Why That Study About Quad Strength Doesn’t Mean What You Think It Does

Last week I told The Odd Story of the HQ ratio, based mainly on this study.

This week we look at a different study (in the spirit of continuity, it made a cameo appearance last week).

What if you were watching TV and an announcer broke in and said:

News flash! The Pope is Catholic! Details at eleven.

Chances are good, you wouldn’t bother tuning in for that newscast.

A certain family of knee studies is pretty much the analogue of that "news." Here’s a specific example (and today’s topic): “Quadriceps Weakness Predicts Risk for Knee Joint Space Narrowing of Women in the MOST cohort.” (Knee joint space narrowing is significant because it signifies a loss of cartilage and accompanies the progression of osteoarthritis.)

This time, let’s start with the conclusion and work backward. That’s because you’ll probably think it means something that it doesn’t. And that part is important to get right from the outset.

Here’s the wording, which is fairly simple and straightforward (note: I’ve snipped an introductory phrase -- which I’ll return to, not to worry, for any careful readers):

Quadriceps weakness was associated with increased risk for tibiofemoral and whole knee joint-space narrowing.

Here’s how you probably translate that:

A patient who has knee pain should strengthen the quadriceps to prevent tibiofemoral and whole knee joint-space narrowing.

In a moment, we’ll look at why that’s a potentially dangerous misinterpretation. Before that, what is the family of knee studies that this belongs to?

Others that resemble it sound like this: Quadriceps Weakness Predicts Higher Levels of Knee Pain Among Osteoarthritis Patients. Quadriceps Weakness Predicts Worse Outcomes After Knee Surgery. Quadriceps Weakness Predicts Increase in Knee Cartilage Lesions. Etc.

So if quadriceps strength protects against all this bad stuff, why shouldn’t someone with chronic knee pain who has some of this bad stuff already, but doesn’t want it to get worse, strengthen the quads? What makes that a potentially dangerous misinterpretation?

The problem is, such studies don’t support that interpretation, if you just look at the raw data and the basic conclusions.

For example, the MOST cohort study essentially says this: If you start with two people, Jane with a quad strength of 80 “units” (I’m using this word as a shorthand; feel free to substitute in your favorite measurement) and Jill 40 units, Jill will be at increased risk of knee joint space narrowing.

The study is silent on whether, if both women have knee pain, Jane should try to strengthen her quads to say 120 units and Jill should try to reach 60 units, to lower their risk. Assuming the study implies this is potentially dangerous. Maybe quad strengthening exercises would damage their knees further!

(Brief digression: It’s probably time to capsulize the study. Here goes: a total of 3,856 knees were tracked over a 30-month period; subjects were 50 to 79 years old; all had knee osteoarthritis or risk factors for it.)

That greater quad strength, to some degree, protects knees seems like a yawner of a conclusion, like “discovering” the Pope is Catholic. When you suffer from constant pain though, the question isn’t do strong quads help -- it’s how do you get there, and more importantly, how do you heal your bad knees that hurt all the time. Unfortunately, there’s no such thing as a quad strength fairy who, with one tap of her magic wand, can bulk up your weak quads.

In the absence of such a fairy, you have to build up the muscles the old-fashioned way, through exercise. While such strengthening isn’t impossible when you have a tender, easily irritated joint, it’s much harder and much trickier. The worse your knee pain, the more difficult the task will be.

But listen: You’re chasing the wrong objective anyway! Physical therapist Doug Kelsey nailed this one: “Having stronger muscles is helpful but weak muscles are not the primary problem.”

It’s your bad knees that need fixing.

Two more things:

(1) What was the introductory phrase I left out?

The full sentence of the conclusion reads (my bold):

In women but not in men, quadriceps weakness was associated with increased risk for tibiofemoral and whole knee joint-space narrowing.

I omitted the women vs. men part because (1) I didn’t want to have to deal with that confusing aspect of the study yet (2) The reason the researchers cite for the difference -- namely, that men have higher strength and this may provide a greater reserve -- strikes me as being sensible. In other words, if the quad strength differences were large enough, you’d see the same effect in the male population too.

Still, including this omission only fortifies my side of the argument.

If quad strength was found to protect against joint space narrowing in women, but not in men, you might surmise that quad strength may not have that much of a protective effect after all. In fact, the MOST cohort researchers cite another study that “found no significant differences in mean quadriceps strength between women with and without tibiofemoral osteoarthritis worsening.”

So while I think it’s obvious that strong quads help shield knees from harmful impacts that lead to everything from knee pain to bone spurs, the advantage they confer may not be that great.

(2) The introduction for the MOST cohort study makes a telling confusion.

