A Study in Hipness
By Beth W. Orenstein
Radiology Today
Vol. 23 No. 1 P. 10

Researchers discover that MRI can find adverse tissue reactions in asymptomatic patients who have undergone hip resurfacing and arthroplasty.

More than 450,000 total hip arthroplasties are performed each year in the United States, according to the Agency for Healthcare Research and Quality. The most common reason is advanced osteoarthritis. “At the end of the day, we can’t cure osteoarthritis, but we can treat it,” says Hollis G. Potter, MD, chairman of the department of radiology and imaging and the Coleman Chair in MRI Research at the Hospital for Special Surgery (HSS) in New York. “As the population gets older, more people will require hip and knee implants,” Potter says. She adds that joint replacements also are going into younger people, many of whom lead active lifestyles, “which means the demands on the implants are greater.”

The technology and surgical techniques of joint replacement have vastly improved since what is believed to be the first hip replacement, performed in Germany in 1891 with the results presented at the 10th International Medical Conference at Berlin. Marius Nygaard Smith-Peterson, of Massachusetts General Hospital, a Norwegian-born American orthopedic surgeon, is credited with making the first mold arthroplasty in the 1920s. His implant, made of glass, was a hollow ball that fit over the femoral head. Metal-on-metal hip prosthesis was introduced in the early 1950s and was used for decades. According to the American Academy of Orthopaedic Surgeons, since 1966, more than 1 million metal-on-metal prosthetic hips have been implanted in people around the world. It is estimated that about one-half of those were in the United States.

Today, four types of total hip replacement devices are available with different bearing surfaces: metal on polyethylene, ceramic on polyethylene, ceramic on ceramic, and ceramic on metal. Since May 2016, no metal-on-metal total hip replacement devices have been approved by the FDA for use in the United States.

One of the reasons metal-on-metal total hip replacements are no longer used is that some people have adverse reactions to them. When the surfaces rub against each other, tiny metal particles can wear off and enter adjacent soft tissue. There also may be wear and corrosion at the connection between the metal ball and the taper of the stem. Additionally, some of the metal ions (cobalt and chromium) from the implant can seep into the bloodstream, inciting an inflammatory host-mediated reaction.

The newer materials used in hip replacements are considered safer. However, some patients still have adverse reactions, Potter says.

Mathias Bostrom, MD, chief of the adult reconstruction and joint replacement service at HSS, says the issue is as follows: “Where you have dissimilar materials touching each other, you could have problems.”

Visualizing Soft Tissue
Over the past decade, registries in Australia and the United Kingdom showed some patients who had received metal-on-metal total hip implants had adverse reactions to them. The adverse reactions to metal-on-metal implants caught Potter’s attention. “We were very interested in this, and I’ve been working on it for many years now,” Potter says. Her interest, she says, comes from the fact that most of the reactions are soft tissue related, “and MRI is the best tool we have available for looking at soft tissue reactions.” Potter’s interest has led to what some are calling groundbreaking research to detect adverse tissue reactions in people who have undergone hip replacements or hip resurfacing.

Hip arthroplasty patients who have reactions may develop hip/groin pain, weakness, swelling, numbness, or changes in their ability to walk. Patients also may experience systemic symptoms involving the skin, kidney, heart, thyroid, and nervous system. However, not every patient with an adverse reaction to metal debris will have pain or other obvious symptoms, making an adverse tissue reaction difficult to diagnose.

Patients can undergo a cobalt-chromium blood test to look for elevated blood ion levels, but the results do not always correlate with adverse tissue reactions. Not all patients with adverse reactions have elevated metal blood levels. “All the blood test tells you is that you have circulating metal ions in your blood,” Potter says. “It doesn’t tell you whether you’re actually having a reaction to those metal ions.”

Matthew F. Koff, PhD, an associate scientist at HSS in the department of radiology and imaging MR division, says patients’ genetic makeup typically determines whether they will have an adverse tissue reaction, “but, unfortunately, there’s no way to predict whether someone is genetically predisposed.” Researchers have tried to test patients before they receive an implant—using skin testing and metal sensitivity testing—but neither seems to indicate who will have an adverse tissue reaction to the metal and who will not, Koff says. Even patients who have skin reactions to metal earrings or jewelry are not necessarily predisposed to adverse tissue reactions, the researchers say.

