teaser
Off-label use of drug-eluting stents (DESs) is common and associatedwith higher ischaemic complication rates, according to studies justpublished.
Dr Nirat Beohar of Northwestern University’sFeinberg School of Medicine, Chicago, USA, and colleagues examined datafrom the DES cover Registry – a prospective multicentre observationalstudy conducted at 140 US centres that includes 7,752 patients whounderwent percutaneous coronary intervention (PCI) in January–June 2005.
Of5,541 patients receiving DESs, 2,588 received stents for off-label oruntested indications – so “approximately half of all DES use occurs inoff-label or untested settings,” the authors observe.
Off-labelindications included use of stents for restenosis, bypass graft lesionand long lesions. Untested indications included left main, ostial,bifurcation or total-occlusion lesions.
Within 30 days, theprimary composite endpoint was significantly more frequent in patientswith off-label use than those receiving stents for standard use.
“Bothobserved and adjusted analyses indicated a greater-than-twofold higherrisk of death, MI or stent thrombosis,” the authors write.
The risk was not significantly higher with untested use, however (adjusted hazard ratio = 1.45).
Excludingearly events, the endpoint was not increased at one year with eitheroff-label or untested use (HR=1.10 and HR=0.91, respectively).
However,off-label and untested use was associated with a 1.5-fold increase intarget-vessel revascularisation (TVR) via PCI or coronary artery bypasssurgery at one year.
In a related article, Dr Htut Win ofBaylor College of Medicine, Houston, USA, and colleagues report thatoff-label use is more common than on-label use – and causes apersistently higher rate of adverse angiographic and clinical outcomes.
DrWin’s team assessed in-hospital and one-year rates of major adversecardiac events in 3,323 patients enrolled in the Evaluation of DrugEluting Stents and Ischemic Events registry, who had high-riskangiographic and clinical features, and were implanted with DESs.
More than half the patients (n=1,817, 54.7%) had at least one of nine potential off-label characteristics.
Thecomposite endpoint of death, MI, or TVR occurred significantly morefrequently in these off-label than the on-label patients (10.9% vs5.0%, adjusted odds ratio = 2.32, p<0.001). This was drivenprimarily by a higher rate of MI in the off-label patients.
Thecomposite endpoint was still higher with off-label DES use at one-year(17.5% vs 8.9%, adjusted HR = 2.16; p<0.001), although mortality wassimilar between groups by this time.
Meanwhile, stentthrombosis was significantly more frequent in the off-label versuson-label group both in-hospital (0.4% vs 0.0%) and at one year (1.6% vs0.9%, HR=2.29; p=0.05).
The authors conclude: “Cliniciansshould be cautious about extrapolating the benefits of DES comparedwith bare-metal stents that were observed in randomised clinical trialsto higher risk settings that have not been assessed.”
JAMA 2007;297:1992-2000;2001-9
