TMS Introduction

A short history of rTMS and the brain, starting at 1985 and ending at 2005 was published by The Regence Group. The following is a citation from the web-site of the Regence Group.
“Transcranial magnetic stimulation” was first introduced in 1985 as a new method of noninvasive stimulation of the brain. The technique involves placement of a small coil over the scalp; a rapidly alternating current is passed through the coil wire, producing a magnetic field that passes unimpeded through the brain. Transcranial magnetic stimulation was initially used to investigate nerve conduction; for example, transcranial magnetic stimulation over the motor cortex will produce a contralateral muscular-evoked potential. Interest in the use of transcranial magnetic stimulation as a treatment for depression was prompted by the development of a device that could deliver rapid, repetitive stimulation. In contrast to electroconvulsive therapy, transcranial magnetic stimulation does not require anesthesia, and does not induce a convulsion. Specifically, early studies suggested that transcranial magnetic stimulation of the left prefrontal cortex was associated with antidepressant properties.
While devices for transcranial stimulation have received approval by the U.S. Food and Drug Administration (FDA) for diagnostic uses, at the present time, no device has received FDA approval for transcranial magnetic stimulation of the brain as a therapeutic procedure. One device, NeoPulse (Neuronetic, Atlanta, GA) has received approval in Canada and Israel as a therapy for depression.

Policy/Criteria

Transcranial magnetic stimulation of the brain is considered investigational as a treatment of all psychiatric disorders, including, but not limited to depression.

Scientific Background

The published literature regarding transcranial magnetic stimulation (TMS) as a treatment of depression or other mood disorders is comprised of several small controlled trials of limited follow-up, consisting of differing patient populations, location of stimulus, and stimulation parameters.
A representative sample of the sham, controlled trials is reviewed below.
Pascual-Leone and colleagues performed a sham controlled study that included a crossover design, enrolling 17 patients with medication-resistant depression of psychotic subtype. (2) Nine of the patients had previously responded to electroconvulsive therapy (ECT). An attempt was made to withdraw medication before the TMS therapy but that was not possible in all patients, and some patients required reintroduction of medications during the trial. To create sham stimulation, the magnetic coil was held obliquely to the scalp surface, which mimicked the sensation of „real” TMS, but did not induce an intracerebral current. Each course of TMS consisted of 5 sessions over 5 consecutive days. The patients received 5 different courses of TMS, both real and sham; each applied at different scalp positions. The authors concluded that stimulation of the dorsal left prefrontal cortex had marked antidepressant effect, with 11 of 17 patients showing a decline in Hamilton Depression Rating Scale scores of 50%.
George and colleagues reported the results of a placebo-crossover trial in 12 medication resistant depressed patients who sequentially underwent TMS of the left prefrontal cortex or sham treatment. (3) Each treatment course consisted of 10 sessions over a 2-week period followed by crossover to the other study arm. There was a modest decline in Hamilton Depression Rating Scale scores when the subjects received active treatment.
Klein and colleagues conducted a randomized placebo controlled trial in 70 patients with depression who were assigned to receive active or sham TMS. (4) The stimulation parameters used in this study, described as „slow” (<1 Hz) were different than the above studies, which used „fast” (>1 Hz) stimulation. In addition, sham TMS consisted of stimulation over the right (as opposed to left) prefrontal cortex using a differently designed coil. Treatment consisted of 10 daily sessions over a 2-week period. At the end of the study, 41% of those in the treatment group reported at least a 50% reduction in Hamilton Depression Rating Scale scores compared to 17% of those in the sham-treated group.
Loo and colleagues reported conflicting results from a double-blind study of 18 depressed adults who were randomly assigned to a 2-week course of real or sham TMS, using the same stimulation parameters as Pascual-Leone, reviewed above. (5) Both groups improved significantly during the 2-week study period.
While these studies suggest the potential of TMS as a treatment of depression, larger placebo-controlled trials of a homogeneous group of patients are needed to further define the optimal stimulation parameters and validate a treatment effect. All of the above studies only examined the treatment effect immediately after the study ended, so durability of results is also unknown. The role of TMS in the overall treatment of depression requires further study. For example, it is not known whether TMS would be used as an alternative to electroconvulsive therapy or as an adjunct to partially effective pharmacologic therapy. Finally, at the present time, there are no TMS devices that have received FDA approval as a treatment of any neuropsychiatric disorder, including depression.
An updated search of the MEDLINE database through February 9, 2005 identified several additional published studies and review articles; however, none alter the conclusions reached above. In a meta analysis of 16 published trials, a Cochrane Review concluded that there is no strong evidence of benefit from TMS when used in the treatment of depression, finding no difference between TMS and sham TMS based on results of the Beck Depression Inventory or Hamilton Depression Rating Scale. In addition, the Cochrane Review found electroconvulsive therapy was more effective than TMS. Other studies and reviews of studies found no or modest clinically significant differences between TMS and sham TMS treatment. Studies comparing ECT to TMS found that response rates and relapse rates for depression were comparable or that ECT was more effective. Several studies found no or minimal effect of TMS on other neuropsychiatric disorders such as other mood disorders, post-traumatic stress disorders, Tourette’s Syndrome, and schizophrenia”.

An excellent paper on TMS in cognitive neuroscience was written by Vincent Walsh and Alan Cowey (2000), from the Department of Experimental Psychology, University of Oxford and is cited below.
“Transcranial magnetic stimulation and cognitive neuroscience”
Transcranial magnetic stimulation has been used to investigate almost all areas of cognitive neuroscience. This article discusses the most important (and least understood considerations regarding the use of transcranial magnetic stimulation for cognitive neuroscience and outlines advances in the use of this technique for the replication and extension of findings from neuropsychology. We also take a more speculative look forward to the emerging development of strategies for combining transcranial magnetic stimulation with other brain imaging technologies and methods in the cognitive neurosciences.
Transcranial magnetic stimulation (TMS) is now an established investigative tool in the cognitive neurosciences, and several groups have begun to exploit its potential in the study of perception, attention, learning, plasticity, language and awareness.
It is also finding applications in the study and treatment of movement disorders, epilepsy, depression, anxiety disorders, stuttering and schizophrenia. Despite the breadth and depth of the published research, the considerations behind the use of TMS and its value in addressing neuropsychological questions remain poorly understood. In this article we confront some of the most common confusions about TMS and show how it can be used to complement and extend existing techniques.
The use of TMS in clinical neurophysiological studies is highly advanced and has been reviewed elsewhere. Likewise, parameters for the safe use of TMS have been established and have been documented extensively in other sources that are required reading for those contemplating the use of TMS. Our aim is not to provide a technical introduction. Here we focus on the role of TMS in the cognitive neurosciences and propose a conceptual framework for the future application of TMS to this area.

TMS and the brain

TMS operates on Faraday’s principle of electromagnetic induction. Faraday showed that an electrical current passed through one coil could induce a current in a nearby coil. The current in the first coil produces a magnetic field that in turn causes current to flow in the second coil. In TMS that second coil is replaced by brain tissue and the induced electric field elicits neuronal activity. The key features of the technique are that the TMS machine delivers a large current in a short period of time — the current in the TMS coil then produces a magnetic field which, if changing rapidly enough, will induce an electric field sufficient to stimulate neurons or change the resting membrane potentials in the underlying cortex. In short, TMS can be used to induce a transient interruption of normal brain activity in a relatively restricted area of the brain.

The mechanism of interference

Perhaps the most common source of confusion over TMS is exactly how it interferes with cortical information processing to induce such a temporary lesion. As far as neuropsychological studies are concerned, the effect of TMS can be thought of as inducing ‘noise’ into neural processes. If a group of neurons are involved in a given task (for example, identifying a shape or matching a picture to a word), introducing a TMS pulse is highly unlikely to selectively stimulate the same coordinated pattern of neural activity as performance of that task. Rather, TMS induces activity that is random with respect to the goal-state of the area stimulated. In other words, TMS induces disorder rather than order into the information processing system, thereby disrupting task performance.
This ‘neural noise’ concept underpins what has become known as the ‘virtual patient’. By careful application, TMS can be used to transiently recreate the deficits seen in some neuropsychological patients, or can be used to create deficits that are rarely, if ever, obtained in neurological patients.

