Sham-controlled studies indicate effectiveness of TMS for negative schizophrenia symptoms


A systematic review and meta-analysis published in Nature Schizophrenia has concluded that data from sham-controlled studies to date indicate the effectiveness of transcranial magnetic stimulation (TMS) in treating the negative symptoms of schizophrenia. In addition, this analysis’ authors—Søren Østergaard (Aarhus University Hospital, Aarhus, Denmark) and colleagues—state that, “although it appears that targeting the L-DLPFC [left dorsolateral prefrontal cortex] and using a stimulation frequency >1Hz are the most efficacious settings, the optimal treatment parameters are yet to be established”.

Østergaard and colleagues begin by noting that pharmacological treatment is a “cornerstone of care” in schizophrenia and other psychotic disorders. However, they continue, while the ‘positive’ symptoms associated with schizophrenia, such as delusions and hallucinations, respond relatively well to these medications, drugs are somewhat less effective in treating the absence/lessening of normal functions, or so-called ‘negative’ symptoms, including affective flattening, alogia, apathy and social withdrawal.

As such, “identification and development of efficacious treatments of negative symptoms is a priority”, the authors claim, noting that a non-invasive neuromodulation technique like TMS could offer hope in this regard. TMS is currently approved by the US Food and Drug Administration (FDA) for the treatment of major depression—a mental health condition that can cause similar deficits in normal functioning to those seen in schizophrenia.

In light of recent trials, some of which have involved the use of novel stimulation parameters or neuronavigation to improve targeting, Østergaard and colleagues conducted a systematic review and quantitative meta-analysis of double-blind, randomised controlled trials (RCTs) evaluating TMS as a treatment for negative symptoms in schizophrenia patients. This involved searching MEDLINE, EMBASE, Web of Science, and PsycINFO for relevant studies in May 2021. Only studies involving adult participants (aged ≥18 years) were included, while co-initiation of other treatments with TMS, such as pharmacological therapies, was the main exclusion criterion.

Østergaard and colleagues ultimately identified 57 studies, with a total of 2,633 participants and spanning 15 different countries, that were eligible for inclusion in their meta-analysis. The authors report that all of the patients in these studies had schizophrenia or schizoaffective disorder, and the majority of eligible trials involved rTMS (n=48), while theta-burst stimulation (n=9) and deep-TMS (n=2) were seen less commonly. The effect of TMS on negative symptoms versus sham controls in each study was quantified by the standardised mean difference (SMD) with 95% confidence intervals and pooled across studies using an inverse variance random effects model.

Outlining individual study results, the authors state that 18 studies showed a statistically significant superior effect of TMS compared to sham treatment, while one study found a statistically significant superior effect with sham treatment. The remaining studies did not show a statistically significant difference between their treatment groups. There was “considerable heterogeneity” between the included studies, Østergaard and colleagues also report.

The pooled analysis of these results showed statistically significant superiority of TMS (SMD=0.41)—corresponding to a number needed to treat of five. In addition, follow-up data from at least four weeks after the end of treatment yielded an SMD of 0.27, the authors write. The effect in participants with predominantly negative symptoms was substantial (SMD=0.50), while no effect on depressive symptoms was observed (SMD=0.02). The SMD seen with different types of TMS did not differ in a statistically significant way either, according to Østergaard and colleagues.

Regarding stimulation sites, stimulation of the L-DLPFC had a statistically significantly larger effect when compared to other sites (SMD=0.55 vs SMD=0.04), although this may have been confounded by the “considerable methodological heterogeneity” in the ‘other sites’ category, the authors claim. They also state that, while stratified analyses suggested that TMS targeting the L-DLPFC and using a stimulation frequency >1Hz was most efficacious, the optimal treatment parameters “are yet to be established” in clinical trials.


Please enter your comment!
Please enter your name here