Some years ago, I treated a young woman who was bothered a lot by hearing voices. The voices were negative: calling her names and commanding her to hurt herself. She suffered a lot and years of psychotherapy together with clozapine, the most effective antipsychotic medication, could not relieve her of these terrible voices.
She is not the only one. Many people struggle with chronic, frequent voices that are hard to bear. Some of them have psychotic disorders, others ‘personality disorders’ or a combination of the two. In the nineties, we hoped to add a new tool to expand the options for solution for this type of hallucinations. Many studies, including several from my lab, tried to reduce hallucination severity with repetitive Transcranial Magnetic Stimulation (rTMS). Although small initial studies from the renowned Professor Hoffman showed efficacy (Hoffman et al, 2005; Hoffman et al, 2003), we and others could not replicate efficacy in larger samples.
It is not to say that rTMS did nothing on hallucinations, some people, did experience some improvement, sometimes a lot. But placebo effects were also high and it was difficult to show superiority of rTMS above placebo. I must say that I kind of gave up on rTMS as I was not sure about its mechanisms and if it could really make the difference for people with severe hallucinations. Until I saw the study from Plwenia and colleagues that appeared in The Lancet Psychiatry this September. They did nothing new really, but they simply did it better: larger sample size, better blinding, longer treatment duration. Needless to say: it is a German study, actually a collaboration of seven excellent German hospitals.
While early studies of rTMS for voices showed promise, we were not able to replicate results.
Methods
The German consortium were able to include 138 adults (43% women) with chronic severe auditory verbal hallucinations (voices) and all had a diagnosis within the schizophrenia-spectrum. Participants had been diagnosed with schizophrenia or schizo-affective disorder for a mean period of 12 years, indicating they had probably been struggling with voices for a long time. Moreover, 38% of them used clozapine, the strongest of all antipsychotics and had hallucinations nevertheless, which indicates that these people had really tried a lot to get rid of their voices.
These adults were randomised 1:1 between active rTMS and placebo, usually called ‘sham’. rTMS was given with a theta-burst frequency, which means that rapid and short bouts of magnetic field changes are induced at the participant’s skull, with a coil pressed gently to the skin. These magnetic field changes can pass the skull and induce a small electrical current that can interact with the underlying brain tissue. rTMS is not painful and when applied correctly very safe. The investigators provided 600 of such pulses to the right side of the head and 600 to the left. They repeated this procedure on the working days of 3 weeks. The place they targeted was close to Wernicke’s area (and its right-sided homologue), which is the major language centre of the brain. We know that this part of the brain is active when patients hear voices, both in the left and in the right hemisphere.
Their primary outcome was change in the Psyrats, a questionnaire that quantifies the severity of auditory hallucinations. Psyrats assessments were completed at baseline and at the end of the end of each of three treatment weeks. Follow up assessments were completed one, three and six months after the end of treatment. Secondary outcomes were PANSS (a measure of positive and negative symptom severity) and the Global Assessment of Functioning (a measure of psychosocial functioning).
Wernicke’s area (and its right-sided homologue), which is the major language centre of the brain, was targeted.
Results
130 patients were included in the final analysis following eight withdrawals. 56 out of 60 people in the rTMS group and 54 of 64 in the sham group received what was considered to be the protocol compliant amount of 12 sessions.
As with earlier studies, Psyrats scores for both groups improved suggesting a reduction in auditory hallucinations. The mean change score was -6.36 for real rTMS and -3.74 for sham, significant at p=0.042. The adverse-events were few, mostly headache, which was noticed just as often in the sham condition as in the real rTMS condition, suggesting it may be more related to expectations than to the actual magnetic fields.
No statistically significant differences in functioning were observed between the groups (as measured with the GAF) with similar findings for negative symptoms as measured with PANSS. However, there was a significant reduction in PANSS positive symptoms including a significant reduction in the PANSS item related to auditory hallucinations.
