Extracorporeal Shockwave Therapy in Insertional Achilles Tendinopathy

Participants aged 18-70 years old who were diagnosed as insertional Achilles tendinopathy by foot and ankle specialists: had symptoms of pain and swelling at Achilles insertion more than 6 months; failed other standard conservative treatments for 3 months, i.e., rest, medication, activity modification, stretching exercise, and heel lift orthosis; be able to complete questionnaires; and willing to participate were enrolled. A computerized block-of-four random numbers were generated into 2 groups; ESWT and placebo (sham) group. Investigators concealed the treatment allocation by sequential sealed opaque envelops. The sealed opaque envelops were opened just before the assigned treatment began. The outcome assessment in this study was seven times at pre-intervention, 2nd, 3rd, 4th, 6th, 12th and 24th weeks. The participants had to complete questionnaire of VAS and VAS-FA Thai version. The investigators used mean and standard deviation for continuous variables, and frequency (%) for categorical data. For non-normally distributed variables, median and range was presented. Continuous data of both interventions was compared using either unpaired t-test or Mann-Whitney U test based on normal distribution. While categorical data was compared using Fisher's exact test. Repeated measures with Bonferroni post hoc analysis was applied for each intervention according to times of follow-up, and between groups. Bartlett's test for equal variances was checked the assumption before using analysis of variance. If the data did not meet the assumption, non-parametric Kruskal-Wallis was performed. The investigators used STATA 15.0 Program, College Station, Texas, USA for all statistical analysis. P-value < 0.05 defined as a significant level. Sample size was estimated using STATA 15.0 Program based on alpha error 0.05, beta error 0.2, mean VAS in controls 5.0, mean VAS in intervention group 3.5, standard deviation (SD) of both group 2.3,(13) pre-intervention assessment for 1 time, post-intervention assessment for 6 times, correlation between the first and the last VAS assessment 0.8. The calculated sample size was 12 per group. With 20% increment for loss of follow-up, the final sample size was 15 per group. All of them remained in the randomized groups and analyzed as intention-to-treat basis.