Defining moments: Biomechanists run OA interventions through the research gauntlet

As the obesity epidemic continues to fuel the spread of osteoarthritis, researchers are working to ensure that interventions designed to limit the progression of OA actually serve that purpose.

Recent studies, including those presented at the annual meeting of the American Society of Biomechanics, found that gait retraining, valgus bracing, wedged orthoses, and variable-stiffness footwear all were associated with improvements in peak knee adduction moment-the outcomes measure currently in favor among OA researchers.

In the moment

Research trends may be less arbitrary in origin than, say, celebrity-inspired fashions, but they're no less contagious: once a line of investigation catches on, the number of similar studies being presented and published tends to multiply as quickly as if the idea had been endorsed by Paris Hilton herself. And if this current spate of studies is any indication, the concept of peak knee adduction moment as a measure of effectiveness for a knee OA intervention just might be the biomechanics mainstream's version of the tiny dog as fashion accessory.

To be sure, previous researchers have used knee adduction moment to assess OA interventions, from high tibial osteotomy to bracing. But until recently, knee adductor moment was but one of many variables thought to be representative of osteoarthritis progression, along with other factors such as mechanical alignment and joint space width.

The turning point for many researchers may have come in 2002, with the publication of a study from Fukui Medical University in Japan that not only confirmed knee adduction moment as a significant predictor of osteoarthritis progression but also found that knee adduction moment was only weakly correlated with baseline mechanical axis and joint space width. Researchers from the University of Tennessee in Knoxville have consistently used fluoroscopically determined condylar separation as a measure of offloading in OA braces, most recently in a much-publicized comparison of five commercially available devices presented at the 2005 meeting of the American Academy of Orthopaedic Surgeons (see "COX-2 inhibitor fallout offers opportunity for OA knee bracing," April 2005, page 53) and published in a supplement to the September 2005 issue of the Journal of Bone and Joint Surgery. But for many other OA investigators, knee adduction moment has become the outcomes measure of choice.

Sultans of sway

The most impressive reductions in knee adduction moment have come from an intervention that does not involve a device. Researchers from Stanford University, led by the husband-and-wife team of Lars and Anne Muendermann, found that training subjects to walk with precisely timed trunk sway decreased knee adduction moment in those patients by two-thirds. By comparison, the greatest reductions seen in patients with medial compartment OA, achieved with high tibial osteotomy, have been reported to be between 30% and 50%.

"We were very surprised to see such large reductions," said Lars Muendermann, PhD, a postdoctoral researcher in Stanford's Biomotion Laboratory, who presented the initial phase of the group's gait retraining work at the ASB meeting.

In that initial phase, the investigators analyzed the gait of 13 healthy adults who were instructed to perform an exaggerated mediolateral trunk sway without walking more slowly than they normally would. They found that both the magnitude and phasing of trunk sway contributed to decreases in peak knee adduction moment (relative to walking normally), with the optimum phasing occurring approximately 30 ms after heel strike. Sway magnitudes greater than 5 degrees were associated with knee adduction moment decreases of at least 50%, Muendermann said.

Once the optimal sway phasing parameters had been established, the researchers conducted another study of 19 subjects who had been trained in the phased-sway technique and found that peak knee adduction moment was 66% lower while sway-walking than while walking normally. Sway-walking was also associated with greater hip and knee abduction moments.

The changes in adduction and abduction moments were consistent with those seen in medial OA patients in a September 2005 study by the same group, which was published in Arthritis & Rheumatism, suggesting that sway-walking mimics the compensatory gait strategies that patients with medial compartment OA adopt. However, the healthy sway-walking subjects did not demonstrate the same decreased flexion angles at heel strike and increased ground lateral reaction forces that had characterized the OA patients' gait in the previous study-which may be related to decreased hip adductor strength often seen in patients with OA, said Anne Muendermann, PhD, a research associate in the same department, who presented the second phase of the group's results at the ASB meeting.

The training is relatively simple, requiring fewer than three practice trials. Questions remain regarding how long an individual with OA might be able to maintain a large magnitude of trunk sway while walking, but the Muendermanns emphasized that even smaller amounts of sway could reduce knee adduction moment as effectively as an orthotic device (for which previously reported reductions have ranged from 8% to 13%).

"The reductions we showed were huge. To achieve the reductions seen with other interventions, you may not have to increase trunk sway as much as we did," Anne Muendermann said.

Sole patrol

As effective as walking with trunk sway might be for decreasing peak knee adduction moment, more self-conscious patients may be averse to trying it, at least in public. But the latest in a series of studies by a different group of Stanford researchers (sans Muendermanns) suggests that a variable-stiffness shoe may be a less conspicuous option.