Spatial resolution

Another important source of confusion is the spatial resolution of TMS. The magnetic field produced by TMS is not spatially focal (in theory it is of infinite extent, like the earth’s gravitational field). However, the distribution of the induced electric field can and has been modelled, and progress has been made in relating the induced currents to specific sites of activation with a gate functions of medial cortex or sub-cortical structures. One should also be aware that stimulating deeper cortical structures (for example, in the sulci) may also stimulate overlying cortex. A potential solution to this problem is to stimulate areas that are accessible in non-human primates but not in human subjects. In cognitive neuroscience the chronometrics of information processing are central to many theories and experiments. For cognitive studies then, an understanding of the temporal resolution of TMS is at least as important as an account of its spatial selectivity. When a TMS pulse is delivered over an area of cortex, the effect is to simultaneously activate many neurons. At the point of maximal activation, the stimulated area will have its lowest signal-to-noise ratio with respect to the task it is trying to perform.
However, as neurons recover, the signal will increase, and whether or not TMS continues to have an effect will depend on the level of signal required for the task. Note that the interaction between the TMS signal and the contribution of an area to a task makes it highly unlikely that the time at which TMS has its maximal effect will correspond with the peak times reported in event-related potential (ERP) or subtractive or magneto-encephalographic (MEG) experiments).
An effectively disruptive pulse will interfere with processes that contribute to the build up of the ERP/MEG signal, so if the signal represents a neural event that is essential to the task, the time of TMS interference will typically precede ERP peaks and is more likely to coincide with single unit data. In other words, where an ERP result reports a peak at, say 300 ms, this may reflect the contribution of more than one neural event with a group maximum at 300 ms. When TMS is applied over the areas that contribute to this signal, it may disrupt processing of the individual components that may be maximal before, at, or after the reported peak at 300 ms.

Neuropsychology and TMS

Studies of neuropsychological patients have provided some of the most important revelations of brain–behavior relationships in the past century. They have made important contributions to the backbone of our understanding of the temporal lobe memory systems, the visuo-spatial functions of the parietal lobes, the different roles of the two cerebral hemispheres, the role of occipital lobes in resolution of a few millimeters. Indeed, there are now both indirect and direct demonstrations of the considerable specificity that can be achieved by this technique. Consider the indirect case first. The spatial and functional localization of positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) are achieved, in part, by comparing the effects on blood flow of different task conditions. The final spatial locations of, for example, ‘the motion area’, or an area important for memory processes or processing words, are then inferred from the differences between the activations produced by task conditions that vary only in the process under consideration.
Similar inferences can be applied to TMS. Here, however, the number of sites that can be compared is more restricted. This limitation provides a conceptual constraint on the application of TMS because a hypothesis is required for every comparison. The subtraction approach follows the logic of lesion analysis in humans and non-human primates, and that of functional neuro-imaging.
In most studies of cognitive function the TMS coil will be a figure-of-eight shape, which induces a maximum electrical field that peaks under the intersection of the two windings. The efficacy of the TMS pulse depends, in part, on the orientation of the underlying cell bodies and fibers with respect to the flow of the induced current. So, to increase confidence in the localization of effective stimulation, one can compare the behavioral effect of stimulation at several stimulation sites as described above, and determine the localization of the behavior in question by subtractive inference.
Alternatively, one can use a task control. Using task controls is of particular interest in cognitive studies as it may be sufficient to show that two processes are functionally dissociated in space or time. Anatomy and function can, however, be combined in TMS studies and some recent advances in combining TMS with neuro-imaging testify to its functional precision.
The first combined studies of TMS and PET showed that TMS induced neuronal activity under the site of stimulation, but that in addition, it also had effects at anatomically connected sites distant from the coil (much as a stimulus in an imaging experiment will activate many regions of the brain). Such studies may therefore have a future use in determining the functional connectivity of the human brain. More recent work has shown that the effects of TMS at the primary site correspond impressively with the activation produced by self-induced behaviour. For example, Seibner et al.applied 2 Hz repetitive pulse TMS (rTMS) over the left sensorimotor cortex of subjects at 140% of the motor threshold. They also asked subjects to imitate the arm movement caused by the rTMS and compared changes in regional cerebral blood flow in the two conditions. Both voluntary and TMS induced movement increased blood flow in the motor cortex contralateral to the arm movement, but voluntary movement also elicited greater activity in the supplementary motor area (FIG. 1).But both conditions excited the same connected cortical areas. George and colleagues have also shown the similarity of brain activations caused by voluntary and TMS-induced movements. So it is clear that TMS could be used to examine changes in connectivity as a function of learning in cases where the areas activated by action and TMS are in correspondence.
There are, however, several constraints that should guide the design of TMS experiments.
One important constraint on TMS is that the effects of stimulation are limited to superficial cortical regions and cannot be used to investigate functions of medial cortex or subcortical structures. One should also be aware that stimulating deeper cortical structures (for example, in the sulci) may also stimulate overlying cortex.

Temporal resolution

In cognitive neuroscience the chronometrics of information processing are central to many theories and experiments. For cognitive studies then, an understanding of the temporal resolution of TMS is at least as important as an account of its spatial selectivity. When a TMS pulse is delivered over an area of cortex, the effect is to simultaneously activate many neurons. At the point of maximal activation, the stimulated area will have its lowest signal-to-noise ratio with respect to the task it is trying to perform. However, as neurons recover, the signal will increase, and whether or not TMS continues to have an effect will depend on the level of signal required for the task. Note that the interaction between the TMS signal and the contribution of an area to a task makes it highly unlikely that the time at which TMS has its maximal effect will correspond with the peak times reported in event-related potential (ERP) or subtractive or magneto-encephalographic (MEG) experiments.
An effectively disruptive pulse will interfere with processes that contribute to the build up of the ERP/MEG signal, so if the signal represents a neural event that is essential to the task, the time of TMS interference will typically precede ERP peaks and is more likely to coincide with single unit data. In other words, where an ERP result reports a peak at, say 300 ms, this may reflect the contribution of more than one neural event with a group maximum at 300 ms. When TMS is applied over the areas that contribute to this signal, it may disrupt processing of the individual components that may be maximal before, at, or after the reported peak at 300 ms.

Virtual neuropsychology

Although patients may be available, their deficits are often transient. It is here that TMS can be used to formally reproduce the basis of the deficit of interest. For example, Fierro et al.applied rTMS (25 Hz for 400 ms) over the right parietal cortex to induce a transient neglect syndrome.
Subjects were asked to judge whether a briefly presented line was bisected centrally, or to the left or the right of centre. Patients with right parietal damage and left hemineglect typically underestimated the length of the left side of the line. Normal subjects, without TMS, tend to overestimate the length of the left side of the line (known as a ‘pseudo-neglect’). Right-sided TMS reduced this pseudo-neglect, that is, it caused subjects to underestimate the left side of the line relative to their own normal judgment. Here, then, is a transient, but formal and reproducible recreation of an effect associated with neglect that can be used to test the theories resulting from classical neuropsychological studies. The protocol adopted by Fierro et al. might also be more powerful if it could be developed in a reaction-time paradigm, which might allow the application of single pulse TMS and therefore chronometric analysis of the neglect syndrome. This study illustrates how TMS can be used to advance neuropsychology. First, Fierro et al. addressed a robust and widely studied phenomenon, the first step in any convincing extensions of neuropsychological findings. Second, they observed a surface difference between the TMS-induced deficits and the deficits seen in patients. This latter point is a feature of many TMS studies and far from driving a wedge between real and virtual neuropsychology, it demands that each approach take note of the differences observed with permanent and transient lesions. From the standpoint of making inferences about normal brain function, a dialogue between the results of these two disciplines is not an optional extra — it is an absolute necessity. The differences between real and virtual lesions may be accounted for by the effects of diaschisis (changes in activity acquired a wide repertoire of compensatory strategies to cope with the deficit. As Lomber pithily, but accurately, observes, this “spectre of compensation” means that lesion studies may “examine the capability of other cortical circuits in the absence of the removed cortical tissue and not the true functions of the vision and some aspects of functional specialization of vision. However, in the study of patients with brain damage or monkeys with specific brain lesions, one is studying an abnormal brain.

