatomic characteristics of 707⁎ legs operated on for popliteal artery aneurysm
| Characteristic | Surgical approach | ||
|---|---|---|---|
| Posterior | Medial | Endovascular | |
| Number of legs | 60 | 621 | 26 |
| Acute surgery | 12 | 181 | 4 |
| Elective surgery | 48 | 440 | 22 |
| Aneurysm diameter, median mm (range) | 30 (10-150) | 30 (10-100) | 30 (15-80) |
| Run-off | |||
 No open artery | 1 | 49 | 1 |
| Isolated crural artery | 4 | 99 | 2 |
| Popliteal artery with continuity to crural vessel(s) | 41 | 417 | 20 |
| Information missing | 14 | 56 | 3 |
| Graft position | |||
| Anatomic | 58 | 218 | |
| Extra-anatomic | 0 | 381 | |
| No graft used | 2 | 8 | |
| Information Missing | 0 | 14 | |
| Graft type | |||
| Interposition/inlay graft, n (%) | 39 (65) | 124 (21) | — |
| Bypass | 21 (35) | 476 (79) | — |
⁎Ten legs were treated with other miscellaneous techniques, not included in this presentation. |
Valid data on the precise proximal and distal extension of the aneurysm was available in 656 legs (92%). The proximal extension of the aneurysmal arterial segment was the popliteal artery in 545 (83%) and the superficial femoral artery in 111. The distal extension was the popliteal artery in 635 (96.8%), the tibiofibular trunk in 17 (2.6%) and a crural artery in four (0.6%). Inflow and outflow anatomy of the open surgical reconstruction, depending on approach and graft-type, are given in Table II.
Table II. Type of graft and inflow and outflow in 673⁎ legs operated on for popliteal artery aneurysm with posterior or medial approach
| Posterior approach | Medial approach | ||||
|---|---|---|---|---|---|
| Vein reversed, n (%) | Prosthetic graft, n (%) | Vein in situ, n (%) | Vein reversed, n (%) | Prosthetic graft, n (%) | |
| Total | 38 | 22 | 184 | 314 | 115 |
| Inflow | |||||
| Common FA | — | — | 87(47) | 12(4) | 19(17) |
| Superficial FA proximal | — | — | 49(27) | 16(5) | 12(10) |
| Superficial FA distal | 5(13) | 3(14) | 32(17) | 168(53) | 39(34) |
| PA artery above knee | 33(87) | 19(87) | 16(9) | 118(38) | 45(39) |
| Outflow | |||||
| PA below knee | 36(95) | 20(91) | 113(61) | 227(88) | 96(83) |
| Tibiofibular trunk | 1(2.5) | — | 23(13) | 26(8) | 11(10) |
| Crural vessel | 1(2.5) | 2(9) | 48(26) | 11(4) | 8(7) |
⁎This information was missing in 8 operated on legs. |
Although the proximal and distal extensions of the PAA were within the popliteal fossa in 76%, making it possible to operate on the leg with a short bypass from the popliteal artery above knee (Pop-AK) to the popliteal artery below knee (Pop-BK) or the tibial-fibular trunk, this was only done in 62% of the legs. A subgroup analysis was performed of those legs actually operated on with a short by-pass in which it was possible to operate on with either a posterior or a medial approach (Table III). There were no significant differences between these two groups concerning age, sex, or distribution between acute and elective treatment. Among the 195 legs that were operated on with a long bypass with inflow from the common femoral or the proximal superficial femoral artery, 34 legs (17.4%) underwent an associated reconstruction of a femoral artery aneurysm.
