The relative increase in risk to comparative medicines was found to range from 27 to 33 percent, however the absolute risk increase for suffering a heart attack was only 0.27 percent, the authors of the meta-analysis said in the study published on Monday in the Archives of Internal Medicine online edition.

Pradaxa, known chemically as dabigatran, is approved in the United States for stroke prevention in patients suffering from an irregular heartbeat known as atrial fibrillation, and in Europe to prevent strokes caused by blood clots in patients who have undergone hip and knee replacements.

The medicine from the privately held German drugmaker is seen as an alternative to decades old warfarin, the highly effective but difficult to use blood thinner.

Pradaxa, which is being touted in a large direct-to-consumer advertising campaign, is expected to compete with two other new blood clot and stroke preventers - Xarelto from Bayer AG and Johnson & Johnson, and Eliquis, or apixaban, from Bristol-Myers Squibb Co and Pfizer Inc.

All three of the new anticoagulants are expected to be widely used and generate billions of dollars in sales.

Doctors from the Cleveland Clinic conducted the meta-analysis based on data from seven clinical trials of Pradaxa involving more than 39,500 patients. The studies compared Pradaxa to either warfarin, enoxaparin - another widely used blood thinner sold under the band name Lovenox - or a placebo.

The meta-analysis also found that the rate of death was lower for patients who received Pradaxa than those who got other medicines in the clinical trials - 4.83 percent versus 5.02 percent in the control groups.

"The overall benefit and risk balance of dabigatran use appears to be favorable in patients with atrial fibrillation because of reduction in ischemic stroke," the study authors wrote. However, they caution, "clinicians should consider the potential of these serious harmful cardiovascular effects with use of dabigatran."

They called for further investigation of the Pradaxa heart risk, especially if used in populations at high risk of heart attack or acute coronary syndrome - an umbrella term for patients who have heart attacks or chest pain typically caused by a blocked coronary artery.

Boehringer said it did not agree with the method used in the meta-analysis that combined trial data from different patient populations and various comparator medicines or placebo.

"Based on the totality of the data, we concluded that myocardial infarction (heart attack) is not an adverse consequence of Pradaxa treatment," said John Smith, Boehringer's senior vice president for clinical development and medical affairs.

"These are data that we've previously reported and we feel there is no adverse signal here," Smith added.

However new medicines often turn up additional safety issues not seen in carefully controlled clinical trials once they become widely used in the general population.

European regulators have said that Pradaxa should be used with caution and at low doses in patients over 75 years of age and in those with moderate kidney impairment due to reports of fatal bleeding. Pradaxa is not recommended for those with severe kidney function problems.

The study authors recommend that doctors "retain a critical view as a powerful drug enters clinical use on a potentially massive scale."

In a related editor's note on the meta-analysis findings, Dr Rita Redberg wrote: "These additional concerns deserve serious consideration in weighing the risks and benefits of dabigatran and highlight the importance of continued collective analysis of data after drug approval."

(Reporting by Bill Berkrot; Editing by Michele Gershberg and Richard Chang and Carol Bishopric)

• A Permanent Economic Emergency