The problem of repeat procedures
Restenosis has been a vexing problem for interventional cardiologists: despite the often crying immediate angiographic procedural success of angioplasty, a substantial fraction of treated patients come back within months after procedure, often unpredictably for repeat procedures.
Although there has been much debate regarding the optimal definition of restenosis, the clinical significance is clear: Even in the stent era, approximately 15% of patients required repeat revascularisation of the initially treated vessel with success within 6 to 9 months. Some patient subsets even required repeat procedures more often, particularly diabetics.
From this standpoint, the availability of drug-eluting stents has represented a major clinical advance: The RAVEL trial1, followed by a host of other randomised clinical trials2 have established that in using angiography and intravascular ultrasound techniques, the amount of luminal loss in the months following implantation of drug eluting stents was substantially reduced compared to that observed after placement of bare metal stents. This difference was associated with a marked reduction in the need for subsequent revascularization and therefore considered as associated with improved clinical outcomes.
However, more recent studies have highlighted that there may be a price to pay for this reduction in restenosis: a higher incidence of stent thrombosis, particularly late stent thrombosis. This is clearly temporally linked with discontinuation of antiplatelet therapy 3 and may warrant more prolonged therapy than with bare metal stents. Stent thrombosis is a catastrophic event, associated with a high mortality (up to 45% case fatality3).
Recently, two studies have highlighted that this phenomenon is not trivial: the BASKET LATE data, presented by Dr Pfisterer at the ACC scientific sessions in march 2006 are follow-up data from a pragmatic randomized clinical trial performed in Basel in which PCI patients were routinely randomized to treatment with a drug eluting or a bare metal stent. In that analysis, after discontinuation of dual antiplatelet therapy at 6 months, the clinical event rate in the subsequent months was worryingly high: 4.9% of the patients with drug eluting stent experienced cardiac death or MI in the subsequent year, compared to 1.3% in those patients with bare metal stents (p=0.01). The recent data from the PREMIER registry4 have highlighted that a large number of patients discontinue thienopyridines prematurely (up to 1 in 7 patients in the first 30 days) and this is associated with increased mortality in the next 11 months (7.5 vs 0.7% p<0.0001).
Finally, it is still largely unknown whether late stent thrombosis is a time limited phenomenon. It is possible that late stent thromboses may continue to accrue over time in the long term with drug eluting stents. If this is true, this will soon become a major clinical problem with already a base of millions of drug eluting stents placed in patients.
The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.
Therefore, it is a legitimate concern to wonder whether, with the gradual replacement of bare metal stents with drug eluting stents, if we have traded a frequent, but rather benign disease (ie in-stent restenosis) for a rare but delayed and life-threatening risk of stent thrombosis. Until we have large scale data with long term follow-up, it may be prudent to restrict the use of drug eluting stents to those patients most likely to benefit from prevention of restenosis (eg diabetic patients or patients with multiple long lesions in small vessels) rather than to use a policy of indiscriminate use of drug eluting stents for PCI. Another implication is that future device (and particularly drug-eluting stent) studies should be powered for hard clinical outcomes (eg death and myocardial infarction) rather than for surrogate measures of restenosis, which may not allow to measure the actual clinical benefit or harm derived by the patient from new devices.