Table III. Subgroup analysis of 447 limbs operated on with a short bypass, comparing posterior and medial approach
| Graft | PAG (n = 57) % | MAG (n = 390) % | P |
|---|---|---|---|
| Vein | 67.3 | 80.0 | |
| Prosthetic | 32.7 | 20.0 | .037⁎ |
| Run-off | |||
| No vessel | 0 | 7.7 | |
| One crural vessel | 4.7 | 8.8 | |
| Popliteal + 1-3 crural vessel | 95.3 | 85.5 | .002† |
| Exclusion of aneurysm | |||
| Proximally & distally | 100 | 88.4 | |
| No or only distally | 0 | 11.6 | .003⁎ |
| Patency 30 days | |||
| Open | 93.0 | 93.9 | |
| Occluded | 7.0 | 6.1 | .768⁎ |
| Patency at 1 year | |||
| Open | 87.0 | 90.3 | |
| Occluded | 13.0 | 9.7 | .467⁎ |
| Patency at re-examination | |||
| Open | 81.5 | 83.0 | |
| Occluded‡ | 18.5 | 17.0 | .787⁎ |
⁎Fisher exact test. |
†Kendall τ-b test. |
‡Re-examination after median 7.0 years (range, 2 to 18 years). |
In the PAG, 37 aneurysms (62%) were operated on with an interposition graft with endoaneurysmorrhaphy, 21 (35%) with a bypass, and two (3.3%) with resection and direct anastomosis. The posterior approach was only practiced in 22 centers during the study period. We did not find any differences in outcome of amputation ≤1 year (P = .174), amputation >1 year (P = .140), or frequency of expansion (P = .859) between centers operating on the patients with both methods (posterior or medial) compared with those who used the medial approach only.
In the MAG, shorter bypasses were reconstructed with a reversed vein graft more frequently (70%), and longer bypasses more often were reconstructed with an in situ technique (70%).
We had information on local wound complications at 30 days after surgery on 60 legs in the PAG and on 615 in the MAG. The groups did not differ in frequency of wound infection, deep (graft) infection, seroma, wound rupture, or local neurologic symptoms. All three of the wound ruptures occurred in the MAG.
The frequency of claudication reported at re-examination was similar in the MAG and PAG (37.0% vs 37.9%, P = .933), as well as among those operated on with a vein or prosthetic graft (38.7% vs 33.3%, P = .501).
Graft patency and risk of expansion of the popliteal artery aneurysm
The study design was prospective until 1 year after surgery, with the consequence that valid data on patency >1 year was available only at the time of re-examination. Graft patency at 30 days, 1 year, and at the time of re-examination of the 681 legs in the PAG and MAG according to surgical approach and graft type are given in Table IV.
Table IV. Primary patency, at different time-points, depending on surgical approach and graft type after open surgery of 681 legs with popliteal artery aneurysm
| Vein graft | Prosthetic graft | P⁎ | |
|---|---|---|---|
| 30 days | |||
| PAG | 92% | 86% | .664 |
| MAG | 94% | 87% | .016 |
| P† | .498 | 1.0 | |
| One year | |||
| PAG | 85% | 81% | .719 |
| MAG | 90% | 72% | <.001 |
| P† | .374 | .584 | |
| At re-examination | |||
| PAG | 90% | 57% | .101 |
| MAG | 84% | 69% | .098 |
| P† | .743 | .661 |
⁎Refers to Fischer exact test, comparing graft type within the posterior or medial approach groups. |
†Refers to Fischer exact test, comparing the posterior or medial approach groups, within the subgroups with same graft type. |
Five legs in the EAG were occluded ≤30 days, and six occluded ≤1 year. Among them, two underwent thrombolysis but reoccluded, and the patients presented with claudication. Two were converted to a bypass, which reoccluded, resulting in claudication. One leg was amputated ≤15 days. Eleven legs in the EAG were re-examined, 10 of which were patent, and expansion (type II endoleak) was verified in one leg.
It was possible to compare the aneurysm size at the primary operation with that at the re-examination in 210 legs and to evaluate if there was a flow signal in the treated aneurysm in all 239 re-examined legs. The risk of expansion after operation was 33% in the MAG (57/174) and 8.3% in the PAG (2/24; P = .014). Among the 57 legs in the MAG in which the PAA had expanded, 20 had expanded 1 to 5 mm, 16 expanded 5 to 9 mm, and 21 expanded ≥10 mm. Among those with expansion ≥10 mm, the mean diameter was 45 mm (range, 30 to 95 mm), and the mean expansion rate was 17 mm. The two expanded PAAs in the PAG had expanded from 7 mm to 19 and 39 mm, respectively.
