Brain changes induced by electroconvulsive therapy (ECT) – understanding treatment mechanisms and predictors of clinical response.


Electroconvulsive therapy (ECT) is a procedure, done under general anesthesia, in which electric currents are passed through the brain, intentionally causing a brief seizure. Although ECT has remained the most effective acute treatment for major depressive episodes for nearly 80 years, how it works and why it is effective is not well known. Only a small proportion of patients with treatment resistant depression are offered ECT, and typically 60-80% experience a good response (> 50% symptom reduction).

Several neuroimaging studies of ECT have demonstrated volume increases of brain areas after treatment. The most consistent finding has been volume increase of the hippocampus. However, a link between the treatment induced brain changes and clinical improvement has not been established.

This project will explore how ECT changes the brain, and if the structural (or functional) changes that are seen can explain the effect (or side-effects) of the treatment.

An important reason for our knowledge gap is that single studies have limited samples and lack statistical power to reach firm conclusions. The Global ECT-MRI Research Collaboration was founded with a belief that collaboration and mega-analysis of combined data will lead to new knowledge that can be generalized across individual research sites.

Within GEMRIC we have the world’s largest data sample with clinical and MRI neuroimaging data before and after ECT – our initial study will include up to 550 subjects (including controls). In our next study we aim at increasing the sample size to 2000 subjects. GEMRIC uses state-of-the-art imaging tools, developed in a world leading laboratory (dr. Anders M Dale, UCSD) as well as tools developed by GEMRIC collaborators. Instead of distributing data among collaborators our philosophy is to collect all the data at a common data portal and then share analysis pipelines into this server. This way, data security can be kept at a very high level while at the same time almost any analysis pipeline can be used. Pipelines are set up as Docker containers – which means standardization, independence of operating systems and reproducible results.

New collaborators are welcome!

Project PI Leif Oltedal

Co-PI Ute Kessler; clinical effects of ECT

Co-PI Jan Haavik;  studies of Blood Biomarkers

 GEMRIC Board members and Sites

  • Leif Oltedal, co-ordinator (Bergen, Norway)
  • Katherine Narr (Los Angeles, USA)
  • Christopher Abbott (New Mexico, USA)
  • Indira Tendolkar (Nijmegen, The Netherlands)
  • Bogdan Draganski (Lausanne, Switzerland)
  • Miklos Argyelan (New York, USA)
  • Amit Annand (Cleveland, USA)
  • Annemieke Dols (Amsterdam, The Netherlands)
  • Louise Emsell (Leuven, Belgium)
  • Pia Nordanskog (Linköping, Sweden)
  • Martin Balselv Jørgensen (Copenhagen, Denmark)
  • Ronny Redlich (Münster, Germany)
  • Carles Soriano Mas (Barcelona, Spain)
  • Akihiro Takamiya (Tokyo, Japan)
  • Hauke Bartsch / Anders M Dale (San Diego, USA)