Keep Running Past the Age of 40 and Your Knees Will Fall Apart! True or False?

I'm an enemy of received wisdom, especially on the subject of what constitutes proper care of one's knees, because there are so many mistaken beliefs out there.

For example, you often hear: Keep running long enough and you'll wreck your knees.

After all, how can all that pounding be good for joints where bones meet, their endings protected only by thin pads of cartilage? And if your knees naturally wear out over time, why hasten the inevitable by subjecting them repeatedly to the high-force activity of distance running?

That's the received wisdom on knees and running. The truth appears to be quite different.

While assembling a bibliography for my book about recovering from chronic knee pain, and sorting through my many files, I came across an article by Gretchen Reynolds that I saved a while ago. Dated Aug. 11, 2009, it appears online at the New York Times Well blog. The provocative title: "Can Running Actually Help Your Knees?"

The opening is attention-grabbing:

An article in Skeletal Radiology, a well-respected journal, created something of a sensation in Europe last year. It reported that researchers from Danube Hospital in Austria examined the knees of marathon runners using M.R.I. imaging, before and after the 1997 Vienna marathon. Ten years later, they scanned the same runners’ knees again. The results were striking. “No major new internal damage in the knee joints of marathon runners was found after a 10-year interval,” the researchers reported.

At first blush, this seems incredible. Of all classes of runners, marathoners train the longest; the race itself of course covers 26.2 miles. Over many years of running, their cartilage should be pounded to dust, right?

To be fair, a skeptic might say, "Not so fast! Perhaps the marathoners were relatively young and thus better able to withstand the stresses (just wait til they get older!). Or they may have been the equivalent of professional athletes, with efficient and streamlined bodies. Who else could run so far, so often?"

Yes, good points, but what about this second study that Reynolds cites:

Stanford University followed middle-aged, longtime distance runners (they didn't have to be marathoners). They were largely in their 50s and 60s (no spring chickens there). The trial went from 1984 to 2002. At the end, the knees of the runners were compared with those of a control group, about the same age, who didn't run.

At the time of their first exams, 6.7 percent of the distance runners had mild arthritis in their knees, as opposed to none of the control group subjects. Which seems to be an argument for inactivity.

Except, by the end of the study, the results reversed. Only one in five runners showed signs of arthritic changes in their knee joints; that contrasted with one in three non-runners. Severe arthritis hobbled almost one in ten non-runners, but only one in fifty runners.

What could explain this?

The article suggests that the knee may develop a "motion groove" (those who read my book will discover me discovering that movement is the magic ticket to knee health). Cartilage adapts to the load of running. I would add further: the tissue gets stronger and stiffer and better able to cope with ordinary daily activities without complaining and becoming sore.

Reynolds tosses in a caveat at the end of her article. Once a runner becomes injured, the "exquisite balance" implied by that "motion groove" can become disrupted, leading to a "degenerative pathway" that causes cartilage to wear down and fall apart -- and then comes pain and arthritis. (I've purposefully put some of these phrases in quotations because I'm not totally on board with the author's analysis here -- I smell a Structuralist lurking -- but that's for another time.)

Injury risks do abound for runners and must be guarded against at every turn. Gaining too much weight is a no-no. Being an inconstant runner -- someone who takes off a few months, then tries to pick up at his old level of intensity -- could lead to problems. Ambition can get you in trouble (for example, suddenly going from running three miles twice a week to ten miles three times a week as you begin training for a race). Stubbornly running through early pain signals is stupid.

But running sensibly doesn't appear to be a knee killer. It may help your joints, in fact (though my exercise of choice is cycling, which is easier to do within knee-safe bounds).

SKEPTIC'S CORNER: If you read the comments after the Reynolds article (299 of them!), you will see skeptics weighing in with arguments that can be summarized along these lines: (1) there is selection bias at work, as the runners were all healthy individuals (2) the runners weighed less than the subjects in the control group (3) anyone who is in their 50s or 60s and still running long distances has, through genetics and anatomy, superior knees and cartilage to begin with.

