Major depression disorder (MDD) is the leading cause of disability around the world (Friedrich, 2017). Although there are many pharmacological options available, one third to half of patients are resistant to antidepressants (Rush et al., 2006; Souery et al., 2007). Electroconvulsive Therapy (ECT) is still the gold-standard treatment for treatment-resistant depression (TRD), despite issues regarding stigmatization, accessibility, and potential adverse cognitive effects (Cole et al., 2021).

Repetitive transcranial magnetic stimulation (rTMS) is a non-invasive brain stimulation technique which has demonstrated antidepressant properties via innovative mechanisms, without the side effects of pharmacological antidepressants or ECT (Carpenter et al., 2012). Repetitive TMS has been approved by the US Food and Drugs Administration (FDA) as a treatment for TRD since 2008 (Janicak et al., 2010), and has been designated as a “Grade A first line treatment” for TRD in Canada (Milev et al., 2016). The clinical benefits of rTMS, in terms of response and remission rates, have varied in randomized controlled trials (RCTs) (Bulteau et al., 2019). According to a meta-analysis, response and remission rates were 29.3 % and 18.6 % respectively (Berlim et al., 2014), whereas a more recent meta-analysis reported higher response and remission rates of 44.6 % and 21 % respectively (Cao et al., 2018).

Subjects included in RCTs are rarely representative of real-world patient populations due to the application of strict selection criteria aiming to avoid comorbidities, the risk of suicide, and confounding variables (Blonde et al., 2018). The generalization of RCT results is therefore limited. Since FDA approval, several naturalistic studies (Bakker et al., 2015; Carpenter et al., 2012; Connolly et al., 2012; Galletly et al., 2017; Griffiths et al., 2019; Lacroix et al., 2021; Sackeim et al., 2020) have also reported the efficacy of rTMS in the treatment of depression in real world conditions. The most recent example, published by Sackeim et al. (2020) found, using the self-assessment Patient Health Questionnaire (PHQ), response and remission rates of 65 % and 31,7 % respectively.

However, previous studies either reported fairly small samples ((Abraham et al., 2007; Brakemeier et al., 2007; Dell'Osso et al., 2009; Huang et al., 2008; Nahas et al., 2004; Triggs et al., 1999)), or did not include bipolar patients (Baeken et al., 2010; Carpenter et al., 2012; Pallanti et al., 2012; Sackeim et al., 2020; Triggs et al., 1999), nor careful clinician-rated scales of depressive symptoms (Carpenter et al., 2012; McGirr et al., 2016; Sackeim et al., 2020), or provided little information on concomitant medication (Connolly et al., 2012; Lacroix et al., 2021; Sackeim et al., 2020), comorbidities (Carpenter et al., 2012; Lacroix et al., 2021; Sackeim et al., 2020), or resistance criteria (Kim et al., 2011; Sackeim et al., 2020), or they allowed inclusion of patients after non-response to only one antidepressant (Bakker et al., 2015; Connolly et al., 2012; Pallanti et al., 2012). Most naturalistic studies reported outcomes of only one stimulation parameter (10 Hz over the left dorso-lateral prefrontal cortex (DLPFC)), which does not reflect the diversity of rTMS designs.

We therefore carried out a large and comprehensive naturalistic study of rTMS as a treatment for TRD, with the aim of guiding clinicians in their daily practice.

Our goals were to (i) characterize how rTMS is used in clinical practice, (ii) measure changes in the severity of depression after rTMS, as well as response and remission rates, and, (iii) explore potential moderators of treatment response in the naturalistic use of rTMS.