If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
A Comparison of Outcomes in Patients with Infrapopliteal Disease Randomised to Vein Bypass or Plain Balloon Angioplasty in the Bypass vs. Angioplasty in Severe Ischaemia of the Leg (BASIL) Trial
Corresponding author. University of Birmingham, Department of Vascular Surgery, Netherwood House, Solihull Hospital, Lode Lane, Birmingham B91 2JL, UK.
Affiliations
Department of Vascular Surgery, University of Birmingham, Birmingham, UK
The aim was to compare outcomes in a subgroup of patients with infrapopliteal (IP) disease randomised to infrapopliteal vein bypass (VB) or plain balloon angioplasty (PBA) in the original BASIL trial.
Methods
A comparison of outcomes from patients randomised to VB or PBA undergoing revascularisation for severe limb ischaemia (SLI) because of IP disease with or without femoropopliteal disease. Data were extracted from case report forms from the BASIL trial. The primary outcome was amputation free survival (AFS); secondary outcomes included overall survival (OS), 30 day mortality and morbidity, freedom from arterial re-intervention, immediate technical success, repeat and crossover interventions, length of hospital stay, and quality of revascularisation.
Results
A total of 104 patients were identified in the BASIL study with IP disease, 56 randomised to IP VB, and 48 to IP PBA. Groups were similar at baseline except for more chronic kidney disease and non-steroidal anti-inflammatory drug use in the VB group, and more previous surgical arterial intervention and antihypertensive use in the PBA group. There were no statistically significant differences in AFS or OS; however, clinically important trends were apparent in favour of a VB first strategy. Patients allocated to VB demonstrated significantly quicker relief of rest pain when compared with PBA (p = .005), but no significant differences in improved tissue healing. Median length of index hospital admission was significantly greater in the VB than in the PBA group (18 vs. 10 days, p < .0001) but there was no difference between the two groups in median total hospital stay between randomisation and the primary endpoint (VB 43.5 vs. PBA 42 days).
Conclusions
Further randomised trials, like BASIL-2 and BEST-CLI, are required to determine whether patients with severe limb ischaemia who require IP revascularisation and who are suitable for VB should have bypass or endovascular intervention as their primary revascularisation procedure.
These data reconfirm the need for further publicly funded, unbiased, pragmatic randomised controlled trials, such as BASIL-2 and BEST-CLI, to compare the clinical and cost effectiveness of infra-popliteal vein bypass and best endovascular treatment in patients suitable for both interventions.
Introduction
The UK National Institute of Heath Research (NIHR) Health Technology Assessment (HTA) funded Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL) trial remains the only published randomised controlled trial (RCT) to compare bypass surgery with angioplasty for the treatment of severe limb ischaemia (SLI) comprising rest pain, tissue loss, or both (http://www.nets.nihr.ac.uk/projects/hta/960501).
The BASIL trial randomised 452 patients with SLI, defined as rest pain, tissue loss, or both because of atherosclerotic peripheral arterial disease, to bypass surgery (75% with vein) or plain balloon angioplasty (PBA) (6 patients also had bare metal stents). Around 25% of patients underwent revascularisation for SLI as a result of infrapopliteal (IP) disease with or without femoropopliteal (FP) disease; however, the trial was not powered to demonstrate effects in this subgroup. In patients who survived more than 2 years after randomisation, bypass was associated with a better outcome in terms of amputation free survival (AFS) and overall survival (OS). By contrast, in patients who did not survive for 2 years, there was no significant clinical difference between the two treatment arms in terms of major amputation or mortality.
In August 2012, the UK National Institute for Health and Care Excellence (NICE) published its national Clinical Guidelines (CG 147) on Peripheral Arterial Disease (https://www.nice.org.uk/guidance/cg147).
Although the multidisciplinary expert guideline committee mainly based its recommendations on SLI revascularisation strategies on data from the BASIL trial, it recognised that BASIL was primarily a trial of FP interventions. As such, the committee's view was that there remained considerable uncertainty as to whether SLI patients requiring an IP revascularisation, and who could have distal vein bypass (VB), should receive VB or best endovascular therapy (BET) as their first revascularisation procedure. For this reason, NICE recommended a further RCT in this area and NIHR HTA subsequently funded the BASIL-2 trial, which compares the clinical and cost effectiveness of VB and BET for the treatment of SLI due to IP disease (http://www.nets.nihr.ac.uk/projects/hta/123545).