My thoughts, point by point: (1) The study's authors did recognize the possibility of selection bias, as the runners were indeed all healthy, but two things should be noted. First, could running be part of the reason for that good overall health (exercise has been proven to boost the immune system)? Second, osteoarthritis of the knee isn't considered systemic (unlike, say, rheumatoid arthritis), and so how much effect does good health have on one's knees, if any?

(2) The runners had only a "slightly lower BMI," according to the authors. Yet that difference at the outset is a BMI of 22.3 for runners vs. 24 for non-runners (the gap is maintained through the study), according to a table included with the report. For two six-foot men, that translates into a 164.5 lb. runner and a 177-lb. non-runner. That is significant, unfortunately, and I would say is the biggest flaw in the methodology because of the well-established relationship between excess weight and arthritis.

(3) This, to me, is the least convincing argument. Remember, at the outset, 6.7 percent of the runners had mild arthritis, while none of the non-runners did. If the runners really benefited from superior cartilage/knees, then why did they have worse knees at the beginning and better knees at the end? Also, if you drill down into the results, 44.4 percent of the runners complained of a previous knee injury versus 35.9 percent of the non-runners. So why was that injury figure 20 percent higher for the runners if they started out with superior knees and cartilage?

Is Your Knee Doctor (or Physical Therapist) a Structuralist?

And why should you care?

Doug Kelsey, chief therapist at Sports Center, defines structuralism as "a school of thought that believes the genesis of musculoskeletal complaints is from one or more biomechanical abnormalities."

Further, he says:

For patella pain, the biomechanical abnormalities include a laterally tracking patella, weak medial quadriceps, tight hamstrings, tight iliotibial band, tight calf muscles, weak or tight hip rotator muscles and over pronation of the foot. A Structuralist view would then be to set the mechanics "right" and symptoms would subside.

Sound familiar? Pay close attention to that first one: "a laterally tracking patella." That is, in layman's terms, a kneecap that isn't perfectly centered in the trochlea, or the groove that it's supposed to slide through with the greatest of ease. We'll return shortly to that mistracking patella.

So if your doctor (or physical therapist) says your problems are caused by a poorly tracking patella -- or that you must strengthen hip/butt/quad/whatever muscles and stretch the IT band/quads/hamstrings/whatever in order to redress your body's imbalances -- chances are excellent you have a Structuralist.

But does Structuralism make sense as a model to analyze and treat the majority of people who suffer from chondromalacia or patellofemoral pain syndrome? Logically (and instinctively), the answer is no for several reasons.

(1) Can stretching, one of the solutions in the Structuralist toolkit, really correct biomechanical abnormalities? Stretching temporarily lengthens muscle fibers. Then they contract again. How long must you stretch to achieve a lasting, beneficial effect? Answer: it's unclear. Paul Ingraham, massage therapist and stretching skeptic, does quote this therapy-exercise textbook (in his comprehensive online essay looking at how stretching fails to deliver what it promises):

Several authors have suggested that a period of 20 minutes or longer is necessary for a stretch to be effective and increase range of motion when a low-intensity prolonged mechanical stretch is used.

That's a lot of time to devote to a single stretch, for a single muscle, as he notes. Then how often would you have to stretch like that? Once a day? Once every ten hours? Six hours? And, even supposing that stretching can change your biomechanics, how are you supposed to be able to tell when you've reached the sweet spot, of just the right amount of change, and not too much (after all, you don't want to have your patella start tracking to the left because you overcorrected for its tracking to the right, and too much flexibility does lead to unstable joints)?

(2) If the Structuralists are right -- if your biomechanics are at fault -- why do chondromalacia and PFPS usually strike at older ages: 30, 40, 50? Let's look at the commonly blamed factor of kneecap mistracking. If that's to blame for knee pain, wouldn't it become a problem soon after you learn to walk? Why aren't there more three-year-olds with PFPS?

Okay, that seems a bit silly. Let's take a charitable view of Structuralism. Let's say patella mistracking doesn't manifest until the skeleton has finished growing, in the mid to late teens for most people. Fine. Then why isn't there an onslaught of cases of PFPS when people reach their early twenties, as their adult frame finishes growing and their badly tracking patella dooms them to a life of pain?

