The safety of drug-eluting stents has once again been thrown into the spotlight, with the two main manufacturers of the products, Boston Scientific and Johnson & Johnson, trading blows at the Transcatheter Cardiovascular Therapeutics meeting in Washington, USA.

Debate about the safety of coated stents came to a head last month at the World Congress of Cardiology, when two studies suggested that they were linked to an increased risk of death and cardiac events compared with their bare metal counterparts.

One study found evidence to suggest a significant increase in death and heart attacks in patients receiving Johnson & Johnson's Cypher (sirolimus) stent, and a trend towards an increase with Boston Scientific's Taxus (paclitaxel) product, while the other study indicated that drug-eluting stents might increase non-cardiac mortality.

The studies have divided cardiologists, with some arguing that the widespread adoption of drug-eluting stents on the back of restenosis data was premature, while others insist the studies have not proved that the new products are more dangerous than their older counterparts.

Either way, the debate has had a damaging effect on the $5 billion market for drug-eluting stents, J&J and Boston Scientific both reporting shrinking sales for their stents.

In the latest installment, Boston Scientific kicked off the ruckus with a meta-analysis suggesting that both Cypher and Taxus were associated with a higher rate of late stent thrombosis at one year, which was a statistically-significant finding. After one year, the rate of thrombosis was 1.3% after four years and 1.2% for Cypher, compared to a 0.6% rate in the bar metal stent group.

J&J hit back this morning with a study suggesting that there was no difference in clot rates between Cypher and older, uncoated stents. These data, from a study conducted by the Harvard Clinical Research Institute, showed a rate of clotting of 1.6% for J&J's Cypher at four years, compared to 1.7% for older bare metal stents, although it used a new definition of thrombosis that has not yet been assessed by the US Food and Drug Administration.

The analysis of data from four randomised trials demonstrated that there was no excess mortality or myocardial infarction associated with Cypher, said J&J.