Introduction: Active treatments are advocated for the management of non-specific chronic low back pain (CLBP), although few studies have documented the relative efficacy of differing types of programme. A number of the available treatments comprise exercise routines on specially designed training machines, which are ostensibly better disposed to reverse the compromised trunk muscle function displayed by these patients than are 'free exercise' programmes. However, in using these muscle-training programmes, the physiological or anatomical adaptations that might account for the improved performance are rarely investigated, let alone identified. This is an important issue, because if the 'newly-acquired strength' is mostly specific to performance on the devices on which the patient has trained and been tested, and reflects the skill in executing these particular tasks, this will not necessarily assist the patient during performance of his/her everyday activities. The aims of the present study were (1) to quantify the changes in back muscle performance in chronic LBP patients following 3 months active therapy, and (2) to analyse the corresponding changes in activation and cross-sectional area of the paraspinal muscles.
Methods: 148 individuals (57% women) with CLBP (age 45.0+/-10.0 years; duration of LBP 10.9+/-9.5 years) were randomised to a treatment which they attended 2/week for 3 months: active physiotherapy, muscle reconditioning on training devices, or low-impact aerobics. Pre- and post-therapy, assessments were made of isometric trunk muscle strength in each plane of movement and of erector spinae activation (using surface electromyography) during back extension. In a sub-group of 56 patients, the cross-sectional area of the paravertebral muscles was determined using magnetic resonance imaging (MRI). In all patients, self-rated pain intensity, pain frequency and disability were assessed before and after therapy.
Results: 132/148 patients completed the therapy. Isometric strength in each movement plane increased significantly in all groups post-therapy. Apart from trunk extension, the changes were significantly greater in the devices group than in the other two groups (Fig 1). Activation of the paraspinal muscles during back extension also increased significantly in all groups (Fig 2) and was weakly, but significantly (r = 0.37; p = 0.0001) correlated with increased strength in back extension. Although, at baseline, highly significant correlations were observed between the size of the paraspinal muscles (at L3/4 and at L4/5) and isometric back extension strength (r=0.75; p< 0.0001), post-training increases in strength were not accompanied by corresponding changes in muscle size. None of the improvements in strength showed any relationship with the clinical changes in pain and disability, regardless of whether the latter were examined on an individual basis or in relation to 'outcome groups'.
Conclusion: The superior trunk strength shown by the devices group post-therapy was considered to be attributable, in part, to a 'learning effect', of the type often seen when training and testing are carried out on the same machines. These gains are considered to be mostly 'task-specific'. However, part of the improvement in strength after active therapy (in all groups) also appeared to be due to an increased neural activation of the trunk muscles. These positive effects should be transferable to the performance of everyday activities for which the same muscles are employed, although the percentage improvement is probably not as high as the measured increase in strength might suggest. Possible roles for improved co-ordination and changes in motivation and/or pain tolerance after therapy cannot be excluded. No differences in the clinical outcome were observed between the three therapy groups, and the changes in physical performance after therapy did not correlate with the clinical outcome. It is therefore questionable whether strength measurements have any clinical significance in documenting the success of rehabilitation programmes, other than on a motivational basis. The results of the present study suggest that the value of supervised active therapy programmes does not reside in the reversal of specific muscular deficiencies, but rather in the provision of a source of confirmation/encouragement for the patient, that movement is not harmful, and a foundation upon which to further build. Whether the utilisation of specific training devices, or individual instruction, is necessary to elicit these particular effects is questionable.