INDIANAPOLIS, Indiana—Quadriceps weakness is a risk factor for knee osteoarthritis (OA), and quadriceps strength training (ST) has been associated with symptomatic improvement, but Alan E. Mikesky, PhD, and colleagues report that regular quadriceps ST was not significantly better than range of motion (ROM) exercises at improving strength or slowing joint space narrowing (JSN) in patients with knee OA.¹  The investigators suspect that any beneficial effect of exercise was diluted over the 30 months of follow-up because of attrition and less than optimal adherence to the protocol.

"It's probably safe to speculate that a more intensive exercise protocol may have produced better evidence of the structure modifying potential of exercise in knee OA. This could have been achieved by either shortening the trial, thus permitting closer follow-up and encouragement to subjects, and/or making the exercise protocol more demanding. Our trial lasted 30 months because that was the minimum duration of follow-up considered necessary for a difference between treatment groups in radiographic outcomes to accrue. Most studies of the effects of exercise on knee pain have lasted only 2 to 3 months. To intensify the exercise protocol would have made attrition and adherence even larger problems than they already were. Based on our experience with a doxycycline trial,² we now think that an effective treatment can show up in x-rays in as little as 16 months. So shorter studies of more intensive exercise programs are probably feasible, " coauthor Steven A. Mazzuca, PhD, told CIAOMed.

Subjects with and without knee OA randomized to strength vs ROM training

The research team from Indiana University, Indianapolis, randomized adults (mean age 69 years) to thrice-weekly regimens of either ST (n = 113) or ROM (n = 108) exercises and then measured strength and JSN at baseline and again after 30 months. ST exercises included leg presses, leg curls, seated chest presses, and seated back rows during an initial 12-week period at an exercise facility, then wall squats, standing leg curls, wall pushups, and seated rows (using elastic bands to provide resistance) after switching to at-home training. ROM exercises included flexibility exercises targeted at the neck, shoulders, trunk, elbows, wrist, hips, knee, and ankles.

Clinical outcomes were assessed using the Western Ontario and McMaster Universities OA Index (WOMACTM), the Medical Outcome Study Short Form General Health Survey (SF-36), and the Center for Epidemiologic Studies Depression Scale (CES-D). Isokinetic strength was assessed using a dynamometer. Radiographic outcomes were assessed using fluoroscopically standardized, semiflexed anterior-posterior (AP) views of each knee.

At 12 months after the beginning of intervention, quadriceps isotonic strength was not significantly different in ST vs ROM patients, but isotonic hamstring was somewhat better in the ST subjects (women: +6.3% vs -0.7%; men: +11.8% vs +8.5%, P = .021), but this difference between treatment groups had vanished by the time of the 18-, 24-, and 30-month assessments. Quadriceps isokinetic strength dropped slightly (but not significantly) less in the ST group (P =.090).

Patients were stratified before randomization to produce treatment groups equivalent at baseline with respect to severity of knee OA and quadriceps strength. When the investigators examined the effects of ST vs ROM exercises on patients with vs without knee OA at baseline, they found that the patients with knee OA lost less isokinetic quadriceps strength with ST than with ROM exercises (mean decrease -19.7 vs -14.6 Nm of extension at 60º/second, P = .041).

"Even though strength training did not result in an increase in isokinetic quadriceps strength in the present study, it slowed the rate of loss of isokinetic hamstring strength over 30 months in comparison with ROM exercises," the investigators report.

Unexpected effect on JSN in normal knees seen

There was no significant difference on loss of joint space width (JSW) between the treatment groups, but the 30-month follow-up unexpectedly shows that subjects with normal knees at baseline were significantly more likely to have loss of JSW greater than the .50-mm margin of error in the ST group than in the ROM group (34% vs 19%, P = .038).

"This observation is difficult to explain," Dr. Mazzuca said. "It could be that there were some patients in whom strength training as we studied it (ie, leg extensions and leg curls) resulted in damage to either the knee cartilage or meniscus (which can also appear as early joint space narrowing in a radiograph). However, the normal knees that exhibited JSN were not exercised more frequently than knees with no radiographic changes. Nor did they occur in subjects who achieved relatively large increases in quads or hamstring strength. We can only suggest that this observation be confirmed or disconfirmed in future exercise trials."

ST had no significant effect on changes in knee pain or on the WOMAC or CES-D measures. Subjects with knee OA randomized to ROM had significantly greater decreases in emotional health status, as measured by the SF-36, than those randomized to ST (-1.6 vs -0.4, P = .042).

Study seen as important first step in defining effects of exercise in knee OA

Leena Sharma, MD, of Northwestern University in Chicago, has also studied the relationship between quadriceps strength and knee OA.³ Dr. Sharma told CIAOMed that this is an important paper because "almost no studies have undertaken examination of the effect of an exercise program on joint structure (in the current era of joint structure assessment) in OA joints."

"This particular ST program did not protect against incident knee OA or OA progression," Dr. Sharma commented. "Since there was a decline in strength over the study, even in those in the ST group, the results beg the question: If a given ST program increases or at least maintains strength in knee muscles, might such a program protect against incident OA or progression?"

Dr. Sharma suggested that key questions for future studies include what sort of strengthening program would be required to increase or at least maintain strength in knee OA, and what biomechanical effect such exercises might have on diseased or damaged joints.

"This study suggests that the ST that we recommend in the clinic is unlikely to help our patients to increase their strength. On the other hand, ST that is too aggressive could accelerate OA progression, especially in a vulnerable (eg, arthritic, malaligned, lax) joint. Is there a window here, ie, ST that improves strength in OA knees but is not too aggressive for an already damaged joint?  And, how tiny is this window? If tiny, how safe is it to put forth a more aggressive ST in the public health arena or in the clinic? In theory, a specific exercise could have a protective effect or a deleterious effect on a vulnerable joint. This study implies that more effort needs to be devoted to refining an exercise program that not only helps to preserve muscle strength, but also to enhance task-specific muscle function. The right exercise could indeed protect against worsening of OA. However, the current results suggest that usual strengthening is not an impressive strategy to reduce risk of incident or progressive knee OA.  The current study illustrates how little we know!" Dr. Sharma said.

References

1. Mikesky AE, Mazzuca SA, Brandt KD, et al. Effects of strength training on the incidence and progression of knee osteoarthritis. Arthritis Care & Research. 2006;55:690-693.
2. Brandt KD, Mazzuca SA, Katz BP, et al. Effects of doxycycline on progression of osteoarthritis: results of a randomized, placebo-controlled, double-blind trial. Arthritis Rheum. 2005;52:2015-2025.
3. Sharma L, Dunlop DD, Cahue S, et al. Quadriceps strength and osteoarthritis progression in malaligned and lax knees. Ann Intern Med. 2003;138:613-619.