Focus on Early Detection
The earlier the adverse reaction is detected, the better, because early detection makes it easier to treat, Potter says. Reactions also typically worsen over time, and delaying detection means that patients are in unnecessary pain longer, face more complicated revision operations, and have more challenging recoveries, she says.

Potter and her colleagues have been working on software that makes MRI the go-to tool to detect adverse tissue reactions in total hip arthroplasty and hip resurfacing, which is an alternative to completely replacing the hip. During a hip resurfacing, the thighbone head is trimmed and capped with a metal ball that moves in a metal socket. Hip resurfacing operations use a metal-on-metal bearing hip. Other materials, including ceramic and high-density polyethylene, can be used as well.

In December 2021, Potter, Koff, and colleagues published a study in Clinical Orthopaedics and Related Research, in which they found that MRI identifies adverse tissue reactions in asymptomatic individuals after hip resurfacing arthroplasty. Previous studies have been limited to metal-on-metal implants.

The researchers contacted 2% of the 22,723 patients who underwent primary hip resurfacing arthroplasty and total hip arthroplasty at HSS between March 2014 and February 2019. They enrolled 243 patients who had chosen a hip resurfacing arthroplasty, based on their desired athletic level after surgery, for analysis at baseline. Because some patients withdrew over the course of the study, they were able to prospectively evaluate 206 patients. Only patients who had surgery at least one year prior were enrolled.

The participants underwent MRI scans and blood serum ion testing and were asked to complete assessments using the Hip Disability and Osteoarthritis Outcome Score survey every year for four years—at baseline and years one, two, and three. Their MRIs were read by a single radiologist not involved in their care who looked for the presence of synovitis (joint inflammation), synovial thickness, and synovial volume.

Potter utilized metal suppression software that enabled the reading radiologists to better see what they were looking for. With the software, “you can now see nuances, whether the adverse tissue reaction is an infection or typical of metal-on-metal or a problem with the plastic,” Bostrom says. The number of patients with infections is low: Less than 2% of cases are infections, Potter notes.

Patients May Be Asymptomatic
The HSS researchers found that MRI identified adverse local tissue reactions in 25% of patients who had hip resurfacing arthroplasty. Potter says the finding was surprising because the patients who had reactions had symptoms similar to, or even less severe than, patients with ceramic-on-polyethylene and metal-on-polyethylene total hip replacements. In fact, they found that patients who received hip resurfacing arthroplasty had a significantly larger volume of joint tissue reaction on MRI—their risk of having tissue complications was nearly five times higher than that of patients who had ceramic-on-polyethylene total hip replacements.

“We found that patients can be completely asymptomatic and have high-functioning hip scores while harboring reactions that could start to destroy the soft tissues around the hip,” Potter says. “If you have an adverse tissue reaction, and it goes unchecked, it could destroy the tissues in half your pelvis.” 

Koff adds that the benefit of finding adverse reactions early cannot be underestimated. “If a patient has an adverse tissue reaction and you don’t find it early, if you have to do a revision, the revision surgery is going to take much longer,” he says. “It’s a lot more challenging, the recovery is longer, and there is a greater cost associated with every aspect of it, including the rehabilitation.”

Koff and Potter are also collaborating with Kevin Koch, PhD, at the Medical College of Wisconsin, on using MRI to detect adverse reactions with metal-on-polyethylene and ceramic-on-polyethylene hip implants. “Not all patients have adverse tissue reactions associated with tissue destruction,” Potter says. Particles can break off from metal-on-polyethylene and ceramic-on-polyethylene implants and cause bone loss but not soft tissue destruction. This “polymeric” reaction has “a very characteristic appearance on MRI,” she says.

