Chat with us, powered by LiveChat

RIP Knee Replacement: New Study Highlights Poor Results

POSTED ON IN Latest News BY Chris Centeno

knee replacement research

Some studies are so significant that they massively change how medicine is practiced. While medicine is slow to change, they are the “beginning of the end” for a procedure or a therapy. I have covered several meniscus surgery studies over the last few years that are in that category. This week such a study was published on knee replacement, basically indicating that the real results of the procedure are poorer than we’ve been lead to believe and recommending against the practice for many patients. So is this the beginning of the end for knee replacement surgery?

Knee Replacement by the Numbers

In one 2006 study, it was estimated that approximately 700,000 knee replacement surgeries were completed per year in the U.S. This number was projected to explode to roughly 3.5 million operations annually by 2030. So it’s a safe bet that we’re likely up to at least 1–1.5 million knee replacement surgeries per year in 2017. I have blogged extensively before that the serious complication rate for this procedure is 8–16% and includes death, infection, blood clots in the lung or leg, loosening of the prosthesis, fracture, and the need for more surgery.

Lackluster Randomized Controlled Trial Results

Last year I blogged on the world’s first randomized controlled trial of knee replacement. The research showed that the benefits were much smaller than you would expect, despite the university talking heads proclaiming otherwise. For example, by the end of the year, 3 in 4 patients who had moderate to severe arthritis and who were in the physical therapy comparison group ultimately decided that they felt good enough that they didn’t need knee replacement. More importantly, of those who did have their knee replaced, only 1 in 5–6 patients reported that they got more than 15% functional improvement. For more info, check out my video below on the study:

Surgeons Believe Knee Replacement Works Well

These past few months, a few of my blog posts highlighting research that orthopedic surgery doesn’t work have gotten some pushback by orthopedic surgeons on LinkedIn. One remarked that knee and hip replacements were, in fact, some of the most successful surgeries ever created in the twentieth century. In general, this opinion is generally widely held by surgeons. However, many of my patients who seek me out for stem cell injections to avoid knee replacement are physicians who have “been to the puppet show and seen the strings.” Meaning that they have cared for the walking wounded who either didn’t do well with the invasive procedure or had a serious complication. So when it comes to the point that they need their knee replaced, they want no part of the procedure.

The New Study

This new research looked at patients who were already enrolled in the U.S. Government-funded Osteoarthritis Initiative Study (OAI) or the Multicenter Osteoarthritis Study (MOST). These studies track knee arthritis patients who fill out pain and function questionnaires, get periodic X-rays and MRIs, have chemicals in their knee fluid measured, and so on. The two investigations are some of the largest knee arthritis research studies every conducted. Because they are tracking patients over time, they can look to see if certain interventions are helping these patients or not.

In both arthritis tracking groups, knee replacement surgery demonstrated quality-of-life improvements that were small across multiple types of functional questionnaires. These improvements were better when the patient was more disabled before the surgery. Let me write that again: amputating the knee joint and inserting a prosthesis didn’t help most patients!

The study authors then looked at the cost-effectiveness of the procedure. Knee replacement was not cost effective for most of the patients who received the procedure. It only became so when the patients were more disabled. Again showing that the only patients who benefit greatly from the procedure are those with severe functional limitations (e.g., unable to walk long distances, climb stairs, or otherwise use the knees to function).

What was interesting was that the researchers used a double-validation method across the two data sets. This means that they made sure that these two independently collected sets of extensive information from knee arthritis patients agreed with one another. So they ensured that the poor showing for knee replacement was observed in both the OAI and the MOST studies.

How Could a Bunch of Lower-Quality Knee Replacement Studies Show Efficacy?

There are hundreds of lower-quality studies that purport to show that this or that great new knee replacement device is the best thing since sliced bread. How could all of those studies be in conflict with this one? The simple answer is surgeon bias. Let me explain.

Orthopedics is one of the only fields in medicine where the treating physician’s assessment of the outcome factors heavily into the final scores for most outcome metrics. For example, while a patient may say that he or she got fair relief from a procedure, the surgeon can supercharge that result by adding his input to the score. Hence, many surgical studies have inflated outcome scores. This new study didn’t use physicians invested in the outcome. Instead, it used government workers having patients fill out questionnaires. So any physician or medical-provider portion of any functional metric was filled out without any knowledge that one day some researchers would slice and dice the data looking for knee replacement results.

Why This Study Is the Beginning of the End for Knee Replacement

Knee replacement is about a $100 billion a year industry if you add the cost of the prosthesis, the cost of surgery and anesthesia, hospital charges, and the amount spent on complications. As a result, I expect this most recent study to be fought tooth and nail. For example, if knee replacement stopped being reimbursed by insurance companies, Medicare, and socialized insurance plans, the layoffs in the medical-device industry alone would be staggering. Many hospitals would literally be short on their payroll. Hence, studies like this tend to be only the beginning of the end. For example, the last meniscus-surgery study showing a complete lack of results was just published in 2016, and this surgery is still going strong. The first study showing that arthroscopic knee cleanups were useless was published in 2002, and it took at least a decade before the practice mostly faded away.

The death of knee replacement, however, could be hastened by orthobiologics or the use of platelet rich plasma (PRP), stem cells, growth factors, and cytokines delivered through minimally invasive injections. Based on our existing published research and new research not yet published on the Regenexx same-day stem cell protocol when used to treat moderate or severe knee arthritis, knee replacement now has much less invasive competition. Also, for more mild arthritis, other research has shown PRP works well.

So Based on This Study, Who Should Get Knee Replacement, and Who Shouldn’t Get the Surgery?

