The option of randomized controlled trials is generally not available for surgical techniques. In lieu of the randomized controlled trial a lower degree of evidence is therefore acceptable. The key features of any investigations should remain the randomization, which has independent value even when assessed in an open or a single-blinded fashion. The central problem is one of control, but in an exploratory period abstention from active treatment is enough. In future cases sham procedures may be carried out when possible, providing there are objective documentation and an adequate follow-up. There must be randomization of both patients and lesions to prevent bias. Follow-up should be for a minimum of 3-6 months and it is advisable to include both patient-centered subjective methods of assessment as well as more objective methods. Subjective methods can include health-related quality of life questionnaires developed for skin diseases, e.g., DLQI or Skindex [14, 15], and VAS scores of pain and disease severity. Objective quantification may include standardized clinical scores such as the Sartorius score , ultrasound imaging and photography.
For bilateral HS lesions randomized intra-in-dividual left to right comparisons of physical therapies are possible. However it is not advised to compare lesions in different regions directly, for example axillary and genitofemoral lesions; this is because the results of therapy often indicate different regional susceptibility in patients,
Table 24.2. Factors to consider when planning experimental studies of physical therapies. Controls are essential and should be set up for each lesion and for the patient. Inguinal lesions are not appropriate controls for axillary lesions. (DLQI Dermatology Life Quality Index, IPL intense pulsed light)
possibly because of differing shear forces and other local phenomena. Study requirements are summarized in Table 24.2. Early exploratory studies should similarly be aimed at providing the necessary data for the planning of larger controlled and randomized studies, and should therefore preferably use the same outcome variables.
Was this article helpful?