At re-examination, 50 (88%) of the 57 legs with expanding PAA in the MAG were symptomatic. The most common symptoms were local pain, swelling, and neurologic impairment, and they were more common the greater the expansion rate (P = .023). When analyzing all re-examined legs, the 60 with expanding PAA more often had claudication (50% vs 33%, P = .023) as well as local symptoms as described (70% vs 53%, P = .022), compared with the 179 legs without PAA expansion.
Eight legs, all in MAG, underwent reoperation for local symptoms caused by PAA expansion after a median follow-up of 12.5 months (range, 5 to 79 months). These eight grafts were all patent before as well as after the reoperation, and no leg was amputated. In the MAG subgroup with short bypasses, 11.6% were operated on without ligation or with only distal ligation, and 88.4% were ligated both proximally and distally. No difference was found in the risk of future expansion between PAG and MAG with short bypass (P = .321).
Analyzing the bypasses with outflow from the common femoral artery or the proximal superficial femoral artery (195 legs), as many as 42% were operated on without or with only distal ligation. Thus, the surgeons decided to refrain from proximal ligation more often when performing a long bypass. Among these legs, 51 were re-examined, and frequency of expansion was 30.4% among those ligated and 40.0% among those not ligated (P = .544).
Amputation rate
At re-examination or at the end of follow-up of other reasons, 80 (11%) of 717 legs had been amputated above or below the knee. Of these amputations, 79% (63/80) were performed ≤1 year, for an amputation rate ≤1 year of 8.8%. The amputation rate ≤1 year was higher in the acutely treated group than in the electively treated group (21% vs 4%, P < .001). Overall limb-survival is shown in Fig 2, A, and by approach and graft in Fig 2, B. In the MAG, vein grafts did significantly better than prosthetic grafts (log-rank χ2, 14.6; P < .001). This difference in amputation rate remained when legs amputated ≤30 days were excluded (log-rank, 15.4; P < .001), but no difference was observed when all legs amputated ≤1 year were excluded (log-rank, 0.992; P = .319). No difference in amputation rate was found between vein and prosthetic graft in the PAG (P = .623). Among the limbs reconstructed with vein grafts, there was no difference between in situ and reversed vein. The EAG was too small to permit a meaningful comparison, but results were similar to those of the PAG.
Fig 2.
A, Kaplan-Meier plot of limb salvage during 15 years in 717 operated on legs with popliteal artery aneurysm. B, Kaplan-Meier plot of limb salvage. MAGv, Medial approach group (vein graft); MAGp, medial approach group (prosthetic graft); PAGv, posterior approach group (vein graft); PAGp, posterior approach group (prosthetic group).
A Cox regression model revealed a significant and independent association between long-term amputation rate and age (odds ratio [OR] 1.06/year; P < .001), emergency procedure compared to elective procedure (OR, 2.67; P < .001), and prosthetic graft compared with vein graft (OR, 2.02; P = .008), whereas gender, surgical approach, and anastomotic technique were not associated with long-term amputation rate. The observed association between amputation rate and emergency procedure was lost when legs amputated ≤30 days were excluded, but age (OR, 1.05/year; P = .008) and prosthetic graft (OR, 3.09; P = .001) retained the association. When legs amputated ≤1 year were excluded, the associations were lost between amputation and age (OR, 1.04; P = .098), as well as between amputation and prosthetic graft (OR, 1.89; P = .223).
The median time from operation to amputation of the 17 legs amputated >1 year was 3.1 years (range, 1.1 to 9.8 years). The frequency of late amputation was 3.7% in the PAG (2/54) and 2.6% in the MAG (15/571). Among those legs that underwent late amputation, five were treated with preoperative thrombolysis because of acute critical ischemia, and the other 12 had undergone elective operations.