(3) Finally, here's the big problem with Structuralism, as Kelsey observes: Nobody has perfect biomechanics in the first place. Yet most of us do fine anyway.

Those are three reasons that logically (and instinctively) Structuralism doesn't make sense. But in the world of evidentiary medicine, musings and common sense alone don't constitute grounds for overturning a prevailing paradigm. In the medical world, physicians turn to clinical studies. So let's look at one.

This study ("Patellofemoral Joint Kinematics in Individuals with and without Patellofemoral Pain Syndrome", published in 2006) included three groups: 1. 20 people with PFPS who had clinical signs of patellar malalignment (as evidenced by tests performed during a physical exam) 2. 20 people with PFPS who had no clinical evidence of malalignment 3. 20 people with no knee problems.

An MRI captured images of their knees in various stages of being flexed, to note "patellar motion" as a function of the particular angle their knees were bent. So the MRI could see, for example, whether the kneecap was perfectly centered in the trochlear groove or sliding out to the right or to the left.

If Structuralism was the correct paradigm for understanding PFPS, what would we expect to find? Easy: that the patients with knee pain tended to have kneecaps with the worst tracking.

What the researchers actually found:

No differences in the overall pattern of patellar motion were observed among the groups ... It is clear from the data that an individual with patellofemoral pain syndrome cannot be distinguished from a control subject by examining patterns of spin, tilt, or lateral translation of the patella.

(If you're a Structuralist, that sound you just heard is the floor collapsing beneath you.)

In other words, if you just look at MRIs of how someone's patella tracks, you'll have no idea whether they have PFPS. Someone with a kneecap that tracks perfectly may have PFPS. Someone with no knee pain may have a patella that mistracks. The authors make the point more bluntly in a follow-up letter to the journal where the study was published: "Our findings add to the evidence that patellar mistracking is not a clinically significant factor for most individuals with patellofemoral joint pain."

So there you have it, a crumbling edifice called Structuralism, that your doctor and physical therapist are probably using right now to analyze why you have knee pain and how you should fix it. And is it any wonder that more people aren't healing? And is it any wonder that, in order for me to heal, I didn't need a visit from the angels from above, but rather a cold-eyed rejection of this whole Structuralist approach (I got better through a simple, long process: I strengthened my knees).

So when I ask, "Is your doctor (or physical therapist) a Structuralist?", it's not an idle question. The fate of your knees may hang in the balance.

So What Exactly Is Knee Cartilage Anyway?

In my last blog post, I dropped a rather long, important-sounding word: glycosaminoglycans.

These polysaccharides help to keep cartilage elastic and resilient. And, according to a Swedish study, their content increased in the knee joints of people who exercised.

Now I want to take a whirlwind tour through the make-up of cartilage, which I studied when I was battling knee pain, to put that fifty-cent word in some meaningful context.

First, hyaline articular cartilage (that’s the kind we're looking at, because it pads the ends of the bones that meet in the knee joint) is like a tough, rubbery, wet sponge. It’s four-fifths water (showing you why it’s important to stay hydrated).

Take out the water and what do you have left? There’s a tough, ropy protein called collagen that is found in higher concentrations on the surface than deeper in the tissue (to keep your cartilage tear-resistant where it matters most). Then you have molecules that weave around that collagen skeleton that are called proteoglycans.

Let’s look up close at those proteoglycans. They’re large molecules that consist of a protein core and many long-chain sugar molecules (if you want to visualize a proteoglycan, the easiest way is to imagine a bottle brush -- the spine of the brush is the protein core, and the many bristles are the sugar chains).

What are the sugar chains called? Go to the head of the class if you've already figured this out: glycosaminoglycans. (A quick aside: if you were to tear apart a glycosaminoglycan in your molecular toolshed, you would find it's composed of sugars such as glucosamine. Ah, so you now you know where that glucosamine you’re swallowing for your joint pain is supposed to be going! But unfortunately it never gets there, making glucosamine supplements useless -- more on that some other time).

