Two months after hip replacement surgery, Scott Ebert's leg started giving way. His joint began slipping. Razor blades of pain pierced his foot. An inexplicably large bruise grew along his calf. An excruciating burning feeling filled his hip. His ears rang. Doctor after doctor didn't know why. Finally, blood tests showed the metal from Ebert's artificial hip was leaching into his bloodstream.

Six months after it was installed, Ebert's new joint had to be replaced. He's left with chronic pain.

Ebert's Stryker Rejuvenate hip prosthesis was supposed to have been the latest and greatest, a new modular design fitted perfectly to his anatomy, with a choice of stems to be planted into his thigh bone and a choice of necks to support a choice of artificial balls and sockets.

Instead, Stryker's Rejuvenate and ABGII modular systems were recalled in July. They are the latest entries in a gallery of recent hip implant product failures, failures that are shining a light on the need for a system that detects quickly when medical devices go wrong.

The failures appear to have metal and friction in common: metal-on-metal balls and sockets, metal-on-metal stems and necks. Friction appears to cause wear that exposes bare metal to blood and nearby tissues. These traveling metal ions can promote inflammation, tumorlike swellings, and muscle and bone degradation, according to physicians and research reports.

An FDA panel last summer found almost no good reason to use metal-on-metal hip implants after reviewing 17,000 reports of problems in about 500,000 patients with metal balls and sockets.

Common procedure

For a generation of baby boomers growing old and arthritic, total hip replacement is an increasingly common need. More than 1 million of the procedures are performed in the United States each year, and the number is projected to quadruple by 2030, propelled by the graying and fattening of America.

Research shows that for patients with severe hip arthritis, joint replacement can improve quality of life and reduce pain. More than 90 percent of new hip recipients still have their artificial joints after 10 years, the industry reports. But consumers and their doctors alike are still unclear about which materials and surgical techniques are best. Lack of data leaves marketing campaigns and paid surgeon-consultants to tout their ideas.

Revisions can devastate

Dianne Pingel, 66, used to exercise at the YMCA and go for long walks with her husband. Now, she said, she is lucky to walk to the mailbox.

She is recovering from five surgeries related to hip implants in succession. Her left hip was replaced in July 2011 with a Stryker Rejuvenate modular stem. Her right hip was replaced 14 weeks later, in October, also with a Rejuvenate device. The right one has since been removed, but the left hip device remains. It needs to come out.

Removing an existing artificial hip and putting in another one is much more extensive than a first hip replacement. It requires a larger incision, and the sawing away of significantly more bone. Surgeons are forced to pull out a device often firmly cemented into what's left of the patient's femur, attempting to minimize damage to what bone is still there. If the problem joint is surrounded by dead muscle and tissue, that must be removed too.

"It's a lot for a person to go through, "Pingel said.

Growing number of lawsuits

Pingel is the first Rejuvenate patient in the nation to sue Stryker, but she's hardly the last. Pingel's suit is one of nine that have been filed in Bergen County, N.J., where Stryker is based. The company, in its court response, denied the allegations. A company spokeswoman had little to add.

"As a matter of company policy, we don't comment on legal matters," said Stryker's Jeanine Guilfoyle.

Since taking Pingel's case, the attorneys have added 300 clients, Pingel's attorney said. Faced with a barrage of liability lawsuits, Stryker isn't saying what percentage of its joints have failed.

Obstacles to joint registry

Today, the FDA asks companies to track their products and report back. But there is a lag time, and it's unclear what the failure rate is, because the data don't include the total number sold.