Two big government-funded studies on back surgery for painful herniated disks show no clear-cut reason to choose an operation over other treatment.

The pain and physical function of the patients, who were suffering from a condition called sciatica, improved significantly after two years whether or not they had surgery. However, neither strategy offered complete relief.

The results indicate patients should choose which treatment they get for the ailment, the researchers said.

"In back surgery for this particular condition, there's actually a choice," said lead author Dr. James Weinstein of Dartmouth Medical School. "If you don't want the risk of surgery, you can do watchful waiting" and still get well.

The condition involves disk cartilage bulging between vertebrae in the lower spine and pressing against a nerve. It can cause excruciating burning pain called sciatica, radiating from the lower back into the legs; patients often have difficulty walking.

About 250,000 Americans have disk surgery for sciatica each year, while another quarter-million instead choose physical therapy, painkillers or rest until they feel better. The surgery costs about $6,000, Weinstein said.

The findings, published in Wednesday's Journal of the American Medical Association, are the first from a big government-funded research project on spine surgery. Patients were treated at 13 spine centers in 11 states.

One study involved 472 patients aged 42 on average who were followed for two years after being randomly assigned to surgery or noninvasive treatment, which included education, physical therapy or painkillers. Surgery involved removing part of the bulging disc in a standard operation often done on an outpatient basis.

Patients in both groups had much improved scores on measures of pain, physical function and disability during periodic evaluations; differences between the groups weren't statistically significant.

Ninety-five percent of surgery patients had no complications, but 4 percent required a second surgery within a year.

In the other study, the researchers followed for two years 743 patients who chose surgery or other treatment. It found a clearer advantage to surgery, including quicker relief in the first months. After three months, 82 percent of surgery patients reported major improvement, compared with 48 percent of nonsurgery patients. Those differences shrank over two years, however, and the researchers said the self-reported results should be interpreted cautiously.

In the randomized study, many patients didn't stay in their assigned group: Almost half those assigned to noninvasive treatment ultimately had surgery, and more than one-third of those assigned to surgery ended up choosing less invasive treatment instead.

That patient crossover makes drawing conclusions tricky and may account for surgery showing no superiority over other treatments, Weinstein said.

No one in either study developed a rare but feared disabling condition called cauda equina syndrome, which should ease the minds of patients and surgeons, said Dr. Eugene Carragee of Stanford University Medical Center.

"Sometimes people with mild sciatica have elected to go ahead with the surgery in order to avoid a theoretical neurologic catastrophe," but now patients can avoid surgery with a realistic expectation that they'll feel better in a year or two, said Carragee, who was not involved in the research.

The study shows how tough it is to find scientific evidence that back surgery works better than other treatments.

For one thing, patients willing to be randomly assigned to surgery are probably different than most patients; their pain could be less, for example, making them more inclined to roll the dice and be assigned to treatment other than surgery.

Another problem: Most surgery studies have no placebo group to rule out the benefits that come with patients' faith in surgery.

Using sham surgery as a placebo, in which patients have incisions but no real treatment, raises ethical questions, but has been done in some research — and some patients say they feel better.

"It's critical that we evaluate the real role of surgery in people's lives," said Dr. David Flum of the University of Washington. "Studies like this that don't have a placebo arm make it very difficult to figure out how much of the effect is the operation versus the patients' and the surgeons' hopes for the operation."