The experimental shoe, which is being manufactured by Nike for the purposes of the current research and may ultimately be marketed by the athletic shoe giant, features a sole that is 2.5 times more rigid on the lateral side than on the medial side. On the strength of previous research in which the experimental shoe was associated with decreased peak knee adduction moments in healthy subjects (see "Footwear modifications reduce load on knee, researchers find," October 2004, page 15), the investigators analyzed its effect on gait in 26 subjects with symptoms of unilateral medial compartment knee OA.

They found that mean peak knee adduction moment was significantly lower while the subjects wore the experimental shoe than when they wore a control shoe (with a homogeneous sole stiffness equal to that of the medial side of the test shoe) or their own personal shoes. The differences between conditions were significant at three different walking speeds, ranging from 1 m/s to 1.7 m/s.

However, the amount of reduction seen in the knee adduction moment varied within the study population, and not all subjects experienced a decrease, said Jennifer Erhart, a graduate student in the department of mechanical engineering, who presented her group's findings at the ASB meeting. For example, at normal walking speed, the change in knee adduction moment ranged from a 20% decrease in one subject to a 7% increase in another; four patients experienced increases.

The results suggest that patients with greater baseline knee adductor moments may benefit most from this type of intervention.

"Individuals with a high knee adductor moment have greater reductions with the intervention shoe," Erhart said. "Gait testing is a good screening tool to identify responders versus nonresponders."

State of brace

Intersubject variability was also an issue for researchers from Raboud University in Nijmegen, the Netherlands, in assessing the effectiveness of the Bauerfeind SofTec OA brace in 15 patients with medial knee OA. After six weeks of brace wear, the mean peak knee adduction moment was lower with the brace than without, but the decrease did not reach statistical significance and was not seen in all patients. The investigators did, however, find that patients with greater degrees of varus deformity at baseline were significantly more likely to experience a reduction of peak adduction moment with brace wear.

In addition, the Dutch researchers reported significant improvements in pain levels and knee function following the six weeks of brace wear in all patients. Mean pain level improved from 6.8 to 4.7 on a visual analog scale, and function improved from 50.1 to 63 on the Western Ontario and McMaster Universities (WOMAC) osteoarthritis index. The findings were published online in September by Gait & Posture.

Angling for relief

Laterally wedged orthoses have received considerable attention in recent years as a potential intervention for patients with medial compartment knee OA (see "Orthoses wedge their way into OA treatment," August 2004, page 49), but findings from the University of Delaware suggest that medially wedged orthoses may be even more effective for the smaller subset of knee OA patients with lateral compartment disease.

Researchers studied 10 patients with lateral compartment OA, each of whom was prescribed a noncustom orthosis with a medial wedge that was individualized (at an angle between 5 degrees and 15 degrees ) so as to cause the least amount of pain during a lateral step-down test; the average amount of wedging was 8.5 degrees . When patients walked at a self-selected speed, the peak knee adduction moment was 26.1% higher while wearing the wedged orthoses than while wearing nonwedged control orthoses. These results compare favorably to adduction moment increases of 6% to 8% previously reported in studies of laterally wedged orthoses for patients with medial compartment OA.

"Lateral osteoarthritis may be more responsive to wedging than medial osteoarthritis," said Joaquin Barrios, DPT, a graduate student in the biomechanics and movement sciences program at the university, who presented the group's results at the ASB meeting.

Weight and gait

Of course, even better than using interventions to slow the progression of OA would be to prevent the disease before it develops. Research from Stanford identifying mechanical similarities between patients with OA and obese individuals without OA suggests that similar interventions could be effective in both populations.

Anne Muendermann and colleagues used magnetic resonance imaging to assess articular cartilage thickness in six overweight (average body mass index of 29.2 kg/m2) and six age- and gender-matched normal-weight (BMI of 22.3 kg/m2) subjects, then incorporated those data into a computer model along with kinematic and kinetic measures collected while the subjects walked at a self-selected speed.

They found that the response of cartilage during gait in the overweight subjects was similar to previous findings in patients with OA. The overweight subjects had a negative correlation between medial/lateral cartilage thickness ratio and adduction moment, which was more pronounced in the anterior region than the posterior region, while the control subjects had a positive correlation. In addition, three of six obese subjects had knee hyperextension at heel strike, which suggests that in those subjects, the anterior aspect of the tibia is more heavily loaded at heel strike, and that the cartilage thickness ratio reflects this change in loading pattern.

The findings not only suggest a possible mechanism by which obesity contributes to knee osteoarthritis; they also suggest that there may be a prophylactic benefit to addressing gait patterns in obese patients that are suggestive of OA even in the absence of symptoms.

"Load-altering interventions may be effective in decreasing the risk of OA in obese individuals," Muendermann said.

Jordana Bieze Foster is a freelance writer based in Massachusetts and former editor of BioMechanics.

---