While drawing comparisons, the researchers cite two other studies. Pay close attention here. In one, “greater quadriceps strength was protective against lateral patellofemoral cartilage loss.” In the other, “strength training did not protect against tibiofemoral joint space narrowing over 30 months in subjects with definite knee osteoarthritis at baseline.” Why the discordant results, they wonder.

Did you notice that one of these three things is not like the others?

Two studies (I’m including the MOST cohort study here) basically say, “If you have stronger quads, you have a lesser risk of bad stuff happening in your knee joint.” But the proposition for the third study begins, “If you attempt to strengthen your quads ...”

Why is this such a critical distinction? Because this gets to the heart of the matter for chronic knee pain sufferers! They’re not spending a lot of time musing, “Hmm, I wonder whether, if my quads are stronger than Jim’s, and he has knee pain too, whose joints will worsen more in a couple of years?” Rather, they’re thinking, “What the heck should I do to protect and hopefully heal my knees?”

That makes the third study the most relevant, hands down. It’s not chasing the obvious and trivial, but setting out to answer the all-important question, “Will strengthening my quads help my knees?”

And its conclusion, remember, was that strength training didn’t help. (Actually, it’s worse than that. Here’s the study abstract and an interpretation of the results. Strength training actually contributed to joint space narrowing among subjects who had no X-ray-indicated osteoarthritis at the outset. One note to keep in mind though: participants were older, averaging 69 years of age.)

While having strong quads is unquestionably a good thing, getting to strong quads -- when you already have knee problems -- may not be so simple or so beneficial. Why? As I’ve already noted, strength training may overwhelm weak knees, hastening cartilage loss and osteoarthritic processes.

Coda: Once again, an objection from my imaginary critic.

All this is okay, but woefully incomplete. What do you ignore studies that do show quad strengthening is good for those with bad knees?

Obviously, only so much can be covered in a single post (even as long as this one turned out to be). But the reply to that objection is really here, in my big picture essay on evaluating treatments for patellofemoral pain syndrome.

I don’t think quad strengthening is necessarily a failure as an approach. Not at all. If your joints are robust enough, if the exercises are smart and not too strenuous, it may work fine. The problems, I think, arise with weaker knees.

So I return to the idea of okay-better-best. Quad strengthening is, in my estimation, somewhere between “okay” and “better” on the spectrum of knee pain treatments. But don’t you want what’s best? That, I’m convinced, is strengthening the joint first. After all, the joint is what’s weak.

The Odd Tale of the HQ Ratio

My latest post lashing out at physical therapy for obsessing over “quad strengthening” (to treat knee pain) prompted this comment:

There is a lot of evidence that weak quadriceps lead to increased knee pain. Ditto for *tight* quadriceps, which are often caused by over-exertion of weak quadriceps. Below are some sources from scientific journals.

There are a lot of causes for knee pain. For chrondomalacia patella, the root cause is often a misaligned patella, improperly tracking over its surrounding cartilage. I argue, based off of my own experience and research, that a first step towards addressing this misalignment should be achieving an optimal balance of strength between the hamstrings and quadriceps.

A couple of links to studies followed. So I thought, “What the heck. Why not see where this leads?” What I discovered was so remarkably unremarkable that I felt compelled to share it with everyone.

Today’s installment: The Odd Tale of the HQ Ratio.

Way back in 1988, when a first-class stamp cost 24 cents and the drug Prozac was a brand-new treatment for depression, a Finnish researcher by the name of Pekka Kannus published a study: “Ratio of Hamstring to Quadriceps Femoris Muscles’ Strength in the Anterior Cruciate Ligament Insufficient Knee: Relationship to Long-Term Recovery.”

Okay, the title is a bit dense.

The “Hamstring” and “Quadriceps” part refers to a strength ratio between the muscles, known as the (what else?) HQ ratio. (Quick aside: the deeper I get into structuralist hypotheses for chronic knee pain, the more their quest begins to look like a decades-long Easter egg hunt for the structural deficiency that explains away much of the problem.)

Kannus decided to investigate whether an optimal balance of strength exists between the hamstring muscles and quadriceps femoris. His study included 41 subjects, from top-level athletes to nonathletes, who had an anterior cruciate ligament injury to one of their knees.

An average of 8.2 years after the injury, he performed various measures of strength of their hamstring and quad muscles. He also noted the condition of their bad knee, both functionally (Do they walk with a limp? Do they have problems climbing stairs?) and subjectively (Do they often feel knee pain?)

His objective was simple:

The main hypothesis was that the patients with an optimal absolute HQ ratio (50% - 80%) have better outcomes than other patients.

