Three-year data from a study on Abbott Laboratories' (ABT) "Xience" heart stent showed it performed better on certain measures than a Boston Scientific Corp. (BSX) stent, according to details Abbott released Sunday.

Xience is already the top-selling drug-coated stent in the U.S. and Europe behind earlier data showing it outperformed Boston Scientific's "Taxus" devices, so the latest data from the 300-patient Spirit II study - which showed a widened performance gap over the last year - appear unlikely to alter trends in the $4 billion global market for the devices.

But the new data, released at the American College of Cardiology's annual conference, do create a longer-term performance measure for Abbott's device. The data release also marks an opportunity for doctors and Wall Street analysts to see how Xience performed between years two and three, after data from the two-year mark raised some questions about narrowing in the arteries of Xience-treated patients.

Measuring that narrowing at the six-month mark was the Abbott-sponsored study's original main goal.

Johnson & Johnson (JNJ) and Medtronic Inc. (MDT) also make coated stents, which are tiny scaffolds for heart arteries that use medication to fight the growth of scar tissue that can lead to repeat procedures.

Boston Scientific also benefits from good Xience data because it sells a version of the stent, called Promus, and then shares profits with Abbott under a deal connected to an acquisition three years ago.

Patients in the Spirit II trial were randomized so that three people had a Xience stent for every one who had a Taxus. At three years, there was a 6.4% rate of so-called "major adverse cardiac events" among Xience patients and a 14.9% rate among Taxus patients, with the difference considered statistically significant, according to Abbott.

The major events measure is a combination of cardiac death, heart attacks or re-treatment of the same area due to lack of blood supply to the heart. This rate for Xience was the same between years two and three but increased for Taxus.

The Spirit II study, which included patients outside the U.S., mainly used an older version of Boston Scientific's Taxus device, although some patients were treated with a newer version the company sells called Taxus Liberte.

Among individual events, the reduced risk of cardiac death among Xience patients was also considered statistically significant, while the lower percentage rates of heart attacks and re-treatment for Xience patients were not considered statistically significant.

When two-year data were released last year, analysis of a subset of patients who had their arteries imaged with an angiogram showed Abbott's stent lost its early edge on a measure called "late loss," or narrowing in the stented area.

This type of analysis was not performed again at the three-year mark, making the latest trends there unclear. But the rate of events with the devices can offer clues about whether the narrowing led to problems. John Capek, executive vice president of medical devices at Abbott, noted in an interview that clinical events have grown faster in the Taxus arm than in the Xience arm.

The rate of stent-related clots with both Xience and Taxus was not considered significantly different in the new data - the rate was 0.9% for Xience and 2.8% for Taxus. Boston Scientific said it was pleased with the performance of both its Promus and Taxus stents in this study and a separate analysis, also released at ACC, looking at two-year data across two different studies.

-By Jon Kamp, Dow Jones Newswires; 617-654-6728; jon.kamp@dowjones.com