Dynamic connectivity

TMS can also be used to explore brain function in patients. One of the recent controversies concerning visual awareness is whether the specialized secondary visual areas such as V4 and V5 are sufficient for awareness of their preferred attribute (for example, color or motion), or whether they must interact with V1 to generate awareness.
A recent study applied TMS to extrastriate visual area V5 in a patient with almost total destruction of the striate cortex in the left hemisphere. TMS over area V5 can produce the illusory perception of motion in normal individuals; the question is whether it can also elicit motion perception in the absence of V1. The patient perceived normal moving phosphenes when V5 was stimulated in the cerebral hemisphere that had an intact V1, but motion perception could not be elicited from the blind hemifield by stimulating the hemisphere without an intact V1. The importance of V1–V5 interactions was further substantiated by the production of moving phosphenes in a peripherally blind patient by stimulation over area V5. This patient had suffered traumatic destruction of the optic nerves, but V1 was intact in both hemispheres.
This combination of real and virtual lesions is still in its infancy but it is clearly a paradigm that needs further exploration.
Neurophysiological studies have recently recorded the timing of interactions between extrastriate and striate cortex by cooling V5 while recording from V1. The effect of V5 deactivation occurred remarkably early — in the first 10 ms or so of the V1 response. Evidence of similarly fast, or perhaps continuous communication may be observed if TMS can be used to study the dynamics of backprojections in humans. The usual effect of V5 stimulation is to produce a perception of moving phosphenes, but this may be weakened or even abolished if V1 is stimulated within a critical time window of the V5–V1 interaction.
This of course would not imply that movement is perceived in V1, only that it is necessary for movement perception.

Language and memory

The effects of TMS on speech are well known and it is now clear that TMS can induce speech disruption that is dissociated from motor effects. As far as neuropsychology is concerned, however, this area awaits theoretically driven studies on the basis of psychological theories of language function. There have been demonstrations that applying TMS over the left frontal cortex can not only disrupt speech production but also impair verbal recall and picture–word matching. TMS over the posterior regions of language-related systems can also disrupt or facilitate picture naming. However, with the exception of one study based on neuroimaging data, the drive has been phenomenological rather than theoretical. The time course of verbal memory and recall and many of the different stages of verbal processing remain to be explored by TMS. Studies of memory are also in their infancy, and the combination of psychological techniques for the study of episodic memory or the effects of confidence judgements during recall are other areas awaiting investigation.

The future of TMS

The recent technical achievement of combining TMS with PET and MRI has been useful in studying the connectivity of the human brain, in validating the specificity of TMS and in guiding the location of TMS application. Looking to the future, the next step is to combine TMS, fMRI and PET in behavioral studies. One method that has already been used successfully is the application of low frequency TMS (for example, 1 Hz for 15 minutes) 20 minutes before the subject performs a task in the scanner. Low frequency rTMS reduces blood flow in the region stimulated for several minutes and can also produce a concomitant reduction in behavioral performance of tasks that rely on that region.

Figure 1 | Spatial and functional specificity of TMS.
This specificity is evident in the correspondence between blood flow changes induced by TMS over the motor cortex to produce a finger movement, and the activity produced by an intentional movement, which also produces supplementary motor area activity.
TMS has been used in studies of cognitive and sensory functions for over a decade, although much less extensively than functional neuroimaging by PET or fMRI. Its period of being a ‘new and exciting’ technique is over and work with TMS must now be judged within the theoretical frameworks used to evaluate other neurocognitive approaches.
TMS has clearly made a contribution to the understanding of perception, attention awareness and plasticity.
Further progress, however, will depend on the application of TMS in other areas such as the neuropsychology of language and memory.
The combination of TMS with other techniques, in conjunction with formal predictions on the basis of lesions of neural networks should provide fruitful avenues of research.
Many procedural and methodological hurdles remain, however, and the reliability and replicability of TMS effects in cognitive studies is a long way from that achieved in neurophysiological studies. The effects of TMS at anatomically connected areas also require careful consideration when developing good control procedures. It is to be hoped that the accumulation of experience from different groups will allow the further development of this method. One possibility for ‘fast tracking’ the methodological advance of TMS in studies of cognition might be some form of data sharing, as recently suggested for other areas of neuroscience. There are good reasons to approach this with caution, but in some cases, for example, mapping phosphenes on individual MRI scans or comparing the effects of different coils in similar experimental situations, data sharing may lead to faster and more efficient methodological advancement”.

TMS 1993-2004

A very compact and clear review on rTMS studies in depression over the period of 1993-2004, was written by Avery, George and Holtzheimer (2004), using three efficacy columns:
a. Decrease in depression ratings (percent decrease based on HAMD or MADRS scores)
b. Response rate (defined as a ≥50% decrease in HAMD or MADRS score)
c. Remission rate (defined as endpoint HAMD ≤ 8).

Their references and comments are cited below:

1. Hoflich G, Kasper S, Hufnagel A, S R, HJ M (1993): Application of transcranial magnetic stimulation in treatment of drug-resistant major depression— A report of two cases. Human Psychopharmacology 8:361-365.
No therapeutic effect.

2. George MS, Wassermann EM, Williams WA, et al (1995): Daily repetitive transcranial magnetic stimulation (rTMS) improves mood in depression. Neuroreport 6:1853-6.
Mean HAMD decreased from 23.8 to 17.5. Two robust responders, two mild responders, two non-responders (who were also ECT non-responders).

3. Grisaru N, Yarovslavsky U, Abarbanel J, Lamberg T, Belmaker RH (1994): Transcranial magnetic stimulation in depression and schizophrenia. European Neuropsychopharmacology 4:287-288 (abstract).
4 patients had slight improvement, 1 worsened, 5 showed no change.

4. Kolbinger H, Hoflich G, Hufnagel A, et al. (1995): Transcranial magnetic stimulation (TMS) in the treatment of major depression - a pilot study. Human Psychopharmacology 10:305-310.
Both rTMS groups improved slightly.

5. Conca A, Koppi S, Konig P, Swoboda E, Krecke N (1996): Transcranial magnetic stimulation: a novel antidepressive strategy? Neuropsychobiology 34:204-7.
rTMS used as an add-on to medication.

6. Pascual-Leone A, Rubio B, Pallardo F, Catala MD (1996): Rapid-rate transcranial magnetic stimulation of left dorsolateral prefrontal cortex in drug-resistant depression. Lancet 348:233-7.
HAMD decreased from 25.2 to 13.8 after 5 sessions. Study included multiple control conditions, including vertex stimulation, right DLPFC stimulation, and sham rTMS.

7. George MS, Wassermann EM, Kimbrell TA, et al (1997): Mood improvement following daily left prefrontal repetitive transcranial magnetic stimulation in patients with depression: a placebo-controlled crossover trial. Am J Psychiatry 154:1752-6.
rTMS superior to Sham, but small decrease in depression rating.

8. Geller V, Grisaru N, Abarbanel JM, Lemberg T, Belmaker RH (1997): Slow magnetic stimulation of prefrontal cortex in depression and schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry 21:105-10.
6 MDD/3 BMD/1 schizoaffective depressed included. Different group from previous open trial. Improvement is based of HAM-D results 2 hrs after treatment. Seven days after treatment showed 15% decrease. 28 days after showed 34% decrease in 9/10 patients.

9. Figiel GS, Epstein C, McDonald WM, et al (1998): The use of rapid-rate transcranial magnetic stimulation (rTMS) in refractory depressed patients. J Neuropsychiatry Clin Neurosci 10:20-5.
Sample overlaps with Epstein study, but includes new sample >65. Results calculated on 50 pts who completed study. Only 23% of >65 responded; 56% of those <65 responded. No seizures, but in two subjects, muscular contractions spread from a single hand muscle to more proximal muscles of the right upper extremity.

10. Feinsod M, Kreinin B, Chistyakov A, Klein E (1998): Preliminary evidence for a beneficial effect of low-frequency, repetitive transcranial magnetic stimulation in patients with major depression and schizophrenia. Depress Anxiety 7:65-8.

11. Avery DH, Claypoole K, Robinson L, et al (1999): Repetitive transcranial magnetic stimulation in the treatment of medication-resistant depression: preliminary data. J Nerv Ment Dis 187:114-7.
Slight improvement in rTMS group compared with sham. No decrement in neuropsychological tests with rTMS.

12. Klein E, Kreinin I, Chistyakov A, et al (1999): Therapeutic efficacy of right prefrontal slow repetitive transcranial magnetic stimulation in major depression: a double-blind controlled study. Arch Gen Psychiatry 56:315-20.
HAM-D decreased from 25.8 to 13.7 with rTMS and 25.3 to 19.7 with sham. Three dropouts (1 rTMS, 2 sham). Of rTMS patients, 49% were responders (w/ >50% decrease in HAM-D); of sham patients, 25% were responders.

13. Loo C, Mitchell P, Sachdev P, McDarmont B, Parker G, Gandevia S (1999): Double-blind controlled investigation of transcranial magnetic stimulation for the treatment of resistant major depression. Am J Psychiatry 156:946-8.
With rTMS significant decreases in HAMD after 10 sessions, but not different from sham. With rTMS, 44.9% decrease from baseline at one month follow-up.

