An open randomised cross-over study of TNS compared with C was conducted on 21 patients who had previously been on the 1 cycle study. They continued for a further 5 months and then crossed over to the alternate treatment.
There was an advantage for TNS over C in FEV1%predicted change over time. The C slope was − 0.88%per month and the TNS slope 0.35%per month (p = 0.0002). This suggests advantages of TNS over C in a study with a small number of patients. Larger studies are required.