So, to step back for a moment, here was the structuralist paradigm for the HQ ratio at the time:

There is an ideal balance between the strength of a patient’s hamstring and quad muscles, which falls within a range, where the hamstring should be 50% to 80% as strong as the quadriceps femoris. When the ratio lies outside this range, the patient will have more knee problems.

And (drumroll, please), the study found ...

Nothing of the kind.

First, Kannus discovered that the HQ ratio varies a lot naturally, even for healthy knees. Second, and to the point, he learned “absolute HQ ratios do not correlate with long-term outcomes.” So, whether your HQ ratios are 65% (in the middle of the “ideal” range) or 30% or 150%, this has no bearing on how your knee fares. He ends on a rather sheepish note: “No general recommendation can be made about an optimal absolute HQ ratio as a goal for rehabilitation.”

But wait, there is something else.

Kannus did find that the HQ ratio of a subject’s bad knee was consistently higher than that of his good knee (meaning the hamstring muscles were stronger relative to the quads). So even though he couldn’t support the contention that an HQ of 65% in one individual is better than a 30% or 150% ratio in another, he could assert that the appropriate HQ for the injured-knee leg may be that of the same person’s uninjured-knee leg.

Which is ...

A big nothingburger in the Great Deli of Knee Pain Studies.

Why is this conclusion all air, no meat? Kannus himself tells us in his write-up of the research (my bold):

“The higher HQ ratios on the injured side ... resulted from the weak quadriceps femoris muscle ... computerized tomographic studies of cross-sectional areas of atrophied thigh muscles have demonstrated that atrophy of muscle tissue is greater in the quadriceps femoris muscle than in the hamstring muscles after knee ligament injury.”

Got that? After such an injury, this quad muscle weakens faster than the hamstring does. So what would you expect to find in subjects? Hamstring muscles that, relative to the quadriceps, are stronger in the injured knee than in the uninjured one.

Now, did the HQ ratio die an ignominious death after Kannus failed to prove his hypothesis?

Of course not.

Check out this study from 2010, more than 20 years later! It had many subjects (almost 2,000). So sample size wasn’t an issue, unlike in the much smaller Kannus study. Among other things, researchers looked at the relationship between the HQ ratio and knee joint-space narrowing in a population that either had either knee osteoarthritis or known risk factors for it.

The study found (another ironic drumroll, please) ...

No evidence that the HQ ratio is linked to the risk of narrowing space in the knee joint.

Hmm. How much do you bet someone will still be studying the HQ ratio 20 years from now? :)

Coda: Now for an anticipated objection:

This study isn’t a “nothingburger” -- far from it! Kannus made a valuable finding! He showed us that the goal of physical therapy is to restore the HQ ratio of the injured leg to that of the non-injured one. This gives us an identifiable, concrete objective.

So what is wrong with using the HQ ratio as a relative, not absolute, guide? For starters, here are two things:

(1) There are a number of muscle ratios you could focus on, and a number of reasons you could come up with for doing so -- so why this one?

Once your knee is bad and you can’t use your leg properly, many muscle ratios probably get out of whack. What if the hip muscles become relatively stronger than the quadriceps femoris? Why not focus on correcting that ratio instead? I’m sure you can concoct a theory why that would make sense. Here’s one off the top of my head: The muscle imbalance draws the kneecap off its proper tracking pattern, causing a misalignment.

But then again, a creative structuralist might conclude it’s not the hip-to-quad strength ratio that matters most. Maybe it’s the calf-to-quad. Or hamstring-to-gluteus. Or gluteus-to-quad. Or hamstring-to-hip.

The possibilities are endless once you go down the structuralist rabbit hole.

(2) Cause and effect! It’s more likely that bad knees cause a distorted HQ ratio, rather than the other way around. So an out-of-whack HQ ratio means little; it’s just a symptom of having an injured knee.

Here’s an illustration to drive home the point: Suppose I discover something I call the Leg Irregular Muscle Propulsion ratio. This proportion is derived by measuring a subject’s fluidity of motion during ordinary walking, compared with a perfectly smooth gait. The lower the LIMP ratio, the closer a leg is to walking normally.

Further, let’s say I observe that, for a healthy individual, the LIMP ratios of their right and left legs are very similar. But among subjects with knee pain, the LIMP ratio is higher in the injured leg.

Ostensibly, as a physical therapist, I now would have a goal: Restore the bad LIMP ratio to that of the good knee. Teach the patient to walk more smoothly. But that ignores the fundamental reason he doesn’t walk more smoothly in the first place. He has knee pain!

Kind of silly, huh? If you want to fix the LIMP, why not focus on improving what’s causing the LIMP, the bad knee?

Physically Active Children May Have Stronger Knees as Adults

Do you have children?

