Is THR or Hip Resurfacing more durable?

Hip resurfacing is more durable than THR in all patient subgroups. Currently, my 17-year implant survivorship is 99%. Only ceramic-ceramic THR in Young-Hoo Kim’s series from Korea can equal this outcome in young patients (but they still acn't function as well due to the stem).

Even if hip resurfacing had a higher failure rate, I would choose this operation for myself because of the more normal function it allows. In fact, this is the reason why I started to perform hip resurfacing in 1999. At that time early data from Derek McMinn in Birmingham England came to my attention. Metal-metal hip resurfacing seemed to work well in the short run, solving the early failures of metal-on-plastic hip resurfacings encountered in the 1970s.

The problems young people faced with THR were well-known. Many of the early adopters of hip resurfacing reasoned that patients would be well-served if we could buy them 10 years with a hip resurfacing to delay the inevitable THR. Surgeons and our patients took a chance to see how many could get ten years out of a hip resurfacing. This gamble paid off. My first cohort of Corin hip resurfacing patients achieved 89% ten-year implant survivorship and 84% 20-year implant survivorship. Not only did we buy most patients 10 years with a bone-preserving implant, but they had a higher chance of making it 10 or 20 years than if they had a THR instead.

As hip resurfacing has improved, implant survivorship has now reached 99% at 17 years in my practice. The only THR that can approach these results in young patients is ceramic-on-ceramic uncemented THR, which is about as infrequently offered to patients in the US as metal-on-metal hip resurfacing. The main reason that ceramic-on-ceramic is not used is the 5-10% rate of loud squeaking reported in many studies. US patients don’t accept this problem and request a revision for an otherwise well-functioning THR that is squeaking, resulting in a high overall failure rate. Korean and French patients don’t seem to be bothered by squeaking as much because there are few revisions for squeaking reported from these countries. If you can tolerate a 5-10% chance of loud squeaking from your hip, then a ceramic-on-ceramic THR can last as long as a metal-metal hip resurfacing.

In other words, if loud squeaking is considered a failure (and revision is performed) then ceramic-ceramic THR has a higher failure rate than metal-on-metal hip resurfacing. In the Korean and French reports, no one gets revised for squeaking; I wonder if squeaking is not happening, or if the patients are more accepting of this problem.

A related question is whether hip resurfacing is as durable as THR in all patient groups. The answer is that in the last 10 years in my practice, all patient groups have had a lower 10-year failure rate with hip resurfacing than THR except when compared to expert surgeon series of ceramic-on-ceramic THR published from Korea.

In the first 10 years of my hip resurfacing experience, young men (under 60) with osteoarthritis enjoyed the lowest 10-year failure rate, far surpassing THR failure rates in young patients. Women had a higher failure rate than men, but young women still fared better with hip resurfacing than if they had a THR. Older patients had a higher failure rate with hip resurfacing, mainly because THR failure rates in older folks were lower than in young patients. Patients with dysplasia (mostly young women) and osteonecrosis (mostly young men) had a slightly lower failure rate with hip resurfacing than THR.

Now, 17-year failure rates for my hip resurfacings stand at 99% for all groups, surpassing THR in every category except for ceramic-on-ceramic as discussed previously. This even surpasses the 95% 10-year failure rate reported in registries for patients with a mean age of 70 years. This is especially remarkable because my patient cohort is much more demanding on their implants; 40% participate in impact sports. They are more active as a group than the average THR patient, yet their hip resurfacing still is more likely to last longer.

In “benchmark” registry data (England, Sweden, and Australia) the 10-year implant survivorship for THR is 95% at age 70 and 83% at age 50. My hip resurfacing 17-year survivorship is currently 99% far surpassing all registry benchmarks for any age, diagnosis, or sex.

Registry data (benchmark data)

When comparing the durability of joint replacement, the typical measure in most studies is implant survivorship. The accepted statistical method is using the Kaplan-Meier (KM) formula. The endpoint that is typically measured is any failure resulting in revision surgery. Patients with various lengths of follow-up since surgery are entered, revision dates are entered, and dates that patients pass away are entered. The formula then provides the estimated implant survivorship for an interval corresponding to the patients with the longest follow-up. If I have been performing a certain operation for 15 years, I can estimate 15-year implant survivorship even though some patients in the study were operated on only 2 years ago. That is why KM data is labeled as estimated.

The major joint registries (Australia, Britain, and Sweden) all report data this way (most individual surgeon series also use KM statistics, I report my data this way). These three are the best joint registries in the world. In these countries, all joint replacements in the country are recorded with the registry by law. All revision surgeries are also registered. Therefore, when any operation is revised later it is considered a failure, and the time to failure is known. We think that most replacements and revisions are entered, but I have been unable to find any reports verifying how high the capture rates are. I believe that patients who die are also captured from death records. There is also an American registry, but it has a very low capture rate and is, therefore, unable to provide any meaningful data on implant survivorship.

Registry data can be analyzed for different implants, but not for individual surgeons. Therefore, it is helpful in providing average surgeon data. The mean age of THR patients in all registries is 70 and the mean 10-year implant survivorship is 95%. This is what the average surgeon can achieve with the average implant in an elderly patient. But, of course, some surgeons are above average, and some are worse. Individual published surgeon series (scientific papers, not registry reports) can give you a clue on how the individual surgeon performs if compared to registry benchmarks. Most surgeons do not provide this data to patients, but you can search for it on PubMed. Most surgeons never collect this data and can therefore not give you accurate data on their outcomes. My data is reported on this website in “Latest Results” and also in the form of peer-reviewed scientific papers in “Scientific Publications”.

While the three major registries are a great source of benchmark data, they do have their flaws as any scientific report does. Other than the major problem of the high degree of individual surgeon variability, the following are problems that you need to be aware of:

• Some surgeons may be better at performing THR, others better at hip resurfacing.
• If a hip resurfacing suffers a femoral neck fracture it is “revised” to a THR (failure), but if a THR suffers a fracture it may be repaired around the stem with cables, which is a similar magnitude of surgery but is not counted as a revision (negative bias for hip resurfacing).
• Hip resurfacing is a relatively new operation with many surgeons performing very few of them without prior specific training, while THR has been around for over 70 years with many surgeons performing high volumes of them (Negative bias for hip resurfacing).
• Age of the patient has a strong effect on THR durability. Young people have a much higher failure rate. There are several possible reasons for this. First, young people on average are more active and place more wear/stress on any implant. Second, older people are less likely to be revised if their implants fail. They just live with a failed implant because they are unable to tolerate revision surgery or choose not to have a revision. These failures continue to count as a “successful” implants according to registries. Older patients have a much higher rate of dying within 10 years, therefore positively biasing the implant results. Diagnosis may also be a major source of age bias. Most older patients have osteoarthritis, while many younger patients have more complex conditions such as osteonecrosis, dysplasia, and posttraumatic arthritis, which all are known to have worse outcomes with THR (negative bias for hip resurfacing). It is therefore critical to examine the age of patients in any report on THR or hip resurfacing. Younger patients are more challenging. Implant survivorship achieved in older patients simply does not apply to younger patients. Most THR surgeons purposefully ignore this fact when they recommend THR for younger patients and quote them implant survivorship based on a much older group of patients. Benchmark registry implant survivorship for mean age 70 patients is 95% at 10 years. For patients under 50, it is 83% at 10 years and 50% at 20 years. My hip resurfacing survivorship is 99% at 17 years for any age and any diagnosis. THR doesn’t even come close.