Replacement body parts don't always last as long as our original equipment. And that's bad news for an increasingly old and arthritic populace whose knee and hip replacements can give out long before that final game of shuffleboard. "Most hip implants have to be replaced after just 10 or 15 years," notes Urs Wyss, who heads r & d for implant manufacturer Sulzer Orthopedics (Austin, TX). The same goes for knee implants, which take even more of a pounding. In fact, the American Association of Knee and Hip Surgeons tallied a half million replacements of these two joints in the U.S last year, and replacements of earlier replacements represented 15% of the total surgeries.

When artificial joints do fail, worn out plastics often have to take some of the blame. Both hip and knee implants often use ultra-high-molecular-weight polyethylene (UHMWPE) as faux cartilage to buffer metal components that move in relation to one another. Even normal wear-and-tear can cause the polyethylene to throw off particles that settle around the implant site. And should the body's immune system attack this debris, surrounding bone can also be eaten away-a condition known as "osteolysis." Lose enough bone to osteolysis, and the implant components can work their way loose, causing pain at best and a failure of the artificial joint at worst.

Now a new material from Massachusetts General Hospital's Orthopedic Biomechanics and Biomaterials Laboratory (Boston, MA) addresses the painful and costly wear problem. Licensed to Sulzer, which markets the material as Durasul, this patented UHMWPE has gone into more than 5,000 hips over the past two years. And it recently endured 30 million cycles on a high-tech hip simulator with no measurable wear at the sub-micron level. "That's the equivalent of 30 years in the human body," notes Orhun Muratoglu, the Mass General materials scientist who developed the material working with hip surgery authority William Harris. Durasul has done so well in the lab and the field that Sulzer recently won approval from the Food and Drug Administration to use Durasul in knees as well.

Durasul begins life as an off-the-shelf UHMWPE, but a patented treatment adds extra wear resistance. The treatment employs electron beam radiation to create crosslinks-or bonds between the polymer's molecular chains. "With a high enough crosslink density, you can completely eliminate wear," Muratoglu says. In a twist on conventional e-beam crosslinking, Mass General's method starts by warming the polymer to 125C and then adiabatically melting it during the irradiation process.

Muratoglu explains that the warming reduces the amount of energy that has to be put into polymer by the electron beam, while the melting liberates free radicals generated by the irradiation process and trapped within the polymer's crystalline molecular structure. Left entrapped, these free radicals would feed an oxidation sequence that breaks polymer chains and reduces mechanical properties over time.

Mass General also employed melting in two earlier crosslinking efforts. The first of these treatments irradiated molten UHMWPE, while the other included a melt- annealing step following irradiation. All three of the crosslinking approaches yielded high crosslink densities and thus had "excellent wear properties," Muratoglu reports. But each method produces different crystalline structures in the treated polymer, which in turn causes a disparity in physical and mechanical properties. The current method, with the warm polymer and adiabatic melting, did the best job of preserving the properties of untreated UHMWPE.

And property retention counts for a lot given the loads artificial joints have to withstand. According to Muratoglu, the hip has to bear three times the body's weight every step. "That's seven-hundred-and- fifty pounds for a 250-lb person," he says. It's worse for the knee, which has to withstand five times the body's weight. Muratoglu notes that those loads translate to stresses in the plastic components that can exceed 12 MPa in hip implants and 30 MPa in knee implants.

To serve in both knee and hip applications, Durasul will have to stave off two different kinds of wear. In the hip implants, wear typically occurs as the cobalt-chrome ball head at the end of the femur rides within a UHMWPE-lined cup that forms the joint's socket. Muratoglu describes this wear mechanism as "abrasive-adhesive," and it typically results in smaller particles of debris. In knee implants, wear can manifest itself as the small particles too. They also suffer from larger delaminations, or chunks, that can cause mechanical or alignment problems. "Knees are more complex joints," Muratoglu says. "But we expect our material to perform just as well in them as in hips."

Hip Check

To put hip implants through their paces, Mass General's Biomechanics and Biomaterials Laboratory and 15 other labs around the world rely on a simulator from AMTI (Watertown, MA). With three motion axes and software-controlled loading profiles, the machines can simulate walking, stair climbing and other "hip" activities.

For the sake of throughput, the machine consists of 12 stations, each capable of chugging along at up to 2Hz-or two steps per second. Running continuously, the simulator can in a week put an implant through the same number of "steps" that your average person would do in a year. And to better replicate the "wet" working environs of a real hip, the simulator immerses each joint in an enclosed, temperature-controlled fluid bath. Software-programmable servohydraulics and dc motors respectively drive the system's motion and load axes. For control and monitoring purposes, each station has a load cell that simultaneously measures six force and load components.

Andy Vasilakis, AMTI's president, notes that much of design impetus for the simulator came from the doctors and biomechanics staff at Mass General. "They told us the loads and motions they wanted, and we did the mechanical engineering," Vasilakis says. But with so much motion and profiled loading going on, the simulator did call for plenty of mechanical ingenuity on AMTI's part-including hydrostatic bearings to withstand the millions of simulator cycles and a magnetic coupling that allows simultaneous rotational movement and changing axial loads without a more complex alignment mechanism.