If you’re a young and active individual who can’t exercise hard because of knee pain and arthritis, this study shows knee replacement is not the right surgery for you. If you’re an older person who has some trouble climbing stairs but can still walk the mall or a significant distance once your knee “warms up,” this study also shows that knee replacement isn’t for you. However, if you’re someone who can’t walk more than a block without severe knee pain or climb any stairs and you have failed other treatments, knee replacement may be well worth the risks.

The upshot? This paper is yet another major groundbreaking study that shows that a common orthopedic procedure performed in more than a million US patients a year is mostly ineffective. This result adds to a long list of common orthopedic surgeries that are ineffective. So if your knee hurts, consider newer orthobiologic therapies such as platelet rich plasma or stem cells, and stay away from getting your joint amputated and having metal or plastic parts inserted!



    Sarala D. Johnson MD(retired) says

    Have had PRP twice, multiple synvisc 1 Q 4- 6 months, arthrosopic cartilage tears removed at HSS, NY, since 1999(once each knee) over 20yrs ago. Poor spine due to spinal stenosis with subsequent 3 spinal fixations with donor bone at lumbar-saccral region & cervical fixation at the NYU, with spinal screws and bone graft. The latest L-S procedure was at the University of Pensylvannia neuro surgery on 9/8/2015 with with spinal screws and bone graft from my iliac crest, I am still in pain for all (osteoarthritis).
    With great expectantations on Regenexx and , esp, since I was an an athelete into my 30s good at Asian sports, now unable to walk comfortably, I would like Dr Centro to aid me to recover some of my former abilities. AGE could be factor but I hope your trainees will not be hindered by this and have more mindfulness, Thank you I am 73.


    Chris Centeno says

    Dr. Johnson, I would have to take a look at all of your information to give you a sense of whether these procedures can help you and if so, how much improvement would be likely. Best to contact my office.


    Dr. Scott Davis says

    I have surgery scheduled for 4/18/17, but I think I should cancel that for a visit to one of your affiliate physician. I'm not a rich man who can put out $7,000, but if I don't do something about my knee soon, I will not be able to function. I've read and kept up with your updated information for a couple of years now. The only thing that is not clear to me, I saw the blind study for adverse effects and one that has similar issues are those with spinal cord and deformity issues. I am a C3-6, T7-S1 fusion and two pedicle bolts from the bottom of the rods into the hips. Yes, I have spine problems.
    The knee injury prompted arthroscopic surgery few years ago, but I believe that only caused the progress of further destruction of the tissues that cushion the knee joint. After the weekend of prayer and research, I will make a decision on this. Many factors play in the decision of which the study results are probably the most important part to date.
    This was a good study that shows about a 2% problem with MSC v. 12% with the replacement on average. The longevity of stem cells therapy remains a question for me. I understand that after a year I may need some platelet rich plasma or something like for a booster? After that I don't think I know how long will this work for previous patients? Years and decades I hope. Is there any results for that kind of study?


    Chris Centeno says

    Scott, if you're severely disabled from the knee arthritis, as the blog says, TKA might help significantly. Your extensive fusions in the spine would be a headwind for knee stem cell treatment. Most patient reponders who have severe knee arthritis do need booster shots at usually the 2-5 year time frame.


    Dr.somanna from India says

    Highly biased opinion.. The study cited talks about TKR in early oa.. Just like any other procedure wrong indication will give poor results...


    Chris Centeno says

    Dr. Somanna,

    We are actually making the same point. As was stated in the conclusion of the blog, there are patients for whom knee replacement is appropriate. The issue is that they are being done in patients for whom it is not appropriate.


    Diann Harper says

    I have diabetis & am overweight. Would that prevent me from having this treatment?


    Chris Centeno says

    Your weight, no. Your diabetes, maybe. Blood sugar needs to be under good control and there are some diabetes medications that could interfere with treatment. These things would be discussed during the candidacy Review. If you'd like to see if you'd be a good Candidate, please submit the Candidate form:


    Joe Cannatelli says

    Are your comments only related to knee amputations /new stem-ball --to replace knee?

    Our doc is talking about resurfacing both knee ends and adding new surfaces by glue or compression methods also some work behinf kneecap==no parts removed

    Procedure is 30 minutes going in from front and no cutting of tendons--whatever--1 day in hospital 2-3 weeks therapy

    see details on ortho website ROTHMAN Institute HQ Phila w/ multiple offices also in NJ

    Thank you


    Chris Centeno says

    My comments apply to all types of knee replacement. Please see:


    Joe Cannatelli says

    is comment correct that cost of stem cell procedure is in range of $7000 per knee? Includes harvesting the cells?

    If not Medicare approved then hospital/anesthesiologist and PT f/u visits would be additional expenses-correct?

    I think you are on right path with stem cells--blood and bone marrowc harvesting is being used successfully in cancer treatments

    joe cannatelli


    Chris Centeno says

    Regenexx procedures are done in fully equipped procedure suites, not in a hospital, so there are none of the additional charges associated with hospital procedures, and the cost is all inclusive. Prices are not "fixed" but are based on what each individual patient needs. PT following the procedure if prescribed would be handled by your insurance the same way as any other Physical Therapy. Please see: and and and


    Add comment

    Your comment will be revised by the site if needed.

    About the Author

    Chris Centeno

    Christopher J. Centeno, M.D. is an international expert and specialist in regenerative medicine and the clinical use of mesenchymal stem cells in orthopedics. He is board certified in physical medicine as well as rehabilitation and in pain management through The American Board of Physical Medicine and Rehabilitation.…

    View Profile

    Search Blog