BASIL trial provides hard data for treatment of severe limb ischaemia


The long-awaited results from the Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL) trial, published in The Lancet (online November 24th 2005), have been broadly welcomed as much needed hard evidence. Principal Investigator Professor Andrew W Bradbury, Head of Surgery and Professor of Vascular Surgery, University of Birmingham, told Vascular News, “There are issues that need further exploration. Although angioplasty is equivalent to surgery in the short term by intention to treat, in 20% of cases there was primary failure that required re-intervention. From a biased, surgical point of view, you could say that the angioplasty success rate looks good because most patients actually had surgery later. If you look at it on treatment received: surgery is better, even accepting the short-term drawbacks.”

BASIL was a multicentre, randomised, controlled prospective trial, which followed patients for five and a half years. According to the authors, it was the largest and most powerful study of its kind carried out to date. The trial enrolled 452 patients at 27 hospitals across the UK; 228 assigned to bypass surgery and 224 to balloon angioplasty. The aim was to try to determine which out of the two available treatment options given first was the more successful, with a primary outcome of time to amputation of trial leg or death from any cause.

The argument between proponents of each procedure is well documented: surgery has good long-term patency and clinical durability but is expensive and, in the short term, there is a higher risk of morbidity and mortality. In comparison, angioplasty carries a lower risk, costs less, is faster and involves a shorter hospital stay. Furthermore, failed angioplasty does not jeopardise subsequent surgery. On the other hand, bypass surgery needs good-quality veins for grafting, which may not always be available (and are less likely to be available if a bypass operation has already been conducted). The long-term results of bypasses using prosthetic materials are much less satisfactory.

However, these points of view are largely based on unsubstantiated opinion as previous studies into this dichotomy have all had major methodological problems, claim BASIL’s authors. They have been underpowered or conducted in low-risk patients. Thus BASIL aimed to compare the outcomes of a surgery-first strategy with an angioplasty-first strategy in patients with severe limb ischaemia.

Patient choice

It is estimated that around 50-100 people out of every 100,000 suffer from severe limb ischaemia every year. Suitable patients for the trial represent only a small fraction of the total people who suffer from the condition. For instance, at the top six recruiting centres, out of 585 patients presented with severe limb ischaemia, only 456 had the right infra-inguinal form of the disease. Of those, 220 were treated without revascularisation. And of the remaining 236, only 70 were regarded as suitable candidates for the trial. A third of those refused to take part, thus only 48 – or 8% of the original intake – were randomised into BASIL.

For the trial as a whole, the baseline characteristics of patients were similar in each group. Around two-thirds were over 70 years old, three-quarters were current or ex-smokers, while more than 40% had diabetes. In addition, most had a history of clinically significant cardiovascular disease. Despite these conditions, a third of patients were not receiving antiplatelet therapy and only a third were taking statins.


Not all attempted procedures were successful: in 33 out of 228 cases assigned to surgery, the procedure was not attempted; for angioplasty, eight out of 224 did not receive the intervention. Of the 216 who did, 43 (20%) were judged as immediate technical failures; by 12 months, 109 had resulted in clinical failure. Thirty days after the first procedure, there was no significant different in mortality between angioplasty and surgery. However, the latter was associated with a higher rate of early morbidity: 57% vs 41% (difference 15.5%, 95% CI 5.8-24.8) mainly infective, wound and cardiovascular complications. Furthermore, patients assigned to receive surgery spent significantly longer in hospital and needed significantly more care for the first 12 months.

BASIL specifically measured the intention-to-treat outcome. From this perspective, surgery was associated with a lower re-intervention rate than angioplasty (41 [18%] vs 59 [26%]; difference 8%, 95% CI 0.04-15%). If the results are analysed based on the first intervention actually received, there is even bigger difference: 33 (17%) surgery patients required re-intervention against 67 (28%) angioplasty cases.

For those assigned to surgery first, survival to the primary endpoint (amputation-free survival) at one year was 68% and at three years was 57%, while for angioplasty the figures were 71% at one year and 52% at three years, with no significant differences. However, up to six months there was a trend towards a higher rate of all-cause mortality with surgery, while at two years after randomisation, surgery had a significantly reduced hazard in amputation-free survival and all-cause mortality.

