They can improve vision, help regulate the heart and pin together broken bones. They keep clogged arteries open, stabilize blood sugar in people with diabetes, prevent pregnancy and enable the deaf to hear.

Medical devices, from artificial joints and breast implants to pacemakers and insulin pumps, are now--literally--part of life for an estimated 25 million people. Nearly one in every 10 Americans has some type of implant. But their benefits and risks are still poorly understood by consumers, an expert panel concluded last week.

"Consumers are not getting enough information," said Edward N. Brandt Jr., director of the Center for Health Policy at the University of Oklahoma Health Sciences Center and co-chairman of the panel, which was convened by the National Institutes of Health. "Their expectations are built up. They think everybody's troubles are over if they can just get an implant."

Most Americans who receive medical implants experience no problems. But the panel concluded that consumers are not adequately informed of a device's benefits, risks, potential complications and expected longevity. It recommended that patient consent forms be streamlined to include such information prior to surgery. It also called for a national monitoring system that would track implanted devices and retrieve implants that are no longer needed for study.

The information gap about these devices is significant, even for health professionals. Panel member Janet Wittes, president of Statistics Collaborative Inc., a Washington consulting firm, had two intra-ocular devices inserted in her eyes to improve vision. "I have no idea what type or model they are," said Wittes, noting that the information would be helpful in case of a recall or other problem.

Design flaws prompt recalls of medical implants from time to time, and these defective implants are usually studied carefully so that improvements can be made.

But the panel found that no one tracks the vast majority of medical implants that work well. Only by systematically collecting these devices and assessing their normal wear and tear can researchers learn valuable information to help with future innovation and design, the panel noted.

"Science progresses by facing its failures and learning from its successes," wrote the panel. Monitoring "can provide valuable insight into the strengths and weaknesses of implant design."

But there are major obstacles to setting up a uniform monitoring system. To better educate consumers, the panel recommended giving all implant users a card describing the device, model and serial number of their particular implant, including the date it was implanted. The panel urged that the cards also serve as an implant consent form that would allow doctors to retrieve implants for analysis after death.

"Many people understand that medical implants improve quality of life," Brandt said. "Far fewer people recognize the importance of retrieving and analyzing medical implants when they fail or are no longer useful."

Consumers may need to be convinced, however, since retrieval usually requires surgery. "Patients and families must understand the purpose of retrieval" so they will be willing to allow doctors to have the devices and analyze them, the panel concluded.

Family members may balk since collection can delay burial or cremation. There are also religious considerations. A few groups--including Orthodox Jews--oppose post-mortem surgical retrieval.

Paying for retrieval is a major question, too. If an implant is recalled, retrieval is often covered by health insurance, said panel member Kenneth H. DeHaven, director of athletic medicine at the University of Rochester's School of Medicine and Dentistry in New York state. But insurance companies generally don't pay for services needed just to monitor an implant.

Since many medical implant manufacturers are small companies, they don't have the money to pay for retrieval either, the panel found. "There currently exists no industry-wide 'Superfund' to support implant retrieval and analysis," the panel concluded. But even if such a fund existed, it probably couldn't cover the entire cost of retrieving every implanted device, according to the panel.

Some companies have also been reluctant to retrieve implants for legal reasons. Collection of data on used devices could leave them vulnerable to lawsuits. "We need to call attention to the code of silence that has been imposed on the medical profession, on hospitals and on the industry by the threat of litigation," said panel member Arthur R. Kantrowitz, professor of engineering at Dartmouth College in Hanover, N.H.

In Sweden, England and Japan, health officials have established national registries to track implants. Citing concerns about patient confidentiality, the NIH panel stopped short of recommending a similar national system to retrieve and evaluate medical implants.

The NIH panel suggested development of a Web site that would give consumers "rapid access to relevant data" about common medical implants. By clicking on a part of the body, consumers could be linked to a page containing relevant information about implants and their appropriate uses.

"Maintaining and updating the Web site will require considerable effort," the panel noted. "The group responsible for the main site will need to provide critical appraisals of the linked sites so that the user will not be misled by raw data and biased selection. Cooperation between the FDA and NIH on developing and maintaining the Web site would be ideal."

The panel's full report is available online at http://consensus.nih.gov or by calling the toll-free number, 1-888-644-2667.

CAPTION: Surgical implants are increasingly common throughout the body. Among the devices now in use are jaw replacements (1), dental implants (2), permanent contact lenses (3), ear implants (4), heart pacemakers (5), breast implants (6), knee replacements (7), intrauterine devices (8), hip joint replacements (9) and insulin pumps (10).