Immobilization is standard procedure for paramedics in many communities across the country.

Immobilization "shouldn't be applied to every single patient who is shot or stabbed because it uses up precious time and doesn't necessarily benefit the patient," said Dr. Elliott R. Haut, lead author of the study published Tuesday in the Journal of Trauma.

Haut, an assistant professor of surgery at the Johns Hopkins University School of Medicine, said the practice was likely developed for victims of blunt trauma, such as car crashes. They are the vast majority of trauma patients and are more likely to have a spinal injury.

Nonetheless, patients with penetrating wounds such as stabbings and shootings generally are put in cervical collars and secured to a board, even if the wounds are not close to the spine.

A trauma doctor at the Maryland Shock Trauma Center and Maryland's director of emergency medical services agree the findings should be reviewed by policymakers. But they stressed that the benefit of immobilization for some patients also needs to be considered.

"It's a provocative study," said Dr. Thomas M. Scalea, physician in chief at the Maryland Shock Trauma Center. "On the other hand," Scalea said, "if you were among the small number of people who had an unstable spine and became a quadriplegic, you wouldn't think much of the data."

But because the patients could die on the scene, where they don't have access to equipment and staff in emergency rooms, Haut said that they still would likely be better off if paramedics rushed them to the hospital without immobilization, and maybe even without some other pre-hospital treatment such as intravenous fluids.

The researchers examined records from more than 45,000 patients from across the country with penetrating traumas included in the National Trauma Data Bank from 2001 to 2004. They determined that 7.2 percent of patients who were not immobilized had died and 14.7 percent of the immobilized patients had died. They found the risk of dying was still two times higher even after variables such as age, race, gender and injury severity were factored in.

Further, the researchers determined that the chance of benefiting from spine immobilization was 1 in just over 1,000.

Haut said some smaller studies have found similar results, so these new findings were not surprising. But policy is not generally changed after one study.

As a trauma surgeon at Hopkins, Haut said most days he sees a patient with a gunshot wound. A five-minute delay in reaching the emergency room can mean life or death for someone shot in the liver, for example, he said. But someone with a massive head wound is likely to die no matter how quickly he reaches the hospital.

It is hard to determine how many people could be saved under new procedures. But Haut said if one or two in his emergency room could be saved a year, and one or two from each of the other trauma hospitals could be saved, "that could add up to a significant number."

He said he planned to approach the Maryland Institute for Emergency Medical Services Systems, the independent state agency that coordinates the emergency medical system, about changing the policy to allow paramedics to use more flexible guidelines and exercise their judgment. They could immobilize only those patients who have more obvious threats to their spines.

"Paramedics are pretty smart," Haut said. "We already rely on them in tons of situations to make judgments."

The immobilization protocol in Maryland has been in effect since 1994, based on what was then current research, according to Dr. Rick Alcorta, state EMS medical director at the agency. He says it does offer some flexibility, but if there is any doubt or if the patient needs a breathing tube, for example, the person is immobilized.

In general, protocols are supposed to minimize time on the scene. So if there is new literature on reducing that time and saving more lives, it will be examined, he said. The agency has a committee of medical professionals that regularly reviews new data, and updates are not uncommon.

"Whenever we look at data, we need to look at it in a comprehensive way and how it impacts all of our patients in the system," Alcorta said. "Where we can improve the process by modifying the protocols, we will."

Reviewing procedures periodically is a good idea, said Scalea at Shock Trauma, where 400 gunshot victims were taken last year from around the state.

Scalea noted that the Hopkins researchers' findings build on other studies, including at least one at University of Maryland Medical Center. But because the new study is large and uses national data, he said, the issue should be given fresh attention.

Scalea said he takes off the collars of his immobilized patients when it's clear the wound is not close to the spine because the collars can get in his way. Perhaps paramedics could be given more leeway, he said.

But Scalea said state officials who review procedures will have to strike a balance. A small number of patients, such as those with high-velocity gunshot wounds, could develop a spinal injury — or their spinal injury could worsen — without immobilization.

"Hopefully, this gets on the radar ... and they make the best judgment they can," he said.

(Baltimore Sun staff writer Justin Fenton contributed to this article.)