One case involved Sgt. Brandon Huff, who was given Factor VII in a combat support hospital in Mosul last year and later suffered a clot-related stroke. Huff, then 23, was injured on a foot patrol April 20, 2005, when a bomb amputated his left leg and peppered his abdomen with shrapnel.

According to his medical records, he had multiple operations and at least one dose of Factor VII while in Iraq, then was evacuated to Germany. When he woke up, doctors realized that he was paralyzed on his left side, and a scan revealed two blood clots in his brain. He survived, endured months of physical therapy, and has largely recovered.

A clot-based stroke in a 23-year-old man is such an oddity, even among patients with severe injuries, that doctors figured Huff's leg injury was the source of the clots -- even though veins in the legs lead to the heart and lungs, not to the brain. They assumed that Huff must have a congenital heart defect that allowed the clots to transfer across the chambers of his heart and into his arteries. Doctors performed an echocardiogram to confirm their suspicion, but the test was negative and showed Huff's heart to be normal and healthy.

Doctors interviewed by The Sun and told the details of Huff's case were at a loss to explain how a young man with none of the common risk factors for a clot-based stroke -- advanced age, high blood pressure, arteriosclerosis -- could suffer the kind of attack that typically afflicts the elderly.

"I would say it's very likely that stroke was caused by the Factor VII," said Bick, a professor of medicine and pathology at the University of Texas Southwestern Medical Center. "That's very unusual."

Civilian hospitals

Military doctors are not the only ones using Factor VII "off-label" to treat patients with normal blood, going beyond the FDA-approved uses for hemophilia. The medical literature is rife with case studies from trauma centers throughout the world describing the drug's success in stopping bleeding when doctors had abandoned hope.

But use of the drug in civilian hospitals is limited by Factor VII's high cost -- a downside the Army can all but ignore. Factor VII is one of the most expensive drugs in the world, and the Army has bought roughly $11 million worth in the last three years.

Enthusiasm among civilian doctors has tempered for reasons beyond price, however, as research builds suggesting that Factor VII might be inducing unwanted clots and that it might not work as well as the anecdotal reports suggest.

While some small scientific studies suggest that bleeding trauma patients given Factor VII need fewer blood transfusions, none has recorded a statistically significant increase in survival among patients who get it. The study that documented the most promising use for Factor VII -- treating bleeding in the brain -- also reported a complication rate as high as 10 percent, compared with 2 percent for patients given a placebo.

When asked by Novo Nordisk to approve Factor VII for treating people without hemophilia, regulators around the world have indicated that they are concerned about Factor VII's potential for complications.

Novo Nordisk, which sells Factor VII under the brand name NovoSeven, sought permission in Europe to market the drug as a treatment for bleeding in the brain but withdrew the request in April after European regulators expressed concerns about "excessive clotting."

A report published early this year in the Journal of the American Medical Association, using data culled from the FDA's database of adverse drug reactions from 1999 to 2005, attributed 43 deaths to thrombo-embolism -- blood clots floating through the patients' veins or arteries -- after injections of Factor VII, mostly in patients without hemophilia.

The report offered no comparison with complication-free uses to put the deaths in perspective and could not prove that the drug caused the clots, and thus it had limited impact. But it was the first published report in a major medical journal to question the safety of Factor VII. And it suggested that the reports of adverse drug reactions filed with the FDA, which are required by drug manufacturers but voluntary for hospitals, "largely underestimate the actual number of occurrences."

Like virtually all published studies of the drug, the FDA report concluded that a large-scale clinical trial, in which 1,500 or more patients are randomly given either Factor VII or a placebo, is necessary to determine whether the drug is safe for non-hemophiliacs and whether it works. Because the FDA requires patient consent in clinical trials, which is often difficult to obtain from trauma patients, Novo Nordisk expects a large trial in the United States to take several years.

Civilian doctors contacted by The Sun, meanwhile, said they have largely abandoned their early hope that Factor VII would be safe and appropriate for widespread use. Most were surprised to learn that the military has embraced it so enthusiastically.

"It's very hard to justify that kind of use in any circumstance, military or non-military," said Dr. David Kuter, director of clinical hematology at Massachusetts General Hospital, who said his hospital mostly stopped using Factor VII for patients after noticing "an increased rate of death."

"There's just a lack of demonstrated efficacy in any situation," he said.

While they rarely see the types of complex injuries that Army doctors see, civilian doctors generally consider intravascular clots to be more dangerous than severe bleeding. Even patients who become "coagulopathic" -- who continue to bleed after their injuries are repaired surgically because their blood has lost the ability to clot -- have options available, including transfusions and other drugs. The effects of blood clots, however, can very quickly become irreversible and deadly.

Military doctors in Iraq often inject Factor VII into injured patients in anticipation of coagulopathic bleeding later, but in hospitals in the United States it is more often used later in treatment, after other options have been tried, and only after case-by-case consideration.