And more patients treated with the quicker and more local radiation technique, also called brachytherapy, went on to need a mastectomy as well -- but there was no difference in their chance of dying in the few years after treatment.

"The decision of whether a patient was treated with brachytherapy or whole-breast irradiation was the single most important factor in whether a patient had a mastectomy," said study author Dr. Benjamin Smith, who called that result "surprising."

"It had generally been thought that if it was used carefully in the appropriate patients, the risks of mastectomy or recurrence would be basically the same in patients treated with brachytherapy or whole-breast irradiation," Smith, from the MD Anderson Cancer Center in Houston, Texas, told Reuters Health.

Brachytherapy involves putting a small device -- a balloon or a catheter -- into the breast, where it delivers a dose of radiation to kill any extra cancer around the cavity where a tumor was removed.

The technique is frequently used to treat prostate and cervical cancer, but it's only in the last decade that devices have been available to use in the breast, Smith said.

The advantages over whole-breast irradiation -- the other option for women who want to keep their breasts -- is that less tissue gets hit with radiation. And brachytherapy only takes five days to complete, compared to six or seven weeks for whole-breast irradiation.

"It's really nice for women who work outside the home or women who live in rural areas because they can get it done much more quickly," said Dr. Todd Tuttle, a cancer surgeon from the University of Minnesota in Minneapolis.

"It's become very popular in the United States, without a lot of good data so far," said Tuttle, who wasn't part of the research team.

For the new study, Smith and his colleagues analyzed Medicare insurance claims for close to 93,000 older women with cancer who got breast-conserving surgery followed by radiation in 2003 through 2007. About 7,000 of them were treated with brachytherapy, and the rest with whole-breast irradiation.

Over the next five years, four percent of women who'd had brachytherapy got their breasts removed because of a cancer recurrence or for another reason. That compared to about two percent of those who got whole-breast irradiation and needed a mastectomy.

Patients treated with brachytherapy were also more likely to get an infection over the next year, or to have breast pain or fracture a rib during study follow-up, the researchers reported in the Journal of the American Medical Association.

They calculated that 56 women would have to get brachytherapy instead of whole-breast irradiation for one to need a mastectomy. And between nine and 16 patients would need to get the local treatment for one extra woman to have a complication, such as an infection.

"Many surgeons are starting to think twice about this kind of therapy for a lot of women," Tuttle told Reuters Health.

"Although it's very attractive at first because you are potentially treating a lot less of the breast and you're doing it in a much shorter period of time, the benefits may not be there. In fact you may see more patients having long-term complications," he said.

Still, Smith said outcomes were generally "very good" across all patients. Five years after their initial surgery, 87 to 88 percent of women were still alive, regardless of what type of radiation they'd received.

Dr. Jona Hattangadi, a radiation oncologist from Brigham and Women's Hospital in Boston, agreed the risks seemed to be low with both types of radiation.

She pointed out the limitations of a so-called retrospective study -- including that researchers can't prove the brachytherapy, itself, was to blame for extra complications and mastectomies.

It could be, for example, that the types of women who opt for brachytherapy are more likely to need a mastectomy anyway -- such as if they go for longer periods of time without coming in to see their doctor.

The "gold standard" option would be a study that randomly assigned women to get one method of radiation or the other, researchers said. That type of analysis is being conducted now, but results won't be available for another few years.

"Physicians need to be selective in who they choose to offer (brachytherapy) to," Hattangadi told Reuters Health. "It may not be uniformly something that should be given to all patients."

"Some of the trade-offs in convenience, efficacy and complications between whole-breast irradiation and brachytherapy may be more complex than we initially appreciated," Smith said.

That doesn't mean women shouldn't get brachytherapy, especially those who are okay with a slightly increased chance they'll need their breasts removed.

But because the treatment "can be a little bit risky," patients should make sure they're having it done by experienced doctors who will monitor them carefully, Smith added.

And women should be educated about all of their radiation options before going forward with any treatment, according to Hattangadi, who wasn't involved in the new study.

"It's important for any woman really to discuss with her physician the risks and benefits of either approach," she said.

SOURCE: bit.ly/4HWZ7 Journal of the American Medical Association, online May 1, 2012.

• A Permanent Economic Emergency