Emergency-treated popliteal artery aneurysm
Among 235 PAAs treated for acute leg ischemia, 100 were treated with preoperative thrombolysis, which transformed definite treatment from an acute to an elective situation in 59 legs (delayed surgery group [DSG]). Within the MAG, the use of prosthetic graft was lower in the DSG compared with the immediate surgical group (ISG), 6.3% vs 25% (P< .001), and within the PAG, the use of vein was higher in DSG compared with ISG, 91% vs 50% (P< .001). The use of a Fogarty catheter was higher in the ISG than in the DSG (P < .001). A subgroup analysis of emergency vs elective operated legs is shown in Table V.
Table V. Subgroup analysis of acute versus elective operations
| Operation type (%) | P | ||
|---|---|---|---|
| Acute (n = 205)⁎ | Elective (n = 512)⁎ | ||
| Amputation ≤1 year | |||
| Yes | 21 | 4 | |
| No | 79 | 96 | <.001† |
| Run-off | |||
| No vessel | 20 | 4 | |
| One crural vessel | 21 | 15 | |
| Pop BK + 1-3 crural vessel(s) | 59 | 81 | <.001‡ |
| Graft | |||
| Vein | 75 | 80 | |
| Prosthetic | 25 | 21 | .214† |
| Expansion | |||
| Yes | 25 | 31 | |
| No | 75 | 69 | .565† |
| Surgical approach§ | |||
| PAG | 6 | 10 | |
| MAG | 94 | 90 | .176† |
⁎Acute operations include limbs operated on for rupture (n = 24), local symptoms (n = 5) or acute ischemia (n = 176). Elective operations include limbs operated on >36 hours after arrival to hospital (n = 453) and legs converted to elective operation after preoperative thrombolysis (n = 59). |
†Fisher’s exact test. |
‡Kendall’s τ-b test. |
§Endovascular and miscellaneous techniques excluded. |
Discussion
This study is based on a prospective reporting from vascular institutions to the national registry and a retrospective review of the patient files. Complete records were retrieved in 94%. A crucial issue is the validity of the registry is if a bias is introduced by not reporting patients with unfavorable outcome. This seems not to be a problem: 15% to 38% more patients were reported to the Swedvasc than to the SHDR, which is used for reimbursement. Patients often underwent bilateral operations, which offered an opportunity to test the internal validity. In all, 141 (97%) of the 146 contralateral procedures had been reported to the registry. We concluded that missing patients or operated on legs were uncommon.
Surgical decision-making in a patient with a PAA is complex, and several factors have to be considered, including the size and configuration of the aneurysm, the run-off, the vessels available for inflow and outflow, and whether a vein is available for a bypass. When the surgeon decides on which technique to prefer in the individual patient, the risk of continued expansion of the PAA should also be considered.
Although this is one of the largest experiences reported on endovascular repair of PAA, only 26 legs (3.6%) were operated on with this technique, precluding a scientific analysis. The first endovascular approach for PAA was registered in 1994, at the same time when the first case report was published,7 and numbers increased throughout the study period. Four hospitals performed most (n = 21) of the procedures, and this represents a pilot experience of a new technique. One fourth of the legs were either converted to a bypass or were amputated ≤1 year. We conclude that in this preliminary experience, the endovascular technique does not seem to offer any advantage to the patients. The rest of this discussion will focus on technical issues of open surgery.
When all patients in the PAG and MAG were compared, it appeared that the advantage of vein over prosthetic graft was relevant only in the MAG (Fig 2, B). The subgroup analysis excluding the legs operated on with long bypasses did identify a confounder, however, because these differences disappeared when short bypasses in the MAG were compared with the PAG (Table III). We believe the absence of a significant advantage of vein in the PAG is merely a type II statistical error, because the PAG is smaller and the patency of the prosthetic grafts on re-examination in the PAG are even inferior to those in the MAG (Table III).
