Renal, iliac, and femoral stenting were developed just a few years after Dotter first performed peripheral angioplasty in 1964. However, the gap between coronary angioplasty in 1977 and coronary stenting in 1986 is longer, and probably reflects the technical difficulties in manufacturing small stents as well as concern over implanting a foreign body into the coronary tree. Indeed, the early results of coronary stenting were compromised by subacute thrombosis and by bleeding secondary to the

more »

... essive anticoagulant regimens used.' As a result, stenting first established its role in bailout of acute coronary dissection with impending or actual occlusion, where reducing the risks of infarction and/or the need for urgent coronary bypass surgery more than justified the thrombotic and haemorrhagic risks.2' In the ensuing years stents have established themselves as the major adjunct to ordinary balloon angioplasty. In 1989 it seemed that stenting was just one of a large number of innovative developments,4 but other technologies such as directional coronary atherectomy, excimer laser, and rotational artherectomy remain confined to interested centres and particular indications. The widespread adoption of stenting has followed improved anticoagulant regimens, better design of stents, and the realisation that stenting is a relatively easy technique for the operator, as well as data supporting the use of stents for routine angioplasty. Initial clinical experience with stenting was reported in an increasing number of publications from 1986 onwards; in some cases these were large registries with follow up over several years, but all provided data that were essentially observational. The first large randomised controlled trials were STRESS and Benestent, both showing a reduction in angiographic restenosis by elective stenting of native coronary vessels with reference diameter > 3 0 mm (Benestent: without stent 32%, with stent 22%; STRESS: without stent 42%, with stent 32%). The clinical benefits were improved procedural success, reduced abrupt closure, and reduced adverse cardiac events-principally a reduced need for repeat interventionbut there was still an excess of bleeding and consequently a longer in-hospital stay.56 The registry experiences of Colombo et al' and of Morice et al,8 followed by the randomised controlled study from Schomig et al,9 were instrumental in giving operators the confidence to stop using warfarin and to move to aspirin and ticlopidine. As a result more recent registry experience and interim reports from newer trials such as Benestent II show much less bleeding; they also show even lower rates of restenosis, often below 20%.10 What are the current indications for stenting in patients undergoing coronary angioplasty? At present stents are used for suboptimal results, and of course for bailout, but when the results of ongoing trials (table) become available we may consider stenting for every lesion. At the Brompton Hospital the use of coronary stents has plateaued at around 70% of PTCA procedures for the past two to three years; the 30% "non-stent" angioplasties include PTCA for lesions in small vessels, in very tortuous vessels, and lesions with an excellent angiographic result after ordinary balloon angioplasty. At present there are no data to support stenting lesions with an optimal angiographic result after POBA, in which the risk of restenosis is low, but future trials might show that this group will benefit from stenting. Enthusiasm for elective stenting must be tempered by consideration of the costs. Early health economic analyses of the Benestent and STRESS data were heavily influenced by excess bleeding and longer in-hospital stay, which roughly balanced the reduction in need for further revascularisation after stenting. Health economic analysis based on the Benestent II pilot data, with few vascular complications, shows that stent implantation is more cost-effective than POBA.'0 What now are the frontiers for coronary stenting, and who does or does not need a stent? Just considering the list of issues shows how far we have come in the past few years. Stenting small vessels Data from a number of trials organised at the Thoraxcenter, Rotterdam, the Netherlands has been pooled to show that the risk of subacute thrombosis rises sharply with vessels smaller than 3-0 mm. Stents are not alone in this-the results of POBA and all other devices are significantly worse for small vessels. However, modem stent designs, antithrombotic regimens, and increased operator expertise have improved the results of stenting small vessels, and many operators will now confidently tackle 2x0-2*5 mm vessels, deliberately dilating the stented segment to