Although there seems to be broad agreement that this is an important question, recruitment to BASIL-2 has been difficult. This is mainly because, despite a lack of supporting evidence, a significant body of vascular surgeons and interventional radiologists believe that a BET first revascularisation strategy is clearly preferable for most, even all, patients with SLI due to IP disease, and that VB should be reserved for those who cannot have or fail BET.
To generate discussion around the question posed by BASIL-2, clinical outcomes have been compared in a subset of patients randomised to IP VB or PBA within the original BASIL trial.
Methods
During the trial, 452 patients with SLI as a result of infrainguinal disease were randomised between August 1999 and June 2004 from 27 different UK hospitals to a VB first or PBA first strategy. The original BASIL case report forms (CRFs) were interrogated for those patients randomised to IP VB or PBA. Patients in this subgroup had either PBA with or without bare metal stent or VB with the distal target vessel below the popliteal artery (including the tibioperoneal trunk and beyond). Patients with grafts using prosthetic material were excluded (including composite sequential), as this no longer reflects current clinical practice (Fig. 1).
Figure 1Flow diagram of patient selection for analysis.
The primary outcome was AFS; secondary outcomes included OS, 30 day mortality and morbidity, freedom from arterial re-intervention, immediate technical success rates, repeat and crossover interventions, and length of hospital stay. The quality of revascularisation was also assessed in terms of improvement in ankle brachial pressure index (ABPI), number of days to cessation of rest pain, and healing of ischaemic wounds and/or ‘minor’ amputations.
Time to event outcomes were recorded over 5 years of follow-up and are presented using Kaplan–Meier plots. Differences between groups are estimated as hazard ratios (HR) computed from the log-rank statistic. The authors present 95% confidence intervals for each HR to describe the range of treatment effects compatible with the observed data. Differences in other outcomes are assessed using t tests, chi-square tests and Wilcoxon rank sum tests according to data type and distributional shape. All analyses were performed an intention-to-treat analysis with data censored on 1 July 2007. Analyses were undertaken in SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
The 56 patients randomised to IP VB and the 48 to IP PBA were generally similar at baseline except in the VB group twice as many patients had chronic kidney disease (34% VB vs. 15% PBA) and twice as many were current smokers (25% vs. 11%); they also a had higher use of non-steroidal anti-inflammatory drugs (23% vs. 6%). In the PBA group twice as many patients had a history of previous myocardial infarction (14% VB vs. 29% PBA) and more patients had a history of hypertension on current antihypertensive treatment (61% VB vs. 79% PBA). Previous vascular intervention was more common in the PBA group. In this group, five patients had undergone previous suprainguinal PBA, three patients had previous infrainguinal PBA, one patient had previous infrainguinal surgery, one patient had both previous infrainguinal surgery and infrainguinal PBA, two patients had previous suprainguinal surgery, and one patient had undergone previous minor amputation surgery. Conversely in the VB group two patients had undergone previous suprainguinal PBA, one patient had previous infrainguinal PBA, one patient had previous infrainguinal surgery, one patient had previous suprainguinal PBA with stenting and concurrent infrainguinal PBA, and one patient had previous PBA in an unrecorded anatomical segment. Baseline comparisons are presented in Table 1. Procedural details, describing anatomical variation and target vessel lesions are reported in Table 2, Table 3. Only single vessel recanalisation was attempted in the majority of patients undergoing PBA (85%), with 15% of patients undergoing two vessel recanalisation attempts and no patients undergoing three tibial vessel recanalisation.
Table 1Baseline characteristics.