14. Pridmore S, Rybak M, Turnier-Shea Y, Reid P, Bruno R, Couper D (1999): A naturalistic study of response in melancholia to transcranial magnetic stimulation (TMS). German Journal of Psychiatry 2:13-21.
The average reduction in the Montgomery-Asberg Depression Rating (MADRS) was 38.5 to 16.8 (56% decrease). The mean time from treatment to relapse was 20 weeks.

15. Triggs WJ, McCoy KJ, Greer R, et al (1999): Effects of left frontal transcranial magnetic stimulation on depressed mood, cognition, and corticomotor threshold. Biol Psychiatry 45:1440-6.
5/10 had at least 50% reduction in HAMD. No adverse effects on neuropsychological tests. Motor evoked potential threshold decreased during treatment in 9/10.

16. Pridmore S (1999): Rapid transcranial magnetic stimulation and normalization of the dexamethasone suppression test. Psychiatry Clin Neurosci 53:33-7.
Overall, mean MADRS decreased from 35.8 to 16.8 (53% decrease). All 12 patients were dexamethasone non-suppressors at baseline. 6 of 12 normalized the DST after treatment with rTMS. These 6 had good clinical improvement (MADRS decreased from 31 to 9) and maintained their response for at least 4 weeks.

17. Kimbrell TA, Little JT, Dunn RT, et al (1999): Frequency dependence of antidepressant response to left prefrontal repetitive transcranial magnetic stimulation (rTMS) as a function of baseline cerebral glucose metabolism. Biol Psychiatry 46:1603-13.
2/13 responded (greater than 50% response) There was a negative correlation between the degree of antidepressant response after 1 Hz compared to 20 Hz. Better response to 20 hz was associated with the degree of baseline hypometabolism measured by PET, whereas 1 Hz rTMS tended to be associated with baseline hypermetabolism. 1/10 responded in 20 Hz group. 0/3 responded in sham group.

18. Stikhina N, Lyskov EB, Lomarev MP, Aleksanian ZA, Mikhailov VO, Medvedev SV (1999): [Transcranial magnetic stimulation in neurotic depression]. Zh Nevrol Psikhiatr Im S S Korsakova 99:26-9.
TMS significantly better than control condition.

19. Padberg F, Zwanzger P, Thoma H, et al (1999): Repetitive transcranial magnetic stimulation (rTMS) in pharmacotherapy-refractory major depression: comparative study of fast, slow and sham rTMS. Psychiatry Res 88:163-71.
Improvement in verbal memory scores after fast rTMS, with no change after slow rTMS, and a trend toward poorer scores after sham.

20. Menkes DL, Bodnar P, Ballesteros RA, Swenson MR (1999): Right frontal lobe slow frequency repetitive transcranial magnetic stimulation (SF r-TMS) is an effective treatment for depression: a case-control pilot study of safety and efficacy. J Neurol Neurosurg Psychiatry 67:113-5.
Included 6 healthy controls who had no change in HAM-D score (mean 0.7). Depressed patients had decr in HAM-D from 18.4 to 10.6.

21. Schouten EA, D’Alfonso AA, Nolen WA, De Haan EH, Wijkstra J, Kahn RS (1999): Mood improvement from transcranial magnetic stimulation. Am J Psychiatry 156:669; discussion 669-70.
Poor response, Hamilton score decreased from 23.8 to 18.5. Some patients started medications during the first week of TMS. HAMD data incomplete on 3/7 patients. One responders and one partial responder (25% decrease in HAMD) were taking few or no benzodiazepines; all others were taking higher doses of benzodiazepines.

22. Berman RM, Narasimhan M, Sanacora G, et al (2000): A randomized clinical trial of repetitive transcranial magnetic stimulation in the treatment of major depression. Biol Psychiatry 47:332-7.
In rTMS group, 1/10 responded (decrease in HAM-D from 48 to 7); in sham group 0/10 responded.

23. Eschweiler GW, Wegerer C, Schlotter W, et al (2000): Left prefrontal activation predicts therapeutic effects of repetitive transcranial magnetic stimulation (rTMS) in major depression. Psychiatry Res 99:161-72.
rTMS significantly better than sham, also used near infrared spectroscopy.

24. George MS, Nahas Z, Molloy M, et al (2000): A controlled trial of daily left prefrontal cortex TMS for treating depression. Biol Psychiatry 48:962-70.
6/10 responded (greater than 50% decrease in HAMD). 3/10 responded to 20Hz. 0/10 responded to sham.

25. Grunhaus L, Dannon PN, Schreiber S, et al (2000): Repetitive transcranial magnetic stimulation is as effective as electroconvulsive therapy in the treatment of nondelusional major depressive disorder: an open study. Biol Psychiatry 47:314-24.
Comparison with ECT. 7/16 responded to rTMS;12/18 to ECT. Among nonpsychotic depressed 5/8 responded to rTMS; 5/10 to ECT. Among psychotically depressed, only 2/8 responded to rTMS; 7/8 to ECT.

26. Pridmore S, Bruno R, Turnier-Shea Y, Reid P, Rybak M (2000): Comparison of unlimited numbers of rapid transcranial magnetic stimulation (rTMS) and ECT treatment sessions in major depressive episode. Int J Neuropsychopharmacol 3:129-134.
Gave unlimited number of rTMS sessions (mean 12.2, SD 3.4). Compared to group of 16 pts receiving unlimited ECT treatments (mean 6.2, SD 1.6). No significant difference between the groups in HDRS improvement. Pts receiving ECT had significantly greater decrease in Beck Depression Inventory scores.

27. Conca A, Swoboda E, Konig P, et al (2000): Clinical impacts of single transcranial magnetic stimulation (sTMS) as an add-on therapy in severely depressed patients under SSRI treatment. Hum Psychopharmacol 15:429-438.
Each patient was given five stimulations to each of several sites during each session; there were 10 sessions total. sTMS was given as add-on to citalopram and trazodone therapy.

28. Garcia-Toro M, Mayol A, Arnillas H, et al (2001a): Modest adjunctive benefit with transcranial magnetic stimulation in medication-resistant depression. J Affect Disord 64:271-5.
rTMS added to current antidepressant treatments. 5/17 (29%) of patients initially randomized to rTMS were responders (>50% decr in HDRS). 15 sham non-responders crossed over to receive active 90% MT rTMS; 4/14 (29%) patients that completed 4 weeks of treatment were responders. The 9 non-responders were treated with 10 additional sessions of 110% MT rTMS; 3/9 (33%) were responders.

29. Manes F, Jorge R, Morcuende M, Yamada T, Paradiso S, Robinson RG (2001): A controlled study of repetitive transcranial magnetic stimulation as a treatment of depression in the elderly. Int Psychogeriatr 13:225-31.
Studied patients >50 y/o (mean age 60.7 yrs, SD 9.8 yrs). Neuropsych testing used MMSE; no significant difference between groups pre- or post-treatment. 6 responders (3 to rTMS and 3 to sham) had significantly greater frontal lobe volume than non-responders.

30. Garcia-Toro M, Pascual-Leone A, Romera M, et al (2001b): Prefrontal repetitive transcranial magnetic stimulation as add on treatment in depression. J Neurol Neurosurg Psychiatry 71:546-8.
Studied rTMS vs. sham as add-on treatment to sertraline for a MDE. All but two patients received benzodiazepines. Differences in response at 2 weeks in HDRS and BDI, but not at 4 weeks (2 weeks after last treatment). Non-responders to sham were crossed over to receive 90% MT rTMS with identical parameters. Non-responders to active 90% MT rTMS crossed over to receive 110% MT rTMS.

31. Lisanby SH, Pascual-Leone A, Sampson SM, Boylan LS, Burt T, Sackeim HA (2001): Augmentation of sertaline antidepressant treatment with transcranial magnetic stimulation. Biol Psychiatry 49:81S.
Compared 10 Hz LDLPFC rTMS to 1 Hz RDLPFC rTMS to sham rTMS, all as add-on therapy to sertraline 50 mg. Remission in the active TMS group combined was 25% vs. 8% in the sham group (NS). Degree of medication resistance negatively correlated with response and remission.

32. Szuba MP, O’Reardon JP, Rai AS, et al (2001): Acute mood and thyroid stimulating hormone effects of transcranial magnetic stimulation in major depression. Biol Psychiatry 50:22-7.
No efficacy data presented. Patients receiving active TMS showed greater mood improvements with acute sessions of TMS than patients receiving sham. Subjects are a subset of a larger study evaluation twice daily vs. once daily rTMS.