I have one, age three and a half. The way she uses (and abuses) her little body never ceases to amaze me. I sit and watch, without comment, as she takes part in marathon sessions of charging at the couch and slamming into the cushions.

One time she advanced across the rug, whipping her head violently back and forth, trying to induce dizziness. I recall thinking, “My God, my God -- your neck!”

Is she active? Do bears you-know-what in the woods?

If you have a similarly high-energy toddler, or older child, there’s good news. All that running and jumping about may be laying the foundation for stronger knees later in life.

That, anyway, is what a study of 298 Australians found. Subjects had their strength and fitness tested as children, then researchers did MRIs of their knees 25 years later. The findings, according to this summary article:

Childhood physical activity, including physical work capacity, leg and hand muscle strength, sit-ups, and long and short runs had a significant, consistent association with greater tibial bone area. In addition, higher childhood physical work capacity measures were associated with greater tibial cartilage area.

(The tibia, by the way, is the bone in your lower leg that meets the larger femur in the knee joint.)

Why this beneficial effect? One investigator in the study, Graeme Jones, speculates that it’s as simple as “bone area gets larger to cope with the extra demands put on it by higher levels of physical activity, and then this leads to more cartilage, as cartilage covers the surface of bone.”

Does this mean though that, if you’re an adult who spent his childhood watching cartoons instead of playing kickball, you’re out of luck now?

That seems highly unlikely to me. One thing that impressed me, after doing a lot of reading about knee cartilage while working toward my own recovery, was how dynamic the tissue is. It’s constantly changing -- sometimes for the worse, true, but often for the better too.

It has been shown that the knee cartilage of triathletes is thicker than normal. Is that because they were active as children? I very much doubt that. More probably, it’s because their present-day physical activity is busily modifying their cartilage in a lot of good ways.

Why Is My Story Relevant If Your Knee Pain Differs From Mine?

Sometimes I get a reaction like this to my book:

Interesting story, but it doesn’t really apply to me. The author hurt his knees cycling, while my knee pain is due to nothing related to sports/no identifiable cause/insert something else here. Also, the author struggled with knee pain for about a year, while I’ve had problems for 2 years/5 years/insert some other length of time here. And the author’s symptoms were burning sensations around the kneecaps and difficulty sitting while mine are problems walking/swelling/insert some other symptoms here.

First, don’t misconstrue what I’m about to say as a marketing pitch (Step right up and buy this book! A panacea for all your knee woes, of any kind, of any severity!) Clearly my story isn’t relevant for certain types of knee problems (e.g., a torn ligament). It’s also not useful for a certain kind of person who mainly wants details on specific exercises to do or who seeks an “other” directed solution to his problems.

But for those suffering from that mysterious condition called patellofemoral pain syndrome, or who have been told they have chondromalacia, or who have chronic knee achiness, diffuse burning, pain of unclear origin -- and for those who believe they themselves must be very involved in fixing their problems -- I think it can be of great benefit. Here’s why:

(1) “All happy knees are alike; all unhappy knees are unhappy in their own way” isn’t quite true. (Apologies to Tolstoy.) Which is to say, I think there are a lot of similarities among knees afflicted with patellofemoral pain syndrome, chondromalacia or osteoarthritis.

For example, many problems will relate to damaged cartilage, and they’ll be frustrating to deal with, because cartilage has no nerve endings. This was a major frustration that I faced certainly.

The upshot: What unites us may be more significant than what separates us.

(2) Some good practices are just universal, regardless of your particular symptom set -- for example, the importance of movement or weight loss (or weight control at least, if you’re thin already).

It’s hard for me to imagine a chronic knee pain patient for whom these good practices aren’t relevant. In the book (and on this blog), I’ve supplied more context -- from medical studies, from my own experience, from examples drawing on common sense -- for why they matter so much.

(3) You need the right kind of optimism to succeed, if you’re going to triumph over knee pain.

More specifically, you need optimism grounded in reality. It’s not the optimism of the terminal cancer patient who, with a 1 in a million chance of a cure, stubbornly vows, “I’m going to get better.” It’s the optimism of a reasonable, intelligent person who thinks he or she can get better because there’s plenty of evidence supporting that belief.

The book is largely about providing the proper context for such optimism. You can get better because I did is part of the message. But you can get better because defects in bad knees can heal -- that’s a bigger, universal message worth shouting from the rooftops. And scientific evidence for that assertion is in the the book (and on this blog).

I believe my story does have a wider relevance than here’s-one-thing-that-worked-for-one-guy. My hope is that others will have their own success stories, following similar principles, and together we can force doctors and physical therapists to rethink how they treat chronic knee pain.