The dynamic measurement and conservative treatment of thoracic hyperkyphosis

Abstract

Age-related hyperkyphosis of the thoracic spine is a problem which potentially affects all adults. It can result in movement dysfunction and may lead to mechanical failure of the thoracic spine, especially in the presence of osteoporosis, due to overwhelming forces exerted by gravity and muscular contraction. A number of studies have endeavoured to evaluate exercise-based programmes aimed at reducing hyperkyphosis in older adults. However, the multimodal nature of these programmes may reflect the uncertainty about which strategies are most effective.

Stroke is a condition which affects 322 000 people in Australia at any given time. Rehabilitation strategies for stroke have commonly excluded resisted strengthening strategies because of fears of increasing spasticity. However, recent studies have failed to confirm this concern. Loss of back extensor strength (BES) is a feature of stroke which is detrimental to function. The effect of resisted BES exercise on function in people with stroke has not been examined.

This thesis describes a number of studies that each inform the design and execution of a randomised controlled trial (RCT) which aimed to establish the relative effectiveness of BES exercises and postural re-education ireducing hyperkyphosis. The preliminary studies included three experiments validating the flexible electrogoniometer (FEG) as a tool to measure thoracic kyphosis, a survey looking at the normal practice of Australian physiotherapists with respect to thoracic hyperkyphosis; three experiments using surface electromyography (sEMG), kinematic and force measurements to determine whether sitting or prone lying was a better exercise positions for strengthening the thoracic erector spinae (TES); an ultrasound study looking at the anatomy of two sEMG recording sites; and a study validating the myometry used in the RCT.

The three FEG validation studies included: a bench test for accuracy, a test-retest reliability study and a study of concurrent validity comparing FEG angle to corresponding Cobb angles. The studies indicated that the FEG is a reliable instrument with excellent day-to-day reliability (ICC{u2082},{u2081} = 0.92; p < 0.0001 ). When compared with the Cobb angle for concurrent validity, the FEG was found to have the best agreement with the Cobb angle for the section of spine between mid end-blocks (r = 0.814 - 0.821, p = 0.001) with an absolute difference of 3.5{u00B0}{u00B1} 6.9{u00B0}.

A stratified cross-sectional mailed survey was used to examine how Australian physiotherapists from varying practice groups assess and manage hyperkyphosis. It revealed that postural re-education, stretching and strengthening were the interventions most frequently used to treat thoracic kyphosis but that the measurement tools used to evaluate treatment effectiveness were primarily subjective.

A prospective observational study which used real time ultrasound to image the muscles overlying the erector spinae at T3 and L4 established that the thoracic erector spinae (TES) could not be accurately recorded with sEMG. Therefore, a comparative analysis of the relative contributions of the TES and lumbar erector spinae (LES) was achieved by comparing the forces developed during prone and seated extension and the levels of LES activation. The results indicated that the TES were recruited to a greater extent during seated extension with scapular retraction than they were during prone extension. In addition, a kinematic study comparing the two exercises showed that prone extension primarily resulted in hyperextension of the lumbar spine with limited thoracic extension.

A test-retest study of a seated myometry method for testing BES showed that it had excellent day-to day reliability (ICC{u2082},{u2081} = 0.96 (95% CI 0.83 - 0.99)). The minimum difference needed to detect a real difference in force generated between measurements (MD) was 20.7N for extension with retraction.

The RCT was subject blinded and utilised a 2X2 factorial design to compare the effects of postural re-education and strengthening. Both stroke and non-stroke (normal) subjects were included although the majority of the subjects were normal. The results of the RCT indicated that, overall, the strengthening intervention resulted in better outcomes in terms of physical ability but that there was no significant reduction in kyphotic angle. The results also suggest that the angular changes which did occur mainly occurred in the upper thoracic spine. Thoracic spine movement frequency was found to be very low in both the sagittal (0.001Hz) and coronal (0.002 Hz) planes which may have implications for the nutrition of the intervertebral disc. There were no differences between the stroke and non-stroke cohorts in terms of their responses to the intervention.

The main clinical significance of this work is the discovery that an increase in back extensor strength does not necessarily result in a clinically significant decrease in thoracic kyphosis, especially at the apex of the curve. Further research is required to explore the best conditions in terms of load and position for thoracic extension strengthening for decreasing kyphosis. The effect of intervention on movement frequency is potentially an area of significant interest with respect to reducing the rate of disc disease.