Vein bypass (n = 56)
Plain balloon angioplasty (n = 48)
p
Age, years (mean, SD)
74.0 (11.3)
77.2 (7.3)
.09
Male sex
41 (73%)
32 (67%)
.5
Comorbidities
Chronic kidney disease
19 (34%)
7 (15%)
.007
Stroke/TIA
12 (21%)
8 (17%)
.5
Myocardial Infarction
8 (14%)
14 (29%)
.06
Angina
11 (20%)
10 (21%)
.9
Dialysis
2 (4%)
0 (–)
.2
Diabetes (DM)
24 (43%)
23 (48%)
.6
Insulin dependent DM
10 (42%)
8 (35%)
.6
Smoking status
Current
14 (25%)
5 (11%)
.1
Ex
29 (52%)
27 (56%)
Never
13 (23%)
16 (33%)
Clinical presentation
Tissue loss
46 (82%)
41 (85%)
.7
Medical therapy
Aspirin
38 (68%)
25 (52%)
.1
Clopidogrel
4 (7%)
3 (6%)
.9
Antihypertensive
34 (61%)
38 (79%)
.04
Statin therapy
22 (39%)
12 (25%)
.1
Warfarin
4 (7%)
5 (10%)
.6
Paracetamol
36 (64%)
29 (60%)
.7
Opiates
31 (55%)
28 (58%)
.8
NSAIDs
13 (23%)
3 (6%)
.02
Gabapentin
2 (4%)
1 (2%)
.7
Amitriptyline
5 (9%)
2 (4%)
.3
Previous intervention
Endovascular
5 (9%)
7 (15%)
.4
Surgical
1 (2%)
6 (13%)
.03
p value in bold demonstrates statistical significance (95% confidence)
Primary technical success (during the BASIL trial technical success was documented by the responsible surgeon or interventionalist at the time of procedure, the criteria for this was left to the discretion of the investigator) was higher for VB (86%) than for PBA (73%) (RR for technical failure 0.53, 95% CI 0.24–1.16, p = .11). In four patients randomised to VB, surgery was discontinued due to a lack of an appropriate IP target vessel; another four patients suffered graft failure within 24 h. Three patients had attempted graft thrombo-embolectomy; one graft was salvaged as a result of this, whereas two patients underwent delayed major lower limb amputation. In the PBA group, it proved impossible to cross and/or dilate the IP lesion in 10 (21%) patients; in two patients, there was distal embolisation/vessel rupture that halted the procedure; and one procedure was abandoned due to pain. From these 13 patients, four went on to have VB, four had no further intervention, three had a repeat attempt at PBA, one had a below knee amputation (BKA), and one patient underwent common femoral endarterectomy and profundaplasty.
AFS was 32% lower in the PBA group than VB group (HR 0.68, 95% CI 0.42–1.10, p = .1) (Fig. 2) and OS 40% lower in the PBA group (HR = 0.60, 95% CI 0.36–1.02, p = .06) (Fig. 3). Removing patients with prior arterial intervention from the analysis did not have any effect on outcome in terms of AFS (p = .5) or OS (p = .3). Similarly, when viewed as a whole, cohort patients with IP disease with prior intervention did not demonstrate significantly poorer AFS than patients with de novo lesions (p = .3). Time to arterial re-intervention (Fig. 4) or time to major amputation (Fig. 5) were similar in both arms, but clinically important halving or doublings of events could not be ruled out for either treatment. Mortality 30 days after the index procedure was not significantly different (4 patients, 7% VB group vs. 1 patient, 2%, in the PBA group, p = .2). The most common causes of death were pneumonia and heart related disease (12 patients and 20 patients respectively) (Table 4). Patients in the VB group were more likely to suffer morbidity within 30 days with more wound infections, sepsis and cardiovascular complications than patients in the PBA group (RR 2.86, 95% CI 1.25–6.53, p = .01) (Table 5).
Figure 2Amputation free survival. HR 0.68, 95% CI 0.42–1.10, p = .1.
In the VB group, nine patients had below knee amputation (BKA) and four had an above knee amputation (AKA) compared with six BKA and four AKA in the PBA group. Only one patient, who was in the VB group, had revision of a BKA to a higher level. Distribution of tissue loss was similar between the groups with most patients having gangrene or ulceration restricted to the digits (74% VB, 71% PBA).
The median length of index hospital admission was significantly greater in the VB group (18 vs. 10 days, p < .0001) but there was no difference in total median hospital stay between randomisation and the primary endpoint (43.5 days VB vs. 42 days PBA).
In the PBA and VB groups, 85% and 77% patients had pre- and post-procedure ABPI measurements. The mean pre-intervention ABPI in the VB group was 0.53 (SD 0.38), which improved to 0.93 (SD 0.42) post intervention. In the PBA group mean pre-intervention ABPI was 0.43 (SD 0.27), which improved to 0.85 (SD 0.32). The mean improvement in ABPI was 0.42 (SD 0.36) and 0.37 (SD 0.38), respectively (p = .18).
In terms of the quality of revascularisation, time to healing of tissue loss was 70% higher in favour of VB than PBA, although this was not statistically significant (HR 1.69, 95% CI 0.88–3.26, p = .1) (Fig. 6). Relief of ischaemic rest pain was more than twice as common in the VB group (HR 2.19, 95% CI 1.27–2.78, p = .005) with more patients pain free at 5 years (Fig. 7).
Figure 6Time to healing of ischaemic tissue loss. HR 1.69, 95% CI 0.88–3.26, p = .1.