33. Catafau AM, Perez V, Gironell A, et al (2001): SPECT mapping of cerebral activity changes induced by repetitive transcranial magnetic stimulation in depressed patients. A pilot study. Psychiatry Res 106:151-60.
rTMS was associated with increases in regional CBF in the DLPFC, but these changes were not associated with change in HAMD.

34. Janicak PG, Dowd SM, Martis B, et al (2002): Repetitive transcranial magnetic stimulation versus electroconvulsive therapy for major depression: preliminary results of a randomized trial. Biol Psychiatry 51:659-67.
Patients received either rTMS or bitemporal ECT (4-12 treatments). There was a raw difference in mean change in HDRS between the groups (55% with rTMS, 64% with ECT), but no statistically significant difference. There was a 46% response rate with rTMS and a 56% response rate with ECT (not statistically significant).

35. Dolberg OT, Dannon PN, Schreiber S, Grunhaus L (2002): Transcranial magnetic stimulation in patients with bipolar depression: a double blind, controlled study. Bipolar Disord 4:94-5.
Preliminary report.

36. Shajahan PM, Glabus MF, Steele JD, et al (2002): Left dorso-lateral repetitive transcranial magnetic stimulation affects cortical excitability and functional connectivity, but does not impair cognition in major depression. Prog Neuropsychopharmacol Biol Psychiatry 26:945-54.
Parallel groups study of 5 Hz, 10 Hz, and 20 Hz rTMS in 15 patients (1 dropout due to incr psychosis). All groups showed decr in HAMD21. 2/4 responders (>60% decr in HAMD21) in 5 Hz group; 1/5 responders in 10 Hz group; 2/5 responders in 20 Hz group. SPECT showed incr ant cingulate activation, incr connectivity between LPFC and basal ganglia 1 hr after stimulation (no changes on right).

37. Padberg F, di Michele F, Zwanzger P, et al (2002a): Plasma concentrations of neuroactive steroids before and after repetitive transcranial magnetic stimulation (rTMS) in major depression. Neuropsychopharmacology 27:874-8.
18/37 patients responded (50% or greater decr in HAMD), 12/37 patients remitted (HAMD 9 or less). Plasma concentration of neuroactive steroids unchanged with rTMS and unrelated to response.

38. Padberg F, Zwanzger P, Keck ME, et al (2002b): Repetitive transcranial magnetic stimulation (rTMS) in major depression: relation between efficacy and stimulation intensity. Neuropsychopharmacology 27:638-45.
Compares 100% MT rTMS to 90% MT rTMS to sham. % decr for HAMD scores. Also, for MADRS scores: 4% decr with sham, 15% decr with 90% MT rTMS, 33% decr with 100% rTMS. 3/10 responders (>50% decr in HAMD) and 2/10 partial responders (>25% decr HAMD) with 100% MT rTMS, 2/10 responders and 1/10 partial responder with 90% MT rTMS, 0/10 responders and 2/10 partial responders with sham rTMS. Patients receiving rTMS had substantially fewer days in the hospital post-treatment (43 days for 100% MT rTMS, 61 days for 90% MT rTMS, 135 days for sham rTMS).

39. Boutros NN, Gueorguieva R, Hoffman RE, Oren DA, Feingold A, Berman RM (2002): Lack of a therapeutic effect of a 2-week sub-threshold transcranial magnetic stimulation course for treatment-resistant depression. Psychiatry Res 113:245-54.
No statistically significant difference between rTMS- and sham-treated patients. Authors suggest this may relate to subthreshold rTMS intensity.

40. Conca A, Di Pauli J, Beraus W, et al (2002a): Combining high and low frequencies in rTMS antidepressive treatment: preliminary results. Hum Psychopharmacol 17:353-6.
Complex design assessing relative efficacy of high frequency LDLPFC rTMS vs high and low freq LDLPFC rTMS vs high LDLPFC and low RDLPFC rTMS. No difference in response between groups. All patients were at least Thase-Rush Stage 4 tx-resistance. Two patients (left-handed women) developed psychotic sx with high freq rTMS; sx resolved after discontinuation of rTMS.

41. Rosenberg PB, Mehndiratta RB, Mehndiratta YP, Wamer A, Rosse RB, Balish M (2002): Repetitive transcranial magnetic stimulation treatment of comorbid posttraumatic stress disorder and major depression. J Neuropsychiatry Clin Neurosci 14:270-6.
Study of rTMS for PTSD comorbid with MDD. No benefit for treating symptoms of PTSD, but patients did show an antidepressant response.

42. Schiffer F, Stinchfield Z, Pascual-Leone A (2002): Prediction of clinical response to transcranial magnetic stimulation for depression by baseline lateral visual-field stimulation. Neuropsychiatry Neuropsychol Behav Neurol 15:18-27.
Subjects are part of a larger trial. Study was designed to test whether lateral visual field stimulation predicts response to rTMS.

43. Mosimann UP, Marre SC, Werlen S, et al (2002): Antidepressant effects of repetitive transcranial magnetic stimulation in the elderly: correlation between effect size and coil-cortex distance. Arch Gen Psychiatry 59:560-1.
Open study of older depressed patients (mean age 56.4±12.7 years, range 40-74 years). Coil-cortex distance index (CCDI; ratio of distance at LDLPFC to distance at motor cortex) negatively correlated with HAMD decrease.

44. Mottaghy FM, Keller CE, Gangitano M, et al (2002): Correlation of cerebral blood flow and treatment effects of repetitive transcranial magnetic stimulation in depressed patients. Psychiatry Res 115:1-14.
Subjects are part of larger trial comparing left and right 10 Hz and 1 Hz rTMS in unipolar depression.

45. Dragasevic N, Potrebic A, Damjanovic A, Stefanova E, Kostic VS (2002): Therapeutic efficacy of bilateral prefrontal slow repetitive transcranial magnetic stimulation in depressed patients with Parkinson’s disease: An open study. Mov Disord 17:528-32.
Patients with depression and Parkinson’s disease received open, 0.5 Hz rTMS over both left and right DLPFCs. Patients received 200 stimuli per day for 10 days (100 stimuli per hemisphere).

46. Schule C, Zwanzger P, Baghai T, et al (2003): Effects of antidepressant pharmacotherapy after repetitive transcranial magnetic stimulation in major depression: an open follow-up study. J Psychiatr Res 37:145-53.
Responders to rTMS showed a significant worsening in mood following cessation of rTMS treatment. This was reversed by mirtazapine.

47. Fitzgerald PB, Brown TL, Marston NA, Daskalakis ZJ, De Castella A, Kulkarni J (2003): Transcranial magnetic stimulation in the treatment of depression: a double-blind, placebo-controlled trial. Arch Gen Psychiatry 60:1002-8.
Compares 10 Hz LPF stim with 1 Hz RPF stim with sham. Initial trial was with 10 sessions: 14-15% decr in MADRS in both rTMS groups (1/20 patients in LPF group with >50% decr), 1% decr in sham. 15 patients with >20% decr in MADRS by 10 sessions went on to receive a total of 20 open rTMS sessions: 40% decr in MADRS in LPF group (4/8 patients with decr >50%), 57% decr in RPF group (4/7 patients with decr >50%).

48. Hoppner J, Schulz M, Irmisch G, Mau R, Schlafke D, Richter J (2003): Antidepressant efficacy of two different rTMS procedures High frequency over left versus low frequency over right prefrontal cortex compared with sham stimulation. Eur Arch Psychiatry Clin Neurosci 253:103-9.
Patients were started on an antidepressant medication 2 weeks before rTMS, which was used as an add-on treatment.

49. Herwig U, Lampe Y, Juengling FD, et al (2003): Add-on rTMS for treatment of depression: a pilot study using stereotaxic coil-navigation according to PET data. J Psychiatr Res 37:267-75.
Left vs. right prefrontal location of stimulation guided by PET-identified prefrontal hypometabolism (when present). 11 of 25 patients had right prefrontal hypometabolism at baseline. 1 of 25 had left prefrontal hypometabolism at baseline. 13 of 25 had left=right metabolism at baseline or no imaging data available. There was no evidence that using baseline prefrontal hypometabolism to guide treatment location was beneficial. Treatment groups were combined for analyses.

50. Zwanzger P, Baghai TC, Padberg F, et al (2003): The combined dexamethasone-CRH test before and after repetitive transcranial magnetic stimulation (rTMS) in major depression. Psychoneuroendocrinology 28:376-85.
Study of combined dexamethasone-CRH test before and after rTMS.