So why should you give two hoots if your cartilage has a low content of glycosaminoglycans or a high content?

Well, when you subject your knee to load, that tough, rubbery tissue that is cartilage gets compressed. Remember, it’s not like a plate of metal, but rather a tough sponge with a high proportion of water. It needs some way to protect itself from intense forces being transmitted through the knee joint during everyday activities. Otherwise, your cartilage could get chewed up pretty fast.

Of course, there is that tough collagen, which helps. But the glycosaminoglycans also play a key role. On the atomic level, they carry a negative charge. So they repel each other when pushed closer together.

So let’s say you jump in the air to grab a Frisbee. When you land, knee cartilage has to absorb that load of your body hitting the ground. The pressure will expel water and synovial fluid from the tissue and push the glycosaminoglycans closer together. But as they’re pushed closer, that negative charge causes them to repel each other more strongly (you know how intense this pushback can be if you've ever tried to place the like poles on a pair of magnets in contact with each other).

So having plenty of glycosaminoglycans is critical for keeping the tissue resilient.

Knee cartilage 101! Important stuff to know.

You Can Strengthen Knee Joints: Scientific Proof

Check out this hard-luck group of 30 people:

They're all between 35 and 50 years old. They've all undergone surgery to repair a torn meniscus (a disk of cartilage in the knee joint). 87 percent of the group are aware of their knee problems at least monthly, and most have pain, stiffness and functional limitations.

Not a happy bunch.

They were subjects in a Swedish study to gauge the effect of exercise on something called "glycosaminoglycans" in knee cartilage. Big word, but important stuff: these polysaccharides are critical for keeping cartilage elastic and resilient.

The subjects were split into a control group (whose level of activity didn't increase over the course of the study) and a group that exercised. Each week, the exercisers were expected to go to at least three group classes, supervised by trained physical therapists.

The classes consisted of warm-ups, such as cycling, rope skipping and jogging on a trampoline. After that the subjects did "individually progressed weight-bearing strengthening exercises." These included such activities as repeatedly sitting down and standing up while holding a barbell bar (with no weight on it) and doing lunges while holding dumbbells.

After four months, researchers used a special enhanced MRI to peer into the joints of all the subjects.

The exercisers were found to have a significantly higher content of glycosaminoglycans than members of the control group. Also, they had improved scores on a special scale designed to assess pain, joint function and quality of life for people with troubled knees.

In Arthritis & Rheumatism (the November 2005 issue), the study's researchers, Ewa M. Roos and Leif Dahlberg, wrote that:

The changes imply that human cartilage responds to physiologic loading in a way similar to that exhibited by muscle and bone.

So weak or damaged cartilage can change, and for the better. Furthermore:

In a cartilage matrix with low [glycosaminoglycan] content, as in cartilage disease, insufficient viscoelasticity may cause cause progressive denaturation of collagen molecules, collagen loss, and subsequent development of [osteoarthritis].

Is having more glycosaminoglycans all that matters in beating knee pain? Of course not. But if you want robust knees that can withstand the rigors of daily living -- and that includes the "rigor" of sitting -- you need good, healthy shock absorbers, or cartilage. And good cartilage needs plenty of glycosaminoglycans to keep it stiff and functioning well.

So what the heck are glycosaminoglycans, and why do they sound so suspiciously like "glucosamine"? Ah, not a coincidence at all. I'll look at that next time, because answering the question properly requires a bit of a "fantastic voyage" into cartilage, on a cellular level.

(Note: observant readers will wonder about the exercises prescribed in the Swedish story. Aren't some of them "quad strengthening" -- which I've made no secret of disliking? Yes, but a few things: 1. They were done after a period of supervised warm up. 2. They were done under the supervision of physical therapists, who were monitoring each individual's form (a breakdown in form signals someone is doing an exercise beyond his or her capability). 3. They were specifically tailored to individual patients. 4. I'm not saying I endorse these particular exercises for knee pain sufferers; I'm just saying knees benefit from movement -- and here's evidence.)