Skip to main content

Archived Comments for: The effectiveness of extra corporeal shock wave therapy for plantar heel pain: a systematic review and meta-analysis

Back to article

  1. Comparing apples to oranges to peaches

    Jan D. Rompe, OrthoTrauma Clinic

    28 May 2005


    I read with great interest what Dr. Thomson and co-workers call a rigorous effort to make a precise estimate of the effectiveness of ESWT.

    The problem of their evaluation is that the question they try to answer is wrong from the very beginning: “ Is ESWT effective in the treatment of patients with plantar heel pain when compared to a control group?”

    There is no such thing as one ESWT.

    If you compare apples (a disorder of various intensity and of various duration) to oranges (various ESWT regimens regarding number of sessions, number of shocks applied per session, various energy flux density per shock, various periods between applications) to peaches (various outcome measures, various periods of follow-up) is it not clear that you will find inconclusive evidence?

    Dr. Thomson and co-workers had identified 11 RCTs published from 1996 to 2003. All of these trials had to pass a rigorous review process to get published. Yet, according the authors, only 5 of those 11 RCTs could be included in the meta-analysis (a 6th trial required statistical re-evaluation by the reviewers).

    None of the trials published from our group was included in the meta-analysis, for not collecting morning pain outcomes, and for not giving follow-up data at 12 weeks.

    What kind of reasonable conclusion is possible from such kind of analysis? What practical worth can such an assessment have for the treating physician when (6 out of 11) apples are compared to oranges to peaches?

    The problematic nature of this review becomes even clearer when focusing on the trial Dr. Thomson and co-workers rated best, Michael Haake´s publication from 2003.

    From my point of view the results of this trial, as well-designed as it may be, show only the disastrous effect of a neglected co-variable, the simultaneous application of local anesthesia.

    In an upcoming publication in the Journal of Orthopaedic Research our group evaluated the effect of local anesthesia on the clinical outcome after repetitive low-energy ESWT for chronic plantar fasciitis. 86 patients with chronic plantar fasciitis were randomly assigned to receive either low-energy ESWT without LA, given weekly for three weeks (Group I, n=45; 3 x 2000 pulses, total energy flux density per shock 0.09 mJ/mm²) or identical ESWT with LA (Group II, n=41). Primary outcome measure was: Reduction of pain from baseline to month 3 post-treatment in a pain numeric rating scale [0-10 points] during first steps in the morning, evaluated by an independent blinded observer. Calculations were based on intention-to-treat.

    No difference was found between the groups at baseline. At 3 months, the average pain score was 2.2 ± 2.0 points for patients of Group I, and 4.1 ± 1.5 points for patients of Group II. The mean between-group difference was 1.9 points (95% CI: [1.1 - 2.7 points]; P<.001). Significantly more patients of Group I achieved ≥ 50% reduction of pain compared to Group II (67% vs. 29%, P<.001).

    ESWT as applied should be done without LA in patients suffering from chronic heel pain. LA applied prior treatment reduced the efficiency of low-energy ESWT.

    These results, which had been made available to Dr. Thomson well before his publication, fully confirmed a pilot study from Austria just published in the Z Orthop.

    Labek et al. included 60 patients with a chronic plantar fasciitis in a triple-arm (20 patients per group), prospective randomized and observer-blinded pilot trial. The patients were randomly assigned to receive either active ESWT without LA (3 x 1 500 shocks, total energy flux density [EFD] per shock 0.09 mJ/mm² [Group A]), ESWT with LA (3 x 1 500 shocks, EFD 0.18 mJ/mm² per shock [Group B]) or ESWT with LA (3 x 1 500 shocks, EFD 0.09 mJ/mm² [Group C]). Main outcome measures were: pain during first steps in the morning (measured on a 0-10 point visual analogue scale) and number of patients with > 50 % reduction of pain and no further therapy needed, measured at 6 weeks after the last ESWT.

    Group A improved in the VAS from 6.4 (SD: 1.7) to 2.2 (SD: 2.6) points, group B from 6.7 (SD: 1.5) to 4.1 (SD: 2.4) points, group C from 6.2 (SD: 1.6) to 3.8 (SD: 2.5) points. A reduction of pain of at least 50 % was achieved in 60 % of group A, in 36 % of group B and in 30 % of group C. Group A without LA showed a significantly higher improvement in the VAS and subjective evaluation than groups B (p = 0.007) and C (p = 0.016).

    At 6 weeks success rates after low-energy ESWT with local anesthesia were significantly lower than after identical low-energy ESWT without local anesthesia. Higher energy levels could not balance the disadvantage of this effect. LA significantly influenced the clinical results after low energy ESWT in a negative way. Blinding patients by LA in ESWT studies must therefore be considered a systematic error in study design.

    Keeping this in mind, I congratulate Dr. Thomson and co-workers on their effort. However, I am convinced they were doomed to fail right from the start to provide a sensible answer to an interesting question.

    Sincerely yours,

    Jan D. Rompe, MD

    Professor, Orthopaedic Surgery

    OrthoTrauma clinic

    Kirchheimer Str. 60

    D-67269 Gruenstadt


    Competing interests