Not So Fast on That Knee Replacement

Here’s a short piece by the New York Times that’s well worth the read if, driven to desperation by pain and a bleak prognosis, you’re considering the ultimate in knee surgery: a total replacement of the joint.

You wouldn’t be alone -- far from it. More than 600,000 of the surgeries were performed in 2012. That’s a big jump from the 250,000 of 15 years ago. But what’s most interesting is where the most rapid growth is: among those 45 to 64 years old, who had triple the number of operations as before.

Are all these surgeries beneficial, especially among younger patients?

Researchers analyzing data from major studies found that people with really bad knees were helped by surgery. “Really bad” in this case means advanced arthritis: in other words, severe pain and impaired physical function, like an inability to climb stairs. But others with less serious arthritis saw only a very small benefit.

The upshot? According to Daniel Riddle, the professor of physical therapy and orthopedic surgery who led the studies:

If you do not have bone-on-bone arthritis, in which all of the cushioning cartilage in the knee is gone, think about consulting a physical therapist about exercise programs that could strengthen the joint, reducing pain and disability.

Amen. Surgery sometimes is the best option. But it’s often the best option when it’s the last option.

Three Reasons Why “Mistracking Kneecaps” Probably Isn’t the Reason for Your Pain

Last time I mentioned in passing an interesting Swedish study. It found clinical tests perceived no significant differences between subjects who had knee pain of unclear origin and a control group. So, in other words, the knee pain sufferers weren’t crooked or imbalanced in some way the control group was not.

Intrigued, I managed to locate the study (most are behind paywalls, but luckily, this one was not). It begins with a fairly broad discussion of patellofemoral pain syndrome that won me over with these two lines:

Some practitioners who find no identifiable cause to the pain use both the term PFPS as well as the term AKP (anterior knee pain), but the terms are best reserved to describe the patient who has yet to be evaluated. If no causative explanation for the pain is found, despite a thorough investigation, the term idiopathic anterior knee pain (IAKP) seems reasonable.

Yes, yes, yes! Let’s stop pretending PFPS is a real diagnosis. “Idiopathic anterior knee pain” is more honest and useful. Basically, it means “you have pain in the front of your knee and we don’t know why.”

There is another discussion section, at the report’s end, that is well worth perusing too. The researchers’ skepticism about catchall explanations for PFPS that cite mechanical abnormalities is virtually palpable.

Here are three big problems with the “oh, you’re crooked/imbalanced” line of thinking.

(1) There’s no accepted definition of what constitutes crooked in the first place -- or more precisely “meaningfully crooked” if you will, because I’m sure very small discrepancies in the length of someone's legs (or in whatever) wouldn’t be considered important even by diehard structuralists.

To make this more concrete: Say you believe patellar maltracking causes most cases of PFPS. Well, if a kneecap doesn’t track perfectly by 1/100th of a millimeter (the width of a thin hair), that’s not enough to be significant. But then, what is? 2 millimeters? 6? 10, 20? The fact is, no one has set forth an assertion on this that’s supported by clinical evidence. So we don’t even know what crooked is.

(2) Also we can’t measure it well anyway (a related, overlapping issue). The Swedish researchers report:

“Fitzgerald and McClure (1995) studied four different manual clinical tests for patellofemoral alignment where measurement reliability ranged from poor to fair ... they were unable to find a reliable clinical method for assessing alignment.”

So there’s no accepted definition of malalignment and no good way of measuring it anyway. But wait, it gets worse:

(3) “Fairbank, Pynsent, van Poortvliet and Phillips (1984) reported that in pain-free subjects, between 60% and 80% of the population fall into what is generally classed as lower extremity malalignment.”

So, even when someone does take a stab at defining malalignment, it turns out -- surprise -- that most of us who are pain-free share this “problem.” In that case, if almost everyone is crooked/imbalanced, what’s so special about it?

And the answer just may be: not much at all.

Corrective Exercises: A Waste of Time?

I found a very interesting article not long ago. It very much reminded me of my “awakening” during my struggle with knee pain.

First, start with the entrenched thinking (flawed) on what causes “patellofemoral pain syndrome.”

American Family Physician describes the cause of PFPS as an imbalance of the forces that keep the kneecap in alignment during knee extension and flexion. This imbalance can increase the risk of muscle dysfunction, poor quadriceps flexibility, overuse, trauma and a host of other musculoskeletal problems. In other words, during PFPS, the kneecap does not glide back easily on its “track” to the femur. ... Some health professionals, such as physical therapists and athletic trainers, recommend corrective exercise as a self-care method for patients.

