Sample sizeTo calculate the number of patients, we used the 3. 1. 2 version of PS which was freely available software. We planned for the study of continuous response variables acquired from patients in the MLD (intervention) and control groups. We based our sample size on published findings which included a report that acute pain is of significant difference clinically when (VAS) varies at least 13 mm between groups. In another study of pain in the acute phase after an artificial knee joint operation, the responses within groups of patients were normally distributed with standard deviations of 15 mm25). Therefore, assuming a true difference between the MLD and control group would be repre- sented by a mean VAS for pain of 13 mm, we required 22 patients in each group to be able to reject the null hypothesis that the population means of the MLD and control groups are equal with a probability (power) of 0.8. The type I error probability associated with the test of this null hypothesis is 0.05. Sample sizeTo calculate the number of patients, we used the 3. 1. 2 version of PS which was freely available software. We planned for the study of continuous response variables ac- quired from patients in the MLD (intervention) and control groups. We based our sample size on published findings which included a report that acute pain is of significant difference clinically when VAS varies at least 13 mm between groups.In another study of pain in the acute phase after an artificial knee joint operation, the responses within groups of patients were normally distributed with standard deviations of 15 mm25). Therefore, assuming a true differ- ence between the MLD and control group would be represented by a mean VAS for pain of 13 mm, we required 22 patients in each group to be able to reject the null hypothesis that the population means of the MLD and control groups are equal with a probability (power) of 0.8. The type I error probability associated with the test of this null hypothesis is 0.05.