We found no difference in diameter, but as expected, the extension of the PAA was greater in the MAG than in the PAG. A great proportion could have been operated on with a posterior approach, because as many as 62% of the legs operated on were reconstructed with a short bypass from the Pop-AK to Pop-BK or tibial-fibular trunk. There was a trend towards better run-off in the PAG (Table III); conversely, prosthetic grafts were used more often in this group, which perhaps explains why no differences in patency or amputation rates were observed between the two groups in this subgroup analysis.
We have no valid data on whether the great or small saphenous vein was used for the bypass in the PAG. This detail was not always described in the case records and was not yet included as a variable in the registry. A possible explanation for the lack of superiority in the PAG for graft patency may be that an insufficiently sized small saphenous or a great saphenous vein was used. In these situations, a prosthetic graft with an adequate diameter might be preferable.
Most amputations (79%) occurred ≤1 year in this study, an expected finding and similar to previously published studies.11 The low risk of late amputation is clearly illustrated in Fig 2, A. Late amputations did occur, however. The amputation rate increased from 8.8% at 1 year to 11% after a median follow-up of 7.0 years (range, 2 to 18 years). Independent risk factors for early amputation were age, emergency procedure, and prosthetic graft and were discussed in a previous report.15 The present study did not identify an independent risk factor for late amputation. However, despite the size of the present study and the long follow-up period, the number of late amputations was limited (n = 17), making it prone to type II statistical error.
The primary aim of the management of PAA is to prevent thromboembolism and amputation.1 The secondary objective is to prevent aneurysm expansion.2 At re-examination after a median of 7 years, 33% of primary operated PAAs in MAG had expanded, of which 14% underwent reoperation for symptoms related to expansion. Even those with an expanding PAA who did not undergo reoperation presented with claudication and local symptoms more often than those without expansion, indicating that prevention of future expansion is an important aim in the surgical therapy for this disease. Only 8% in the PAG had expanded at re-examination, and no one was reoperated on for expansion. In the MAG, expansion of primary operated on PAAs occurred even when the artery was ligated both proximally and distally, but through a medial approach it is impossible to control side branches. However, among the eight legs reoperated on for expansion, four had not been ligated proximally and information on ligation technique was missing in the other four, indicating that expansion requiring redo surgery can be prevented by proximal ligation in the MAG.
In the largest previously published study, Jones et al17 re-examined 36 legs operated on for PAA through a medial approach with duplex imaging after a medium follow-up of 46 months. Although the follow-up was somewhat shorter than the present study, the reported expansion rate was identical to that of the MAG of this study, 33%. Other studies evaluating the medial approach for expansion found similar expansion rates after long-term follow-up.18, 19 In contrast, Box et al20 reported exclusion of PAA through a medial approach, as close to the PAA as possible in 17 legs, and found expansion in only one (6%) after a median follow-up of 18 months (range, 6 to 48 months). They concluded that the meticulous ligation technique explained their favorable results, but this may also be the result of chance, given the small sample size.
A subgroup analysis of the short bypasses showed significantly more expansion in the MAG than the PAG. We conclude that this complication, similar to endoleak type II after endovascular repair of AAA, is frequently seen after surgery on PAAs, and that in this aspect a posterior approach offers an obvious advantage.
There are differences in local surgical tradition: only 22 of the Swedish centers (52%) operated on patients with PAA with both a posterior and a medial approach. Most patients in this study were operated on with a medial approach, similar to previous reports.21 There are multiple explanations for this predominance of the medial approach, including local tradition, and the PAA can extend into the femoral segment or be combined with atherosclerotic lesions in the femoral artery.