Sounds simple, straightforward, reasonable. Except:

Despite the prevalence of corrective exercise prescriptions, current evidence shows that this intervention may not always effectively treat knee pain and could be a waste of time.

The problem is, the idea behind corrective exercise is that you’re crooked (your kneecap is mistracking) or that various muscles or tissues are too tight or too loose. But “studies have shown that PFPS may not always be a biomechanical problem.”

A 2006 Swedish study is then described, one that I was previously unaware of. Eighty patients with PFPS were examined. Of those, 29 had no identifiable cause of their PFPS (the others either had “slow bone turnover disease” or a type of pathology of the knee, and a small number dropped out.) For the 29 who didn't have a clear cause for their pain, “researchers could not differentiate between [them and] the control group that had no knee pain and were not diagnosed with PFPS.” So they weren’t identifiably crooked or imbalanced in a way that the control group was not.

Of course there is evidence that exercise can reduce knee pain, but as Paul Ingraham says in the article, that’s “probably not because it’s ‘correcting’ anything.”

I’ve linked to Paul’s website a few times over the years, such as to this essay where he examines the obsession that physical therapy has with crookedness/imbalances. I like his thinking and he’s a good writer. In the article above he is quoted saying, regarding the misalignment theory (the underlining is mine):

Exercises are prescribed in the hope that such things can be corrected, usually by strengthening and stretching.  Unfortunately, a lot of exercising for these goals is often out of tune with how exercise actually does help patients.

I couldn’t agree more with that. And, finally, he does well to note that knee pain comes in many stripes, with many possible causes. So, he notes:

Exercise is no kind of magic bullet. Patellofemoral pain has many faces, many possible causes and complications, and some cases do not respond to any kind of exercise, ‘corrective’ or otherwise.

Read On for the Top Risk Factor for Knee Pain

“Broken record” has a pejorative connotation. No one likes a “broken record” who harps about one thing in particular, all the time.

When it comes to knee pain though, I’d argue that it’s not bad to be a broken record about one thing anyway. And that’s the single factor that, more than any other, predicts whether you’ll have knee pain.

It’s simple:

Being overweight.

A new meta-analysis of existing studies found that “one-fourth of cases of onset of knee pain could be attributable to being either overweight or obese,” according to researchers at the Arthritis Research UK Primary Care Centre.

That may elicit a yawn from you, especially if you think I’m a bit of a broken record on this subject. But here’s the part I found shocking:

5.1 percent of new knee pain/knee osteoarthritis could be attributed to a previous injury and 24.6 could be attributed to being overweight or obese.

Wow! Didn’t see that coming. What a disparity. If anything, I would have expected a previous injury to be a more significant contributor to the onset of knee pain. But it isn’t. And it’s not even close.

If I were an orthopedic doctor, and I had a patient with knee pain who was overweight and who claimed to be serious about doing whatever it takes to get better, I’d say:

“Lose x pounds. That will show that you’re really serious.”

Because if you are serious, and you are overweight, shedding some pounds has to be a No. 1 priority. On this point, the evidence isn’t debatable.

Knee Pain and the Influence of Genetics

Is knee pain a family affair?

A recent study shows there’s apparently a gene-related link to the development of knee pain associated with osteoarthritis.

The study included 219 subjects, average age 48. Roughly half were the children of parents who had knee replacements; the rest belonged to the control group. The knee pain of all subjects was assessed three times: at baseline, two years later, then eight years later.

Even after the proper adjustments were made to control for such factors as age, sex and BMI, “individuals with a parental history of knee replacement had a more than twofold greater likelihood of worsening total knee pain.”

Interestingly, the adjustments were even made for radiograpic and MRI abnormalities. So that meant researchers were comparing people of similar age, sex and BMI who also had similar-looking MRIs and X-rays. Even then, the offspring of the knee replacement group had a twofold greater likelihood of worsening pain.

The study’s authors speculated that “implies that the genetic contribution to knee pain may be mediated through factors outside the joint, possibly involving pain processing.” I interpret that to mean that, if you’re a child of a patient who had a knee replacement, you may be more likely to experience a worsening of pain simply because you may be more sensitive to it -- an interesting and curious finding.

In any event, I’d argue that the takeaway is that, if you’re in this high-risk group, being proactive about not developing knee pain in the first place makes a lot of sense. Build up the leg muscles around your knees. Find good joint-friendly activities (cycling, walking). Take good care of your knees before they start to hurt.