Eight (17%) people in the PBA group went on have bypass compared with eight people (14%) in the VB group who went on to have PBA. Patients who underwent primary angioplasty that failed and went on to VB had no differences in AFS compared with patients undergoing primary VB. Likewise, no difference in AFS was seen in patients who underwent primary VB and went on to have secondary PBA.
Discussion
Despite being underpowered, this analysis of the subgroup of IP patients receiving revascularisation for SLI provides some thought provoking evidence. Although the clinical endpoint analysis did not reach statistical significance there appears to be a positive, clinically important limb salvage and survival advantage in favour of VB in this patient group despite a higher rate of smoking and previous myocardial infarction. Clinical effects in terms of healing of ischaemic wounds were difficult to compare because of lack of detailed information regarding wound severity, depth, extent, or presence of infection. Owing to the data available, it was not possible to apply a validated, robust wound scoring system to describe differences, if any, in terms of extent of tissue loss between intervention groups. From the information available (anatomical location of tissue loss and time to complete healing) there was a trend towards improved healing rates in the VB group, although this was not statistically significant. Time to cessation of rest pain from primary intervention was significantly less in the VB group.
We accept that such results should be interpreted with caution given the BASIL trial was not powered to demonstrate treatment effects in the IP subgroup. Results from this study are however thought provoking and demonstrate the need for further higher quality evidence in this patient subgroup. It is also pertinent to note that despite higher 30 day morbidity in the VB group this did not correlate with poorer limb salvage or mortality. Increase in morbidity did, however, lead to a longer inpatient stay during the index admission.
In both groups the median length of hospital stay on index admission was high, the authors believe that the BASIL trial accurately recorded the standard of care for SLI patients in the UK at that point in time. They believe that this was largely because most patients with SLI were emergency rather than elective admissions and a high proportion had tissue loss (approximately 70% in both groups), which often leads to complex social and wound management issues.
Interestingly, there was little or no difference in the length of time spent in hospital between the two groups from randomisation to primary endpoint, suggesting that the early advantage of a shorter length of stay in patients undergoing PBA is lost over a longer period of time when examining all-cause admissions
In both groups, treatment success was arguably low (VB 86% vs. PBA 73%). The original trial analysis was performed on an intention-to-treat analysis. At the time of randomisation, patients were deemed suitable for either intervention based on adequate radiographic assessment and clinical equipoise between both revascularisation strategies. However, in some patients randomised to VB on exploration of the target vessel, the distal bypass vessel was occluded or not suitable for surgical anastomosis. This would represent a similar situation to a failure or inability to cross a target lesion in the PBA group. The trial was designed to be pragmatic and to reflect variations in practice and assessment throughout the UK as no specific guidelines or evidence were available at the time of the original BASIL trial, even now with recent European guidelines
practices in imaging assessment and treatment methods vary widely.
The BASIL trial has been criticised by some regarding generalisability, approximately 30% of screened subjects were randomised into the trial. This trend is seen in current clinical trials in patients with SLI as patients present with heterogeneous patterns of anatomical disease, which can make comparisons and equipoise between treatment strategies a barrier to recruitment and inclusion despite a clear lack of evidence to support a first revascularisation treatment strategy.
Up to 50% of patients with SLI will have significant IP disease and this proportion may be increasing due to the rising prevalence of diabetes and chronic kidney disease.
Present data indicate that it is far from clear that patients with SLI who require IP revascularisation and who are suitable for VB should have BET as their primary procedure. The technical and clinical failure rate with IP PBA was high; whether this would be materially lower in current practice leading to improved clinical outcomes is a subject of ongoing research.
In conclusion, current data reconfirm the need for further publicly funded, unbiased, pragmatic RCTs, such as BASIL-23 and BEST-CLI,
Dr Popplewell and colleagues present a subgroup analysis of BASIL1 (Bypass versus Angioplasty in Severe Ischaemia of the Leg), a controversial, underpowered trial. They suggest from their analysis that vein bypass is superior to balloon angioplasty for infra-popliteal disease, despite the data in their study not achieving statistical significance.
To submit a comment for a journal article, please use the space above and note the following:
We will review submitted comments as soon as possible, striving for within two business days.
This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
We require that commenters identify themselves with names and affiliations.
30278-2&id=gr1.jpg){kind=link}
30278-2&id=gr2.jpg){kind=link}
30278-2&id=gr3.jpg){kind=link}
30278-2&id=gr4.jpg){kind=link}
30278-2&id=gr5.jpg){kind=link}
30278-2&id=gr6.jpg){kind=link}
30278-2&id=gr7.jpg){kind=link}