A Randomized Clinical Trial of Three Active Therapies for Chronic Low Back Pain (1999 Volvo Award Winner in Clinical Studies).

Mannion AF, Muntener M, Taimela S, et al (Schulthess Clinic, Zurich, Switzerland; Institute of Anatomy, University of Zurich-Irchel, Zurich, Switzerland; DBC International, Vantaa, Finland), Spine. 1999;24:2435-2448.

This randomized clinical trial evaluated the efficacy of 3 therapy programs--modern active physical therapy, muscle reconditioning on training devices, and low-impact aerobics--commonly used to manage patients with chronic low back pain (LBP). Outcomes were assessed in terms of changes in patient reports of pain intensity, pain frequency, and disability and in terms of functional capacity (eg, lumbar range of motion). Psychosocial factors were also evaluated, primarily using questionnaires, to examine their possible role in successful therapy outcomes. Pre- and post-therapy objective and questionnaire measures were evaluated and similar questionnaires were administered 6 months following completion of therapy.

One hundred forty-eight patients were recruited for the study; 137 questionnaires were available for analysis at all 3 time points. The inclusion criteria were as follows: less than 65 years of age, pain of more than 3 months' duration, pain causing absence from work or needing medical attention, ability to travel to the hospital, willingness to adhere to random treatment assignment, ability to understand German or English, and ability to perform a simple pre-inclusion test designed to ensure a minimal ability to complete the study.

Subjects were randomly assigned to 1 of the 3 treatment groups: (1) physical therapy, (2) training with devices, and (3) aerobics and stretching. Subjects were required to attend therapy twice a week for a total of 3 months.

Subjects in the physical therapy group received half-hour individual sessions consisting of isometric exercises and exercises using Therabands and general strength training devices. Subjects were given advice on home exercises and given instruction on ergonomic principles.

The muscle reconditioning group, which used exercise devices, used the David Back Clinic. Subjects attended in groups of 2 or 3, with each session lasting 1 hour. The exercises applied isoinertial resistance to the lumbar spine in the 3 cardinal planes. Relative progression of loading was applied throughout the rehabilitation period. Sessions were preceded by a 10-minute aerobic warm-up.

The subjects in the aerobics and stretching group attended 1-hour small group classes with a maximum of 12 subjects per class. Classes consisted of a 21-minute warm-up period involving whole body stretching and low impact aerobic exercises, followed by 30 minutes of specific exercises directed at the trunk and leg muscles. The last 15 minutes consisted of cool-down and relaxation exercises.

For functional assessment, the authors measured lumbar range of motion using the CA-6000 Spine Motion Analyzer, an electromechanical device that measures spinal movements in flexion and extension, lateral bending, and rotation. Questionnaires were administered to measure pain intensity, frequency, and disability; fear-avoidance beliefs and coping strategies.

Parametric statistics were used for the analyses. Comparability of groups at baseline was assessed using an ANOVA. Changes in variables for the 3 groups at entry, after therapy, and at 6 months were assessed using a 2-factor ANOVA with repeated measures.

The 3 treatment groups did not differ significantly in their adherence rates. There was no significant difference among the groups in baseline sociodemographic and clinical characteristics. Overall, there was a significant reduction in the greatest and the average pain scores recorded posttherapy as compared width pretherapy values; however, there were no significant differences among the 3 groups in the extent of reduction. Significant reductions were found from baseline to post-treatment for all 3 groups; again, the groups did not differ significantly from each other. Disability, according to self-reports using the Roland-Morris Back Pain Questionnaire, showed a significant decrease from baseline to posttherapy; again, no significant difference existed between the exercise groups. At the 6-month follow-up, the training devices and aerobics groups showed a further decline in Roland-Morris scores; however, the physical therapy group's scores actually increased.

Regarding lumbar spinal mobility, the subjects as a whole showed a small but significant posttherapy increase in sagittal plane range of motion, but no difference existed among the groups. The aerobics group showed the best improvement in ranges for lateral bending and rotation, the devices group had a moderate increase, and the physical therapy group exhibited little change. The extent of change in the aerobics and devices groups compared with the physical therapy group were significant. Significant reductions were observed in coping strategies for the whole subject group, including fear-avoidance beliefs about physical activity, praying and hoping, "catastrophizing," and pain behavior. With the exception of disability and fear-avoidance beliefs about activity for the physical therapy group, these effects were maintained through the 6-month follow-up.

The authors concluded that the 3 treatments administered to patients with chronic LBP proved to be equally effective in their ability to reduce pain intensity, frequency, and disability. For the most part, the results were maintained over the 6-month follow-up period. The authors stated that the general lack of treatment specificity may suggest that the therapies were beneficial, not through the reversal of physical weakness, but rather through some "central effect" that may have involved an adjustment of perception of pain and disability. Direct costs associated with the tests were 3 times as great for physical therapy and 4 times as great for the training devices as for aerobics.

Bill Barnes, PT Lincoln Physical Therapy Associates and Work Adjustment Center Lincoln, Neb