Corrigendum to “The effect of electroconvulsive therapy on cerebral monoamine oxidase a expression in treatment-resistant depression investigated using positron emission tomography” [Brain Stimul 12 (3) (2019) 714–723] (Brain Stimulation (2019) 12(3) (714–723), (S1935861X18313950), (10.1016/j.brs.2018.12.976))

Pia Baldinger-Melich, Gregor Gryglewski, Cécile Philippe, Matej Murgaš, Gregory M. James, Chrysoula Vraka, Leo Silberbauer, Theresa Balber, Thomas Vanicek, Verena Pichler, Jakob Unterholzner, Georg S. Kranz, Andreas Hahn, Dietmar Winkler, Markus Mitterhauser, Wolfgang Wadsak, Marcus Hacker, Siegfried Kasper, Richard Frey, Rupert Lanzenberger

Research output: Journal article publicationComment/debate/erratum


An error occurred during decay correction of the radioactive metabolites of [11C]harmine using high-performance liquid chromatography. Consequently, the quantification of monoamine oxidase A volume of distribution and subsequent statistical analyses were incorrect. We reran the analyses based on the correct data which revealed that the numerical changes resulting from the latest computations have no consequences for the main results or the conclusions of this publication that are subsumed below. There were no clinically relevant changes in MAO-A VT following a course of RUL ECT in TRD patients. In the results section, second paragraph, we now state the following: Linear mixed models analysis assessing the effect of ECT on MAO-A VT in depressed patients revealed a main effect of time (F = 20.50, p < 0.001), hemisphere (F = 8.56, p = 0.004) and ROI (F = 83.53, p < 0.001) and an interaction between hemisphere and ROI (F = 2.24, p = 0.015). The three-way interaction as well as other two-way interactions were non-significant. Post-hoc comparisons revealed a significant decrease of MAO-A VT between PET 1 and PET2 (PET1>PET2) and between PET2 and PET3 (PET2>PET3) (both p < 0.001, corrected). The estimates of mean MAO-A VT over all 27 ROIs were 19.92 ± 0.79 (mean ± standard error, SE) at PET1, 19.28 ± 0.78 at PET2 and 18.43 ± 0.79 at PET3. The test-retest variability of mean MAO-A VT values between PET1 and PET2 was 3.26% (Cohen's d = 0.81). Comparably, the effect of ECT on MAO-A VT was 4.53% (Cohen's d = 1.08). Using paired-samples t-test, change scores of MAO-A VT between PET1 and PET2 were compared to change scores of MAO-A VT between PET2 and PET3, showing no significant difference (left hemisphere p = 0.68, right hemisphere p = 0.80). We did not find significant differences in MAO-A VT between medicated TRD patients at baseline and age- and sex-matched healthy control subjects. In the third paragraph of the results section we now state the following: Comparing MAO-A VT between healthy controls and patients at baseline (PET1) revealed a significant main effect of ROI (F = 119.92, p < 0.001) and hemisphere (F = 18.90, p < 0.001) as well as a significant interaction ROI and hemisphere (F = 2.46, p = 0.004). Post-hoc separate analyses per hemisphere, however, revealed only numerical differences between groups, indicating slightly lower mean MAO-A VT in healthy controls compared to MDD patients in the right (17.69 ± 0.90 versus 19.88 ± 0.90, mean ± SE), and left hemisphere (18.07 ± 0.86 versus 19.81 ± 0.86, mean ± SE). An updated version of Table 2 summarizing mean regional MAO-A VT in TRD patients at each time point and HC is shown below. Further information about corrections related to the exploratory analyses performed in this study is available on request. [Table presented]

Original languageEnglish
Pages (from-to)1280-1281
Number of pages2
JournalBrain Stimulation
Issue number5
Publication statusPublished - 1 Sep 2020

ASJC Scopus subject areas

  • Biophysics
  • Neuroscience(all)
  • Clinical Neurology

Cite this