51. O’Connor M, Brenninkmeyer C, Morgan A, et al (2003): Relative effects of repetitive transcranial magnetic stimulation and electroconvulsive therapy on mood and memory: a neurocognitive risk-benefit analysis. Cogn Behav Neurol 16:118-27.
Comparison of patients receiving rTMS or ECT (unclear if patients were randomized). Patients in ECT group allowed to stay on medications; patients in rTMS group were tapered off medications. ECT patients were more depressed at baseline.

52. Brasil-Neto JP, Boechat-Barros R, da Mota-Silveira DA (2003): [The use of slow-frequency transcranial magnetic stimulation in the treatment of depression at Brasilia University Hospital: preliminary findings]. Arq Neuropsiquiatr 61:83-6. Epub 2003 Apr 16.
Three patients received open slow rTMS to the RDLPFC. 2/3 responded and 1/3 remitted. Two patients presented with psychotic features: 1 responded, 1 did not respond. Remitter did not have psychotic features and was less depressed at baseline.

53. Loo CK, Mitchell PB, Croker VM, et al (2003): Double-blind controlled investigation of bilateral prefrontal transcranial magnetic stimulation for the treatment of resistant major depression. Psychol Med 33:33-40.
No significant difference between the two groups. 2 responders in the rTMS groups, 1 responder in the sham group. 6 sham patients crossed over to rTMS; 1 pt in this group responded.

54. Grunhaus L, Schreiber S, Dolberg OT, Polak D, Dannon PN (2003): A randomized controlled comparison of electroconvulsive therapy and repetitive transcranial magnetic stimulation in severe and resistant nonpsychotic major depression. Biol Psychiatry 53:324-31.
No significant difference in HDRS decrease between rTMS and ECT patients. 12/20 ECT responders and 11/20 rTMS responders (>50% decr in HDRS, final HDRS <10, and final GAF >60); no significant difference between groups). 30% remission rate (final HAMD<9) in ECT and rTMS groups. For ECT group, patients received unilateral ECT initially, then bilateral ECT if no response after 6 treatments; 13 patients recv’d unilateral ECT, 7 recv’d bilateral ECT — no significant difference in response rate between these groups.

55. Nahas Z, Kozel FA, Li X, Anderson B, George MS (2003): Left prefrontal transcranial magnetic stimulation (TMS) treatment of depression in bipolar affective disorder: a pilot study of acute safety and efficacy. Bipolar Disord 5:40-7.
No significant difference between rTMS and sham in decrease in HAMD or response rate. No patients developed mania or hypomania during the study.

56. Kauffmann CD, Cheema MA, Miller BE (2004): Slow right prefrontal transcranial magnetic stimulation as a treatment for medication-resistant depression: a double-blind, placebo-controlled study. Depress Anxiety 19:59-62.
No significant difference between the groups. However, the active TMS group showed a significant reduction in HAMD over time whereas the sham group did not. Active TMS responders relapsed in 2-3 months. Sham responders relapsed in 2 weeks.

57. Fabre I, Galinowski A, Oppenheim C, et al (2004): Antidepressant efficacy and cognitive effects of repetitive transcranial magnetic stimulation in vascular depression: an open trial. Int J Geriatr Psychiatry 19:833-42.
All patients were older than 55 years and fulfilled criteria for vascular depression. Mean age 67.9 years, range 56-77 years. 5/11 patients had 25% decrease in HAMD and “almost” had a 50% decrease. Prefrontal atrophy was negatively correlated with HAMD change.

58. Hausmann A, Kemmler G, Walpoth M, et al (2004): No benefit derived from repetitive transcranial magnetic stimulation in depression: a prospective, single centre, randomised, double blind, sham controlled „add on” trial. J Neurol Neurosurg Psychiatry 75:320-2.
rTMS studied as an add-on treatment to antidepressant medication (with medication started at the start of rTMS). No significant differences between the groups. 20 Hz LPF and the combined 20 Hz LPF/1 Hz RPF active rTMS groups were combined for efficacy analyses. Unknown if patients were treatment-resistant at baseline.

59. Nahas Z, Li X, Kozel FA, et al (2004): Safety and benefits of distance-adjusted prefrontal transcranial magnetic stimulation in depressed patients 55-75 years of age: A pilot study. Depress Anxiety 19:249-56.
Study of older patients (mean age 61.2 years) with TMS intensity adjusted for prefrontal atrophy. With adjustment, treatment intensity was near 100% MT (actual intensity 114%, range 103%-141%).

60. Mosimann UP, Schmitt W, Greenberg BD, et al (2004): Repetitive transcranial magnetic stimulation: a putative add-on treatment for major depression in elderly patients. Psychiatry Res 126:123-33.
Study of rTMS in older patients (mean age 62 years).

61. Jorge RE, Robinson RG, Tateno A, et al (2004): Repetitive transcranial magnetic stimulation as treatment of poststroke depression: a preliminary study. Biol Psychiatry 55:398-405.
Study of rTMS in post-stroke depression.

62. Holtzheimer PE, 3rd, Russo J, Claypoole KH, Roy-Byrne P, Avery DH (2004): Shorter duration of depressive episode may predict response to repetitive transcranial magnetic stimulation. Depress Anxiety 19:24-30.
No significant difference between rTMS and sham; however, a significant negative correlation between length of current depressive episode and response to rTMS was found (r=-0.61, p<.05). Patients with a current episode of 4 yrs or less showed a 56% reduction in mean HAM-D whereas those with a current episode longer than 10 yrs showed only a 7% decrease.

63. Boechat-Barros R, Brasil-Neto JP (2004): Transcranial Magnetic Stimulation in depression: results of bi-weekly treatment. Rev Bras Psiquiatr 26(2):100-102.
Study of bi-weekly treatments over four weeks (8 treatments total). 50% decr in HAMD, 60% response rate, and 40% remission rate in this open study. Patients were allowed to continue on medications”.

Long term effects of rTMS

Although 900 publications on rTMS, including 89 reviews, can be found in Pubmed, at the time of our clinical trial long term effects of rTMS had not yet been published. But in 2002, Dannon et al published a preliminary report of their clinical trial on severely depressed patients(Dannon et al, 2002). Patients were followed on a monthly basis and outcomes were determined with the Hamilton Rating Scale for Depression-17 items (HRSD) and the Global Assessment of Functioning (GAF) scales. Medications were routinely prescribed. The 6-month relapse rate of the 21 patients from the rTMS group was 4. Differences on the scores in HDRS were 7.9 ± 7.1 and on the GAF 77.8 ± 17.1 at the 6-month follow up.

Clinical Trial on a 14 weeks effect of rTMS in the Lucas Andreas Hospital in Amsterdam

We conducted a double-blind, placebo-controlled trial of rTMS in 55 patients for 2 weeks, with a 12-week follow up, using random effects regression analysis to correct for unbalanced data, a common problem in longitudinal medical research.

Methods and Materials

Subjects / design
This study was designed as a double-blind placebo-controlled study, conducted over a 4-year period from 1997 to 2001. All inpatients and outpatients who met the DSM-IV criteria for major depressive episode were included. Further inclusion criteria were age over 16 and a score of 20 or higher on the 17-item Hamilton Depression Rating Scale (HDRS). Exclusion criteria were a history of epilepsy and a medical disorder that would preclude the administration of rTMS (Wassermann, 1998). Psychotropic medication was accepted if the dosage of antidepressive medication had not been changed for 6 weeks, and if the dosage of benzodiazepines had not been changed for 2 weeks prior to inclusion. For details of the psychotropic medication received see table 1.
Antidepressive medication had to remain stable during the 14 weeks of the study. Written informed consent was obtained. The ethics committee of the hospital approved the protocol.
Prior to treatment, all patients were assessed with standard clinical, psychiatric, and laboratory tests. Trained medical practitioners rated depression at baseline with the HDRS. Ratings were repeated at weeks 1, 2, 4, 8, and 14, with weeks 1 and 2 being the actual weeks of treatment. Patients were randomly assigned to rTMS or sham condition. Only the neurophysiologist applying the magnetic stimulation knew the chosen condition. All other members of the medical staff, as well as the patients and the data manager were blind to the treatment modality. Only after entering of all data into the data file the code for Sham / rTMS was revealed. Thus this study was double blind and placebo-controlled.