Did My Knees Really Get Better Or Do They Just Feel Better (And Does It Matter)?

A long title but I couldn’t think of a good shorter one.

After reading my story, some people say something like this:

Why don’t you get another MRI (or some other test) that shows whether your knees really healed? That would prove whether your program really worked.

To be sure, this kind of comment has never been phrased in a hostile way. There’s no implication I’m a liar or fraud. Rather, people have a deep curiosity -- the same as I do actually -- about what changes physically occurred in my knee joints between the worst days of my condition (when I was suffering each day in Hong Kong) and now.

If I could do such a test (for cheap), and it could measure such a thing, I’d do it in a heartbeat. Hell, I’d love to know. Armed with the test results, I could probably sell five times more books. ;)

But it’s not possible. Here’s why.

(1) Simply having an MRI done would cost a lot; I’m guessing at least $1,000. And my health insurance company isn’t going to pay for a “there’s nothing wrong but I’m curious about what my knees look like” MRI.

(2) I don’t have an ideal MRI to compare it against anyway. True, I had images taken in Hong Kong in early September 2007, but that was before my disastrous experiment with weightlifting to strengthen my quads (which really trashed my knees). An MRI done in November or December of 2007 might have shown more damage.

But here’s the big reason:

(3) My original MRI exam was only somewhat useful in identifying my problem and determining the extent of it. Actually, “somewhat useful” may be a too-kind phrasing. My MRI basically said I had changes consistent with mild chondromalacia. So it found no giant potholes in my cartilage that a subsequent exam might show had healed.

My suspicion is that if a surgeon had cut open my knee, he would have spotted some kind of more obvious cartilage damage not detected by the MRI. But I never went that route (thank God). So, like many knee pain sufferers, I don’t have a good baseline test that says, “Wow, your cartilage is really messed up!” I suspect I was on the verge of going downhill fast, but I was fortunate to fix my knee pain in the relatively early stages.

So how do I know there was damage in the first place and I didn’t just suffer from some weird neurological ailment?

Well, as I relate in the book, there was a lot of noise from my knees when I dropped into a squat (as if I were about to sit in an invisible chair). My knees were so loud that my doctor felt compelled to be blunt and opine that the joints would never get better. Also, when I went into a deep crouch -- which was hard to do and uncomfortable -- and then straightened up, there was a loud, ugly “ripping” noise.

But now those disconcerting sounds, whatever they indicated, are either gone or much quieter (I can still hear a little crunchiness in my knees, but it’s not painful at all). I’m back on the bicyle, riding as hard as ever. I can sit at my desk again through a long 10-hour-plus day without issues.

Do I have a before/after set of tests that shows the improvement? No. But even if I did, would it matter? I’m not so sure it would. Just from the way I feel, I know something in my bad knees sure as hell healed/improved significantly. And I’m pretty confident of that, whether or not those changes could be detected by an MRI or some other test.

Running Not Only Doesn’t Ruin Your Knees, But May Help Them

One of the most persistent knee pain myths is that, if you're a runner, the sport will eventually exact its due and lead to sore, aching joints. If you run long enough, the pessimists warn, you’ll pay the price.

A recent study (the results of which were presented at the American College of Rhematology’s annual meeting) came to a much different conclusion.

Namely, 29.8 percent of non-runners in the study had symptomatic osteoarthritis compared with 22.8 percent of the runners (I have to assume they controlled for weight in their calculations, as that’s a Statistics 101 sort of thing to remember to do.)

The study was huge (2,683 people, who had an average age of 64.5), which is good, but it did have a kind of squishy longitudinal component. Anyone who was a regular runner at some stage in their life, even if it was only between the ages of 12 and 18, was tagged as a “runner.” That can be problematic, as anyone knows who remembers my analysis in the book of one longitudinal study in particular. Still, this would not be the first study to suggest running is beneficial for knee joints, so I think the conclusion makes sense.

One of the study’s authors does raise a caveat: “This does not address the question of whether or not running is harmful to people who have pre-existing knee OA,” according to Grace Hsiao-Wei Lo.

But that’s okay. That’s a lesser issue. Running generally isn’t a good idea if you already have osteoarthritis in your knees, especially if it’s painful (there are better activities that are lower impact on the joints). The good news is that evidence increasingly shows that running doesn’t cause that osteoarthritis.