From a cost perspective over the first 12 months of follow up, surgery-first was estimated to cost £23,322 (£20,096 hospital stay, £3,225 procedure costs), roughly a third higher than the £17,419 (£15,381 + £2,039) for patients assigned an angioplasty-first strategy.

From this the authors concluded that for patients with a good prospect of living past the next two years, the durability of surgery makes it a better choice. For more severe cases, the lower risks associated with angioplasty in the short term makes it a better bet.

Follow-up research

The report highlights the fact that if a patient is alive with their leg intact at two years after randomisation, they are more likely to remain so if they had been assigned to receive surgery first, rather than angioplasty. Although BASIL was not designed to answer the question of the long-term effectiveness of surgery vs angioplasty, there are inevitably conclusions that can be drawn with regards to which procedure is more suitable in which circumstance.

Dr Trevor Cleveland, Consultant Vascular Radiologist at the Sheffield Vascular Institute, gave Vascular News his opinion on BASIL: “Of all the suitable patients that were presented to the centres, about 30% went into the trial, so these results are not applicable for the majority. However, for those patients [that were treated] it is very useful for all of us to have this hard evidence.”

Cleveland conceded that the trial raised questions about the success rate of angioplasty. However, he added that this was “because it was attempted in complicated scenarios, not just straight-forward lesions, more usually associated with successful angioplasty”. Such a bold move may have raised the failure rate, but as there are lower risks associated with the procedure it is worth a try, he opined. “If we fail with angioplasty, surgery is still there as an option. It’s something we’ve always said but have never before shown.” Cleveland also cautioned about over-interpreting the data: “There are suggestions that, long term, surgery may be more durable, but the data were never set up to show that, it’s an observation… It is noticeable that the outcomes for surgery were surprisingly good – a reflection of the quality of the surgeons in the trial.”

BASIL will run for another two years to get even more follow up data on patient survival. The study authors are also doing a more in-depth economic analysis that they hope will be published in January 2006. “Surgery seems more expensive than angioplasty in the short term, but in The Lancet paper we only discussed costs for the first year. There is a big cost for amputation – there is patient care and the cost of having homes adapted and so forth. If we can prove that surgery decreases the need for amputation then the relative cost will look better,” said Bradbury.

From a radiologist’s point of view, the additional cost in the short term of surgery means that angioplasty has room to manoeuvre, observed Cleveland. “In the present state, the [angioplasty] technology is not ideal for infrainguinal artery work… If we can find a stent or other device that increases the chance of a successful primary outcome and that makes [the procedure] more durable, we have effectively got money to spend, given the lower morbidity of endovascular treatment.”

He added that BASIL highlighted the limitations of current endovascular technology. “Around 20% of the attempted angioplasties failed to get effective channels. In all but 4% of cases the practitioner could get the wire through. Therefore we need to improve on simple balloons.”

Another, less controversial conclusion regarding patient health can be drawn from the trial. The report states: “Clear evidence shows that so-called best medical therapy (consisting of antiplatelet drugs, smoking cessation and lipid-lowering therapy) can retard the development and progression of lower limb arterial disease.”

Bradbury commented that BASIL was a groundbreaking study in many ways; it has generated a lot of new data and there are more results to come. “It was a prospective, randomised, multicentre trial… In the literature, most studies take place in a single centre – often the ‘top centre’. This study gives a better overall picture of the real world.”

Bradbury and Cleveland both agreed it is important that a panel of expert radiologists examine the angiograms taken before the trial procedures. Cleveland said: “Especially for the failed angioplasties, we need to determine why they failed: were these the wrong disease patterns for simple angioplasty?” Bradbury concluded: “The take-home message is that this was a trial of two strategies. We’ve shown that out to two years they are roughly equivalent.”

Vascular News believes the investigators should be commended. This is the first, long-term hard evidence comparing surgery against angioplasty and, while it will take some time for the data to be absorbed and analysed, the results are very welcome indeed.