Group	MAO	TCA	SSRI	Mood  stabilizer	Neuroleptic	Hypnotic	Tranquilizer	Anticholinergic
rTMS	0	0	10	3	6	13	14	2
sham	0	4	17	2	9	14	16	0
Total	0	4	27	5	15	27	30	2

Table 1. Psychotropic medication received during the study.
Some patients received more than one medication.
MAO = monoamine oxidase inhibitor, TCA = tricyclic antidepressant, SSRI = selective serotonine re-uptake inhibitor, rTMS = repetitive transcranial magnetic stimulation.

Procedure rTMS

Subjects received rTMS daily on 10 consecutive weekdays (five sessions per week). The Maglite™ with a round stimulating coil MC-125 (Dantec Medical A/S. Skovlunde, Denmark) for biphasic pulses was used. Stimulation parameters were 20 Hz, 20 trains of 2 seconds, 30 seconds between trains, and 80% motor threshold. Before the first session, the optimal motor point and stimulation threshold for activating the right thenar muscles were determined for each patient by applying a single stimulus on the left side of the scalp. The left dorsolateral prefrontal cortex (left DLPFC) stimulation site was defined as being 5 cm anterior to this optimal motor point, according to the technique of Pascual-Leone et al (1996). This site was marked. During the rTMS session, the coil was centered flat over the left DLPFC. The small hole in the center of the coil permitted exact positioning by visual control of the mark. The sham treatment was performed by angling the outer edge of the coil 45˚ with the inner edge resting on the vertex, thereby inducing a contraction of the scalp and face muscles.

Statistics

Independent sample T-tests were used to investigate differences between rTMS and sham treatment in terms of age and HDRS scores (table 2 and 3). Chi-Square tests were performed to check whether the treatment and sham groups differed in terms of time and reason for drop-out, electroconvulsive therapy (ECT) before rTMS treatment, antidepressive medication during the study, inpatient or outpatient, age older than 65 years, personality disorder, type of depression, or side effects (table 3).

A hierarchical linear model was used to analyze the change in HDRS scores over time, and the difference between sham and rTMS treatment, as well as the influence of other characteristics. This model is also known as a multilevel model, or random-effects regression model (Gibbons et al.1993, Snijders 1996). Its main advantage over the standard repeated measures analysis of variance (as, for instance, implemented in the SPSS GLM module) is that it can handle unbalanced data, that is, it does not require the same number of measurements for each subject. Thus, all available measurements can be analyzed, whereas in a standard repeated measures analysis of variance either subjects with incomplete data are omitted from the analysis, or measurement points are discarded (typically, the later time points because of drop-out). This is undesirable because it results in loss of power and disregards the effect of drop-out.

The five HDRS scores (week 1, 2, 4, 8, and 14) were entered in the hierarchical linear model, using the HDRS score before treatment (‘week 0’) as covariate. The change in HDRS score over time was modeled with a piecewise linear model distinguishing between the treatment period (week 1 and week 2) and the post-treatment period. The large fluctuations between patients and measurements were modeled with random effects. The effect of rTMS on the decrease in HDRS score over the two periods was investigated, as well as the influence of covariates, such as sex, age, and left-handedness.

Discussion

The results of our study involving patients with a DSM IV diagnosis of major depression suggest that rTMS treatment has modest antidepressant effects. Depression scores decreased by 20% in both sham and rTMS groups during treatment but decreased further during 12 weeks of follow-up in the rTMS group only. Two studies conducted under similar conditions (left DLPFC, intensity 80%, 20Hz) revealed superior results for rTMS compared to placebo. George et al (1997)found a small decrease 2 weeks after cessation of treatment for 2 weeks. Berman et al, (2000), found a decrease over the 2-week course of rTMS, but, in contrast to the study of George et al (1997) and our study, benefits were no longer evident 2 weeks after the treatment had ended.

The inter individual variability may be associated with equivalent large variability in human neurophysiologic responses to specific brain interventions (Maeda et al, 2000). The degree of variability of rTMS effects in the human brain can be lowered somewhat by increasing the number of rTMS pulses to 1600 (Maeda et al, 2000), but still remains high.

The rather large number of drop-outs, which is quite common in a longitudinal study, caused missing observations. For this reason, we used the random effects model, which analyzes all available observations and which provides a more realistic approach to the analysis of longitudinal psychiatric data (Gibbons et al, 1993). Random regression models provide solutions for commonly observed problems of missing data, serial correlation, time-varying covariates, and irregular measurement times. Moreover, they accommodate systematic person-specific deviations from the average time trend (Gibbons et al, 1993). Mazumdar et al (1999) paid attention to the problem of missing values in longitudinal clinical trials. In data from clinical studies, the method of carrying the last observation forward has the major objection that strong time trends are often seen, and such time trends can very easily be confounded with treatment effects. Another model that is often used in intention-to-treat- analyses is the repeated-measures-model, which is mainly based on the assumption that observations are missing at random, an assumption that is rarely justified in psychiatric research.

Our rTMS treatment protocol was based on the method used in earlier studies (George et al 1997, Pascual-Leone et al 1996). Most studies, particularly those over the left dorsolateral prefrontal cortex use a figure of eight coil.

At the time this study was started, our method was more usual, but in the light of the current views of “power” of studies with TMS, our study is completely underpowered. Moreover, the sham rTMS we used was a 45 degree angle sham. Angling the coil at 45 degrees is not really a sham because it does induce stimulation of cortical tissue (Lisanby S, et al, 2001). Unfortunately, there appeared to be a fatal flaw in the methodology we used for stimulation. A circular coil creates the strongest stimulation at the outer edge of the donut, the stimulation being weakest or non-existent in the center, so in our method there would be no electrical activity at the DLPFC, but stimulation at distant regions of the cortex.
These methodological flaws might explain the modest improvement at the end of the first two weeks in the active group and the small difference at this point between the two groups. One more point that should be made is that 80% MT is now considered to be a very low stimulus intensity. Mark George found only a small decrease in de HDRS in 1997 when he used 80% MT but a larger decrease when in 2000 he used 100% MT. Others have had similar experiences.
At the time of our trial, improvement after the cessation of treatment had never been demonstrated before, nobody else had taken a routine, close look, such as we did. This may make our study a valuable contribution to TMS research.

References

1. Berman RM, Narasimhan M, Sanacora G, Miano AP, Hoffman RE, Hu XS et al (2000): A randomized clinical trial of repetitive transcranial magnetic stimulation in the treatment of major depression.
Biol. Psychiatry 47 (4): 332-337.

2. Dannon PN, Dolberg OT, Schreiber S, Grunhaus L (2002,in process): Three and six-month outcome following courses of either ECT or rTMS in a population of severely depressed individuals-preliminary report.
Biol. Psychiatry 51 (8): 687-690.

3. George MS, Wassermann EM, Kimbrell TA, Little JT, Williams WE, Danielson AL et al (1997): Mood improvement following daily left prefrontal repetitive transcranial magnetic stimulation in patients with depression: a placebo-controlled crossover trial.
Am. J. Psychiatry 154 (12): 1752 -1756.

4. Gibbons RD, Hedeker D, Elkin I, Waternaux C, Kraemer HC, Greenhouse JB et al (1993): Some conceptual and statistical issues in analysis of longitudinal psychiatric data.
Arch. Gen. Psychiatry 50: 739-750.

5. Kirkcaldie M, Pridmore S, Reid P (1997): Bridging the skull: electroconvulsive therapy (ECT) and repetitive transcranial magnetic stimulation (rTMS) in psychiatry.
Convuls.Ther 13: 83-91.

6. Lisanby SH, Gutman D, Luber B, Schroeder C, Sacheim HA (2001): Sham TMS: intracerebral measurement of the induced electrical field and the introduction of motor-evoked potentials.
Biol Psychiatry 49:460-463.

7. Loo C, Mitchell P, Sachdev P, McDarmont B, Parker G, Gandevia, S (1999): Double-blind controlled investigation of transcranial magnetic stimulation for the treatment of resistant major depression.
Am.J.Psychiatry 156 (6): 946-948.

8. Maeda F, Keenan JP, Tormos JM, Topka H, Pascual-Leone A (2000): Interindividual variability of the modulatory effects of repetitive transcranial magnetic stimulation on cortical excitability.
Exp Brain Res 133:425-430.

9. Mazumdar S, Liu KS, Houck PR, Reynolds III CF (1999): Intent-to-treat analysis for longitudinal clinical trials: coping with the challenge of missing values.
Journal of Psychiatric Research 33:87-95.