Operating on the patient in a prone position complicates the harvesting of the great saphenous vein and the exposure of the femoral artery for inflow,5 as well as exposure of the crural vessels, when necessary.11, 22 Advantages of the posterior approach, however, are that the aneurysm can be completely excluded and the short interposition graft with an end-to-end anastomosis can be performed.2, 22, 23, 24, 25 It is possible to find a vein of a good quality for a short bypass in most patients. Thus, it seems worthwhile to accept the inconvenience of the prone position and to turn the patient during the reconstruction, when necessary, to reap the advantages of operating on the patient with the posterior approach, whenever possible.
Some authors have argued that small PAA resection is not always necessary because the diameter is never large and symptoms of local compression or rupture are rare.5, 11 The results of this study have clearly shown that expansion of PAAs is a clinical problem that needs to be dealt with, confirming the opinion expressed in previous case reports25, 26, 27, 28 and articles focusing on the posterior approach.2, 21, 23
The choice of graft depends on the availability of vein, how the long segment of the popliteal or superficial femoral artery is affected by aneurysmal disease, the necessity of an immediate revascularization in the case of severe acute limb ischemia, and the life expectancy of the patient. In this study, vein was used in 81% in MAG and 63% in PAG. In situ vein was used more frequently in the MAG when a long bypass was constructed, and only reversed vein was used in the PAG. The overall conclusion is that vein grafts do better, especially in the long run (Fig 2, B; Table IV).
Conclusions
Venous arterial conduit and a posterior approach offer long-term advantages, and should be considered the first choice of treatment during surgical decision-making for the patient with popliteal artery aneurysm.
Swedvasc participants
Swedvasc steering committee: Martin Björck, MD, PhD, Ken Eliasson, MD, Claes Forssell, MD, PhD, Ingvar Jansson, MD, PhD, Anders Lundell A., MD, PhD, Lars Karlström, MD, PhD and Jonas Malmström, MD.
Senior advisors: David Bergqvist, MD, PhD, FRCS, Lars Norgren, MD, PhD, and Thomas Troëng, MD, PhD.
The patients were treated at the following hospitals, followed by the name of the responsible surgeon: Borås: Christer Drott, MD, PhD; Danderyd: Karl-Gösta Ljungström, MD, PhD; Eskilstuna: Ingvar Jansson, MD, PhD; Falun: Bo Westman, MD; Gävle: Torbjörn Tuveson, MD; Halmstad: LennartSmith, MD, PhD; Helsingborg: Gunnar Plate, MD, PhD; Huddinge: Björn Wiklund, MD, PhD; Hudiksvall: Gunnar Tydén, MD, Jönköping Erik Wellander, MD, Kalmar Claes Skiöldebrand, MD, PhD; Karolinska Stockholm: Jonas Malmstedt, MD; Karlskrona: Thomas Troëng, MD, PhD; Karlstad: Becke Lundqvist, MD; Kristianstad: Göran Emtersjö, MD; Linköping: Björn Jönsson, MD, PhD; Lund: Else Ribbe, MD, PhD; Malmö: Bengt Lindblad, MD, PhD; Motala: Christian Almström, MD; Norrköping: Monica Svensson, MD; NÄL: Per-Erland Thornell, MD, PhD; Oskarshamn: Kaj Sundqvist, MD, PhD; Sahlgrenska Gothenburg: Lars Karlström, MD, PhD; Skellefteå: Lars-Olof Lundström, MD; Skövde: Anders Lindhagen, MD, PhD; St. Göran: Gunnar Johansson, MD, PhD; Sunderbyn: Helena Andersson; MD; Sundsvall: Anders Henriksson, MD, PhD; Stockholm South Hospital (SÖS): Peter Konrad, MD, PhD; Uddevalla Bengt Frisk, MD, PhD: Uppsala, Martin Björck, MD, PhD; Umeå: Conny Arnerlöv, MD, PhD; Varberg: Peter Pedersen, MD; Visby: Anika Boström, MD, PhD; Värnamo: Arild Stubberöd, MD; Västervik: Åke Aldman, MD; Västerås: Anders Hellberg, MD, PhD; Växjö: Hilding Björkman, MD; Örebro: Björn Stenberg, MD, PhD; Örnsköldsvik: Sture Hägg, MD; Östersund: Thomas Bolin, MD, PhD.
