It has been approved since 1997 for the treatment of some epilepsy patients, and the drug agency has told the manufacturer that it is now "approvable" for severe depression that is resistant to other treatment.
But in the only rigorously controlled trial so far in depressed patients, the stimulator was no more effective than surgery in which it was implanted but not turned on.
While some patients show significantly improved moods after having the $15,000 device implanted, most do not, the study found. And once the device is implanted, it is hard to remove entirely; surgeons say the wire leads are usually left inside the neck.
Proponents say that many severely depressed patients do not respond to antidepressants or electroshock therapy and that those patients are desperate for any treatment to relieve their suffering.
"These people have no other options, so we need to consider anything that shows potential to help," said Dr. Harold A. Sackeim, chief of biological psychiatry at the New York State Psychiatric Institute, who consults for Cyberonics Inc., the Houston company that makes the stimulator.
But Dr. Michael Thase, a psychiatrist at the University of Pittsburgh who consults for the company, said there was "simply not a good enough basis in evidence" for approval. While the device is promising, Dr. Thase said, "the shaky state of the evidence means we have to be very cautious with this and prepare for the possibility that the hoped-for benefit isn't there."
The drug agency has given mixed signals about the stimulator. In August 2004, it told Cyberonics in a letter that the treatment was not approvable, saying more information was needed. But in February, after the company provided more data, the agency changed that position, informing the company that the stimulator could now be approved. The company's stock price has fluctuated as investors try to anticipate the agency's decision, which the company is hopeful will come by the end of the month.
The Senate Finance Committee recently began looking into the F.D.A.'s potential reversal, but Cyberonics officials say they have been assured by the agency that this will have no bearing on its final decision.
In a conference call with reporters and analysts on Thursday, Robert Cummins, the company's chief executive, said no other treatment had been deemed approvable by the drug agency for stubbornly depressed patients. Clearly, he said, "the status quo for millions of Americans, their families, psychiatrists and payers is neither safe nor effective."
Still, some patient advocates and other experts are now questioning how the device has come so close to approval with such limited evidence for its effectiveness.
"I've never seen anything quite like this," said Dr. Peter Lurie, deputy director of health research at Public Citizen, a nonprofit group that is a frequent critic of the F.D.A. and the drug and medical-device industries. "What we could be setting ourselves up for is an epidemic of implantation of a device with no proven effectiveness."
Experts who were involved in the approval process say they were moved by the desperate prognosis for severe depression and by powerful testimonials.
At a critical meeting of an F.D.A. panel last June, six patients with chronic severe depression said they felt much better that the stimulator had been implanted, as part of an investigational study. At that meeting, the panel voted 5 to 2 to recommend that the device be approved.
One patient, Charles Donovan III of St. Louis, said the stimulator had saved his life. "I went from being a complete mental-health vegetable to someone who had the energy and confidence to do this book," Mr. Donovan said in an interview.
But the panel did not hear from patients who did not benefit from the stimulator, according to the transcripts. One of them, Katherine Coram, 57, of Silver Spring, Md., signed up for the trial after seeing a newspaper advertisement about it.
"Believe me, when you're depressed for long enough, you get to a stage where you're willing to try almost anything," Ms. Coram said in an interview.
In the study, doctors implanted the device in 235 severely depressed people. The stimulator sends timed pulses of electricity to the vagus nerve, which has wide connections throughout the brain.
Half of the patients then had their stimulators turned on. The investigators did not know which of their patients had their stimulators on.
After three months, researchers "unblinded" the study and compared levels of depression in the two groups based on standard measures of disease severity, the F.D.A. documents show. They found that 17 of the 111 patients who had implants turned on and completed the trial showed significant improvement. But 11 of 110 who had no stimulation and completed the trial also felt significantly better. The difference between the two groups was small enough to be attributable to chance.
Alan Totah, vice president of regulatory affairs for Cyberonics, said at the meeting, "The primary endpoint did not reach statistical significance." But Mr. Totah said "the results did show a positive trend in favor" of the stimulator.
Hoarseness was a common complaint. Many patients who have had a stimulator on also said that it put a quiver, rumble or other odd inflection in their voices.
"I certainly knew mine was on," Ms. Coram said. "I could feel it. You get this constricting pain in the back of the throat. I couldn't talk sometimes."
Ms. Coram said that she was slightly more functional at work after the surgery but that it did not last. Later, she said, after she took a doctor's advice and had the stimulator's pulse turned up higher, "my life fell apart."
"I was very anxious and agitated, much more so than before," she went on. "I felt suicidal for a while, worse than I had been in 8 to 10 years."
But several members of the panel that voted for approval said that given the alternatives for people like Mr. Donovan and others who did well, the insignificant difference between the two groups was cast in a different light.
"The feeling was that anything that gives these people hope is potentially worthwhile," the chairwoman, Dr. Kyra Becker, a neurologist at the University of Washington, said in an interview. "But the whole meeting was uncomfortable, and everyone wanted to see another trial done, no question about it."
Dr. Becker said that if she had voted her conscience, solely on the basis of the evidence, she would have voted not to approve.
A member who voted against approval, Dr. Richard Malone, a psychiatrist at Drexel University College of Medicine in Philadelphia, said he was bewildered by the recommendation.
"I walked out of there thinking I was nuts," Dr. Malone said in an interview. "It was stunning, but then I find much of life is stunning."
The F.D.A. usually follows the recommendations of advisory panels.
Another reason some psychiatrists are intrigued by the device for depression is a finding in the evidence that some people with the implant might do better over time. In follow-up data, Cyberonics reported to the drug agency that about 30 percent of those in the study showed significant improvement on one measure of depression after six months or more.
"The effect appears to be sustained, which is very significant in these patients, who almost always relapse," Dr. Sackeim said.
But other experts say it is extremely difficult to interpret this long-term evidence. Many patients in the study were taking psychiatric medications, or had electroconvulsive therapy, both of which can improve mood. These and other factors are difficult to control for, despite the company's efforts to do so, they said.
Cyberonics says that the long-term evidence it has provided to the federal agency satisfies requirements for approval, and that senior agency officials have told the company as much.
The agency has a higher standard of proof for approving new drugs than it does for devices. Devices require a "reasonable assurance" that they are safe and effective and that potential benefit outweighs the risk.
F.D.A. officials said they could not comment on any product that was pending approval. But Dr. Donna-Bea Tillman, director of the agency's office of device evaluation, said "it is not the kiss of death" if a product's effectiveness is not supported by a well-controlled clinical trial.
"We consider safety and effectiveness in relation to the alternatives that patient population has, including whether they have any alternatives at all," Dr. Tillman said.
If a device is approved, the agency specifies precisely which patients should have access to it, she said. Whether to recommend it is then left to doctors' judgment.
And that is a prime concern among critics of the approval process, like Public Citizen and other groups concerned about patient protection. Once a product has been approved, the manufacturer can promote it aggressively, and some doctors may recommend it to any patient, whether severely or moderately depressed.
Some people may not wait for approval. Sue Wanemaker, 51, who lives near Denver, said she had a vagus stimulator implanted last April for her depression. Ms. Wanemaker, who also has epilepsy, said she had noticed little benefit, and added that when turned up high the stimulation made her feel suicidal.
She has since had the device turned down. "I'm going to hang in there," she said in a phone interview, "because why not, I've got it in now, and - there it goes!" With that, her voice trilled for a few moments.