10. Pascual-Leone A, Rubio B, Pallardo F, Catala MD (1996): Rapid-rate transcranial magnetic stimulation of left dorsolateral prefrontal cortex in drug-resistant depression.
Lancet 348 (9022): 233-237.

11. Pinhas N. Dannon, Ornah T. Dolberg, Shaul Schreiber and Leon Grunhaus (2002): Three and six-month outcome following courses of either ECT or rTMS in a population of severely depressed individuals — preliminary report.
Biol. Psychiatry 51(8):687-90.

12. Snijders T (1996): Analysis of longitudinal data using the hierarchical linear model.
Quality & Quantity 30: 405-426.

13. Wassermann EM, Lisanby SH (2001): Therapeutic application of repetitive transcranial magnetic stimulation: a review.
Clin. Neurophysiol 8 (112): 1367-1377.

14. Wassermann EM (1998): Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines from the International Workshop on the Safety of Repetitive Transcranial Stimulation, June 5-7, 1996.
Electroencephalogr. Clin. Neurophysiol. 108 (1): 1-6.

Websites

15. http://www.regence.com/policy/
rTMS 1983-2005 by the regence group.

16. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?CMD=Display&DB=pubmed
900 publications on rTMS, including 89 reviews.

17. www.nature.com/reviews/neuroscience
TMS review by Vincent Walsh and Alan Cowey, 2000
Macmillan Magazines Ltd 74 | MONTH 2000 | VOLUME 1.

Curriculum Vitae Marij Bär

Profiel

Binnen de vakgebieden van Psychologie en Gezondheid heb ik wetenschappelijk onderzoek verricht en gecoördineerd, waarbij ik mijn doelen bereikte door mijn enthousiasme, doelgericht plannen, doorzettingsvermogen en zorgvuldige bewaking van het gehele onderzoeksproces. Sinds kort heb ik een eigen praktijk Counseling en Coaching.

Studie

HBO-Klinisch-Chemisch Analist, HBO-opleiding Nucleaire Geneeskunde, Universitaire opleiding Klinische Psychologie, afstudeerrichting Gezondheidspsychologie, Praktijktraining Counselling en Coaching, Academie voor Counselling en Coaching, Post-doc Master opleiding Counseling en Coaching, Benelux Universitair Centrum.

Werk

Medisch-analist, hoofd Isotopenlaboratorium,
Psycholoog-onderzoeker bij onderzoek van alcoholisme, behandeling van depressie en behandeling van paniekstoornis. Mede-auteur van enkele publicaties betreffende de diagnostiek van alcoholisme en biologische markers van alcoholisme.

Momenteel vind ik veel voldoening in de uitoefening van mijn beroep:
Counselor en coach,
Docent Psychologie,
Trainer praktijktrainingen Counseling en Coaching,
Vrijwilliger in Psychiatrische Inrichting.

Results

Fifty-five patients originally entered the study. Two patients stopped after one TMS session because of emergency ECT and extreme dizziness, respectively. One patient stopped after five sessions because of extra medication due to suicide risk. Therefore the data for 52 patients were analyzed: 26 sham and 26 rTMS patients. At week 14 the total number of drop-outs was 11 out of 26 sham patients and 14 out of 26 rTMS patients. Sham and rTMS patients did not differ significantly in age and HDRS score before the start of treatment (week 0). No differences between groups were found for sex, left-handedness, right-handedness or ambidexterity, type of depression according to DSM IV criteria, personality disorder, time or reason for drop-out, ECT before treatment, medication, side effects of treatment, inpatient or outpatient, or age older than 65 years.

During the treatment period, depressive symptoms improved in both groups (decrease >20% HDRS) (Figure 1), but there was no statistically significant difference between the groups. However, in the post-treatment period, differences between the two groups emerged. Twelve weeks after termination of rTMS treatment, HDRS scores of less than 10 were only found in 4 rTMS patients (14%), all of whom were men. Moreover a decrease in HDRS score of more than 50% was found in 5 out of 26 patients (19%) in the rTMS group (4 men, 1 woman), while 4 out of 26 patients (15%) who received sham treatment responded (all women).

	RTMS  N	Sham  N	rTMS Mean Hamilton (SD)	Sham Mean Hamilton (SD)		p-value
Week 0	26	26	25.9  (4.33)	25.9  (5.59)	0.99
Week 1	26	26	22.1  (6.83)	23.8  (6.54)	0.36
Week 2	25	24	21.1  (7.47)	21.9  (7.08)	0.71
Week 4	23	25	20.6  (9.25)	20.2  (8.14)	0.88
Week 8	19	18	15.5  (7.45)	21.2  (9.59)	0.06
Week 14	12	15	14.7  (7.96)	18.7  (8.21)	0.21

Table 2. differences in mean Hamilton scores in rTMS and sham groups in weeks 0-14, analyzed by T-Test (standard deviations in parentheses).

rTMS = repetitive Transcranial Magnetic Stimulation
HDRS = Hamilton Depression Rating Scale

	rTMS  n=26	Sham  n=26	p–value
Mean age	51 (15.4)	52 (13.2)	0.94
Number of males / females	14 / 12	9 / 17	0.13
Clinical patients / day clinical/ outpatients	15/0/ 6	14/1/ 10	0.43
patients receiving ECT before treatment	1	0	0.51
Number of patients >65 years	7	5	0.51
Number of patients left/right/ambidexterity	1 / 24 / 1	3 / 21 /1	0.55
Type of depression categorized according to DSM IV criteria			0.49
Number of patients with personality disorder	15	13	0.24
Total number of drop outs	13	12	0.75

Table 3. Differences in demographic and clinical variables analyzed by chi-square-test (standard deviations in parentheses).

	Week 2		Week 4		Week 8		Week 14
reason	rTMS	Sham	rTMS	Sham	rTMS	Sham	rTMS	Sham
Increase of symptoms			1		2	2	2	1
Strong increase of symptoms								1
Decrease of symptoms		2	1		2	1	2
Strong decrease of symptoms	1			1
Other						1	3	2
Total	3		3		8		11

Table 4. Reasons for dropout after completion of stimulation sessions.
Other = other reasons for dropout; ECT, holiday, no-show, moved to other treatment.

The random effects regression model revealed a significant decrease in HDRS scores during treatment (on average to 89% of ‘week 0’), but no significant difference between the rTMS and sham groups. The HDRS score decreased by about 2.5 points in the first week and by about 1 point in the second week in both groups. Thus there was an improvement of more than 10% in all patients. However, in the post-treatment period the two groups became progressively different, resulting in a significant mean difference of more than 4 points in HDRS score between the sham and rTMS groups. On average, women had higher scores (2 points) than men, and patients older than 65 years had higher scores (about 3 points) than younger patients. Somewhat unexpectedly, left-handed or ambidextrous patients had mean HDRS scores that were 3 or more points lower than those of other patients (see Table 5).

There were large individual differences, particularly in the post-treatment period. The overall between-patient variance was more than 12, indicating a standard deviation of approximately 3.5 HDRS points. The measurement variance was also about 12. The between-patient variance became larger in the post-treatment period, where the variance increased by more than 30 HDRS points, leading to an approximate standard deviation of 6.6 HDRS points. The largest variance, however, was associated with the change in HDRS score post treatment, as indicated by the so-called random slope. Therefore, although the average improvement was 1.10 HDRS points, the change varied enormously among patients (approximate standard deviation over 7 points), implying that in some patients the HDRS score increased after treatment. This large variability may also have masked differences between completers and drop-outs, which were found to be not significant.

Fixed Effect		Parameter Estimate	SE	p
Percentage decrease with respect to week 0		0.886	0.129	<0.00001
Score change week 0 – week 1	Overall	-2.503	3.476	0.24
	Difference rTMS vs. sham	-1.190	1.403	0.20
Score change week 1- week 2	Overall	-1.097	1.838	0.28
	Difference rTMS vs. sham	0.562	1.289	0.33
Post-treatment score change (weeks 4 – week 14)	Overall	-1.10	1.84	0.27
	Difference rTMS vs. sham	-4.400	2.680	0.05
Covariates	Women vs. men	2.129	1.297	0.05
	Older than 65 vs. younger	3.371	1.504	0.01
	Left/double-handed vs. right-handed	-2.88	1.899	0.06
Random effect
Between-patient variance	Overall	12.42	3.722	<0.00001
	Post-treatment (weeks 4, 8 and 14)	31.25	11.49	<0.0001
	Overall post-treatment score change effect	52.49	23.50	0.006
Measurement variance		11.46	1.818

Table 5. Results from the random effect regression model. Parameter Estimates, SE ‘s and p-values.