Author contributions
References
- Popliteal artery aneurysms: a comparison of outcomes in elective versus emergent repair. J Vasc Surg. 2004;39:1171–1177
- . The posterior approach for repair of popliteal artery aneurysms. J Vasc Surg. 2006;43:940–944discussion 944-5
- . The arteriosclerotic popliteal aneurysm. Surgery. 1949;26:41–58
- . An analysis and follow-up study of one hundred popliteal aneurysms. Surgery. 1953;33:284–293
- . Exclusion and saphenous vein bypass of popliteal aneurysms. Surg Gynecol Obstet. 1969;128:829–830
- . Endovascular treatment of popliteal artery aneurysms. Eur J Vasc Endovasc Surg. 2003;26:506–511
- Transfemoral endoluminal stented graft repair of a popliteal artery aneurysm. J Vasc Surg. 1994;19:754–757
- Percutaneous endovascular treatment of peripheral aneurysms. J Cardiovasc Surg (Torino). 2000;41:871–883
- . 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
- . Atherosclerotic popliteal aneurysm. Br J Surg. 1997;84:293–299
- Popliteal artery aneurysms: the risk of nonoperative management. Ann Vasc Surg. 1994;8:14–23
- Swedvasc. The Swedish Vascular Registry. Available at: www.swedvasc.se. Accessed Jan 26, 2007.
- Auditing surgical outcome: ten years with the Swedish Vascular Registry—Swedvasc. Eur J Surg. 1998;164(suppl 581):3–32
- . Nationwide study of the outcome of popliteal artery aneurysms treated surgically. Br J Surg. 2007;in press
- . Popliteal artery aneurysm with acute ischemia in 229 patients (Outcome after thrombolytic and surgical therapy). Eur J Vasc Endovasc Surg. 2007;33:690–695Presentation at the XXth Meeting of the European Society for Vacular Surgery Prague, September 2006.
- . Graft patency is not the only clinical predictor of success after exclusion and bypass of popliteal artery aneurysms. J Vasc Surg. 2003;37:392–398
- . Fate of excluded popliteal artery aneurysms. J Vasc Surg. 2003;37:954–959
- Outcome of popliteal artery aneurysms after exclusion and bypass: significance of residual patent branches mimicking type II endoleaks. J Vasc Surg. 2004;40:886–890
- . Outcome following bypass, and proximal and distal ligation of popliteal aneurysms. Br J Surg. 2007;94:179–182
- . Popliteal artery aneurysms: tried, true, and new approaches to therapy. Semin Vasc Surg. 2005;18:224–230
- . Surgery of popliteal artery aneurysms: a 12-year experience. J Vasc Surg. 2003;37:586–593
- Surgical management of popliteal artery aneurysms: which factors affect outcomes?. J Vasc Surg. 2006;43:481–487
- . Popliteal arterial aneurysms (Their natural history and management). Arch Surg. 1981;116:724–728
- . Fate of excluded popliteal artery aneurysms: should we change our current treatment?. Chir Ital. 2005;57:505–507
- Rupture of excluded popliteal artery aneurysm: implications for type II endoleaks—a case report. Vasc Endovasc Surg. 2004;38:575–578
- . Rupture of a previously bypassed popliteal aneurysm: a case report. J Vasc Surg. 1987;5:874–875
- . An unusual complication of bypassed popliteal aneurysms. Arch Surg. 1983;118:111–113
This project was supported financially by the Swedish Research Council (Grant K2007-64X-20406-01-3), the Swedvasc, Erik, Karin and Gösta Selanders Foundation, and the Research Funds of the County of Jönköping and of Uppsala University.
Competition of interest: none.
CME article
PII: S0741-5214(07)00621-0
doi:10.1016/j.jvs.2007.04.018
© 2007 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.