ATLANTA — For the first time, an osteoporosis drug has reduced deaths and prevented new fractures in elderly patients with broken hips, according to new research.

Some experts called the drop in deaths “striking” but said other drugs could have a similar effect.

In the study, there were 28 percent fewer deaths and 35 percent fewer fractures in the group that got a once-a-year infusion of the bone drug Reclast compared to those who got a dummy treatment.

No other osteoporosis drug study published in at least 15 years has shown such a pronounced reduction in deaths, said Dr. Kenneth Lyles of Duke University Medical Center, the lead author.

The study was released online Monday by The New England Journal of Medicine to coincide with a presentation at a medical conference in Hawaii. It will be published in a later edition of the journal.

The research was funded by Novartis, which makes Reclast, and Lyles has two patent applications for the use of the drug. Under the name Zometa, the drug was approved by the U.S. Food and Drug Administration for cancer patients in 2002. It was approved for post-menopausal osteoporosis last month, under the name Reclast.

More than 300,000 hip fractures occur in the United States each year. Often they trigger a downward spiral — roughly one in five elderly victims die within a year of breaking their hip.

Generally, doctors tell hip-fracture patients to take Fosamax and other bisphosphonates, a class of osteoporosis drugs that stops bone breakdown. But many patients do not take the pills because they cause heartburn and other symptoms. They also are a hassle for elderly patients: You must take it on an empty stomach in the morning, and wait a half hour before eating.

For the study, researchers recruited about 2,000 patients from 23 countries who were not taking oral bisphosphonates. Their average age was 74 and most were women. All had previously broken a hip.

Half of the participants received Reclast, a bisphosphonate given in an infusion.

Over the next two years, 139 of the patients in the placebo group had new broken bones, or about 14 percent. Just 92 of the treated patients had second fractures, or about 9 percent.

More surprising, 141 died in the placebo group, or about 13 percent, compared to 101 in the treatment group, about 10 percent.

The cause of death was never determined for many, including more than half of those who died in the Reclast group.

“There’s no question that we had a death benefit. We can’t tell you why we had a death benefit,” Lyles said.

Some study participants were taking hormones or other drug therapies, but such cases were balanced between the two groups and did not explain the benefit, he said.

The researchers gave vitamin D, which helps strengthen bones, to all study participants to address harmful deficiencies in some patients. But they said the study benefits were over and above the benefit of the added vitamin D.

It’s not likely the Reclast study will be repeated. Scientists say the value of bisphosphonates has now been clearly proven, and it would be unethical to do another study in which the drug is withheld from some participants.

“The reduction in fracture incidence and death was striking and clearly establishes the need for pharmacologic interventions in patients who fracture a hip,” wrote Karim Anton Calis and Frank Pucino, two pharmacy experts, in an editorial accompanying the study.

The study should cause more physicians to prescribe Reclast or other bisphosphonates to hip-fracture patients, said Dr. Roberto Pacifici, an Emory University osteoporosis expert.