In addition, adverse tissue reactions don’t always occur right away. “We’ve seen them at different time intervals,” Potter says. “Generally speaking, if someone is stable after five years or so, it’s rare that a new adverse reaction will occur, but we have seen rare reactions occur in some people whom we thought were fine.” It could be a certain load of metal that stimulates the host mediated response, she says. “We don’t know that yet.”

Koff adds that the research results are significant not only for patients who received a hip resurfacing arthroplasty but also for patients who receive a ceramic-on-polyethylene or metal-on-polyethylene implant. He also notes that, because MRI is nonionizing, an annual MRI would not pose radiation risks for patients.

Surveillance Recommendations
As a result of their findings, Potter and Koff recommend that most patients with hip resurfacing arthroplasty as well as total hip arthroplasty undergo annual MRI to determine whether they are having an adverse tissue reaction. Koff says an MRI is needed, rather than an X-ray, because “the X-ray may look fine. But that doesn’t mean there isn’t anything amiss.”

Bostrom says the Koff-Potter study is cutting edge and groundbreaking. “It will revolutionize the way we assess total joint replacement. I don’t think you can understate that,” he says. For example, one of his patients had a metal-on-metal total hip replacement and minor symptoms, but Bostrom did not think the patient had a serious adverse tissue reaction. “The patient’s plain X-rays were completely normal. He had a CT scan and a ton of other tests done and everything came back negative,” Bostrom says. “The patient had some bursitis and I thought that was the reason for his pain.” After an MRI was performed, Potter called to tell Bostrom that the patient had a “tremendous amount of damage to soft tissue around his implant,” Bostrom recalls. The patient underwent revision surgery, changing his implant’s metal head to ceramic.

However, Bostrom believes that not all hip replacement or hip resurfacing patients need an annual MRI. “Metal-on-metal should be imaged on a fairly regular basis,” he says, “but some implants we know are doing well and so there’s no reason to do surveillance MRI on them.”

Geoffrey Westrich, MD, former research director of the adult reconstruction and joint replacement service at HSS, agrees that it would certainly be worth following high-risk hip replacement patients with MRI. However, he says, if tests show they have no cobalt or chromium ions in their blood, ordering an MRI on an annual basis couldn’t be justified. “It would be extremely costly for the hospital, the patient, and the insurance carrier, and, in this era of cost containment, we have to pick carefully what tests we’re going to have. If the hip has no cobalt or chrome, that’s not the patient I would order an MRI for.”

Westrich also cautions that not all radiologists are able to use MRI to evaluate hip implants as successfully as Potter and those at HSS. “With MRIs, it’s very dependent on who’s doing it and whether they are dedicated to figuring out the best methods of doing metal suppression MRIs,” he says. “MRI quality and technique vary tremendously from location to location. At our institution, we have the highest-resolution MRIs, the most powerful magnets, and the latest technology.” Institutions with older MRI machines that haven’t been upgraded may not have as much success or be as familiar with metal artifact reduction techniques, Westrich says.

Next Steps
Koff says the next steps are to continue to follow the patients in their study and look for adverse tissue reactions in years five, six, seven, and beyond.

The HSS researchers are also hoping to develop a neural network, an AI-based learning tool, that will allow a more automated assessment of the MR images, Potter says. “Not many people are used to looking at MR of metallic implants,” she notes. “It’s like reading a chest X-ray for the first time. There’s a learning curve to figure out what’s normal, what’s infection, what’s an adverse tissue reaction.”

Potter, Koff, Koch, and Robin Ausman, a programmer analyst at the Medical College of Wisconsin, have an abstract, supported by the National Institutes of Health, which they submitted to the International Symposium on Medical Robotics held at Georgia Tech in mid-November. Their abstract says that automated and objective radiomic measures can be used to derive soft tissue classifications performed by expert radiologists.

“This finding has potential significance as a quantifiable diagnostic tool for further scientific exploration into imaging signatures that differentiate and/or predict rapidly deteriorating soft tissue reactions, allowing for more expeditious clinical intervention with revision surgery,” the authors conclude.

— Beth W. Orenstein, of Northampton, Pennsylvania, is a freelance medical writer and regular contributor to Radiology Today.