This new German trial suggests that rTMS for persistent auditory hallucinations is more effective than sham (placebo) treatment.
Conclusion
In sum, rTMS for persistent hallucinations is more effective than sham and is relatively well tolerated.
The authors concluded:
Considering its excellent safety and tolerability profiles, cTBS [continuous theta burst stimulation] provides an additional therapeutic strategy that complements medication and psychotherapy for this frequently and severely disabling condition.
The treatment was well tolerated by participants.
Strengths and limitations
This study has many strengths: it is done very thoroughly and well. It is a multi-centre study, which means a potential bias of one researcher has not played a major role. The sample was large enough and participants were well blinded. Participants were treated over a three week period, which means they had to visit the hospital a lot. When living far away, that may be a challenge. Yet, I do think the frequent treatments have contributed to the positive effect, so I would not recommend to lower the number of treatments. Perhaps the travel schedule could be simplified by providing two treatment sessions per day. For example one before and one after lunch.
The study did though have some limitations. These included:
- A relatively short observation period of three weeks.
- Lack of control in post intervention treatments.
- Relatively high drop out rates post intervention.
These challenges are common to studies with this population and should be considered in future trials.
Participants were treated over a three week period, which means they had to get to hospital sites frequently.
Implications for practice
I was happy with this study, as it is the largest so far and it shows a positive effect. Not a huge effect, but that was not to be expected. I know from clinical practice that several patients with persistent hallucinations fare well with this type of therapy, but it is currently not reimbursed here in the Netherlands or recommended within clinical guidelines. When we look to the field of depression, the effect sizes of rTMS are not larger. However, more large-scale RCTs have been done for that disorder and efficacy has been demonstrated thoroughly. Now, rTMS for depression is reimbursed in many European countries. I have good hopes that rTMS for persistent voices will follow that same path and that patients have yet an extra option to find relief of bothersome hallucinations when psychotherapy, medication and the combination of those two have failed. To further make the case, we also need more research with larger samples that are able to address some of the challenges seen in this study, in particular in relation to loss to follow-up.
Many university medical centres have TMS labs and could provide this type of treatment for people with persistent hallucinations. As it is time-intensive for both patient and staff, it comes with costs. Clinicians and researchers should now talk to national authorities and health cost insurances to regulate reimbursement also for hallucinations. I hope the German group from this study can also provide some data on cost-effectiveness to show that reimbursing rTMS to help people get rid of persistent hallucinations improves quality of life for them and also reduces long-term service use.
This study provides hope that a new and safe treatment may be possible to provide relief to that young woman I worked with all those years ago, and to the many other people whose life is limited by experiencing distressing voices.
This study provides hope for a new treatment for the many people who live with distressing voices every day.
Links
Primary paper
Plewnia C, Brendel B, Schwippel T, Becker-Sadzio J, Hajiyev I, Pross B, Strube W, Hasan A, Campana M, Padberg F, Mayer P, Kujovic M, Lorenz S, Schönfeldt-Lecuona C, Otte ML, Wolf RC, Höppner-Buchmann J, Serna-Higuita LM, Martus P, Fallgatter AJ. Theta burst stimulation of temporo-parietal cortex regions for the treatment of persistent auditory hallucinations: a multicentre, randomised, sham-controlled, triple-blind phase 3 trial in Germany. Lancet Psychiatry. 2025 Sep;12(9):638-649.
Other references
Hoffman RE, Gueorguieva R, Hawkins KA, Varanko M, Boutros NN, Wu YT, Carroll K, Krystal JH. Temporoparietal transcranial magnetic stimulation for auditory hallucinations: safety, efficacy and moderators in a fifty patient sample. Biol Psychiatry. 2005 Jul 15;58(2):97-104.
Hoffman RE, Hawkins KA, Gueorguieva R, et al. Transcranial Magnetic Stimulation of Left Temporoparietal Cortex and Medication-Resistant Auditory Hallucinations. Arch Gen Psychiatry. 2003;60(1):49–56.