Two Versus One High-Frequency Repetitive Transcranial Magnetic Stimulation Session per Day for Treatment-Resistant Depression: A Randomized Sham-Controlled Trial. Response to Andrade and Colleagues
In their recent letter to the editor article,1 Dr Andrade and colleagues have expressed concerns about our study design,2 proposing that the patients involved in the twice-daily groups would have a larger placebo response. This was not the case, there was a marked difference for active stimulation compared with sham stimulation. Moreover, twice per day sham stimulation did not differ significantly from once per day sham stimulation. So, there was not a larger placebo response for twice-daily groups as Andrade and colleagues have suggested. In addition, debriefing data from both the patients and raters indicated that the blind was successfully maintained.
For what concerns sham stimulation, we did not have a coil that mimicked somatosensory experience; these coils were introduced in 20093 and our study began in 2006, so we did not have them in our disposal at the time.
We are also skeptical about the design that Dr Andrade and colleagues proposed, meaning to use both an active and a sham condition within the same day; we are afraid that it will affect patients' blinding; when patients have experienced an active stimulation condition, it is more than certain that they will distinguish it from a sham condition. Then, Dr Andrade and colleagues would have to randomize sham and active stimulation for these patients (whether the active condition is going to be first or second) to investigate if there is a difference between an active-sham and a sham-active within the same day condition, thus, increasing the number of patients needed.
Furthermore, when Andrade and colleagues propose to inform patients in the following way, “…because of variation in the rTMS procedure, they may or may not perceive differences between the first and second sessions during a day…,” we do not really know how ethical it is to withhold information from the patients (meaning that there will be a condition combining an active and a sham condition within the same day). On the contrary, our study design was more straightforward, delivering active or sham stimulation once or twice per day.
Last but not the least, in the study that Andrade and colleagues have chosen as an example of a successful crossover design, Haghighi et al4 did not use a sham coil that imitated somatosensory senses but they tilted the coil 45 to 90 degrees. So, this example seems to be irrelevant to what Andrade and colleagues propose (using a sham coil that imitated somatosensory senses). In addition, in the same study by Haghighi et al,3 blinding has not been checked (patients and raters were not asked if they thought the stimulation condition was active or sham); probably, this was not the purpose of this study.