Acute effects of repetitive transcranial magnetic stimulation on attentional control are related to antidepressant outcomes
===========================================================================================================================

* Marie-Anne Vanderhasselt
* Rudi De Raedt
* Lemke Leyman
* Chris Baeken

## Abstract

**Background:** Repetitive transcranial magnetic stimulation (rTMS) applied over the dorsolateral prefrontal cortex (DLPFC) is a new treatment procedure that holds promise of more insight into the pathophysiology of depression because the DLPFC may play an important role in the interplay between emotional and attentional information processing. We sought to investigate whether acute neurocognitive effects of rTMS are related to antidepressant outcomes.

**Methods:** Between January 2005 and May 2007, we examined the effects of a single session compared with 2 weeks of rTMS over the left DLPFC on cognition and mood in therapy-resistant patients with depression. We used a crossover placebo-controlled double-blind design and differentiated rTMS treatment responders and nonresponders. We used a task-switching paradigm to measure cognitive function.

**Results:** After 2 weeks of high-frequency rTMS over the left DLPFC, depressive symptoms improved in more than half (53%) of our therapy-resistant population. After a single session, mood did not improve but attentional control was increased solely within our group of treatment responders.

**Limitations:** Our results should be interpreted as preliminary because our sample was small and because the cognitive task we used has not been tested for validity and reliability. In addition, despite minimal stimulation of the DLPFC during sham stimulation, it is possible that the stiumlation was partially active. Finally, benzodiazepines may have had impairing effects on the attentional task.

**Conclusion:** Cognitive reactivity after a single session of rTMS may hold promise as a predictor of beneficial treatment outcomes. Moreover, within the group of responders, attentional control appears to play an important role in the progress of mood disorders.

## Introduction

Repetitive transcranial magnetic stimulation (rTMS), a non-invasive method of neuronal depolarization of specific areas of the human brain, is a relatively new technology that holds promise for therapeutic advances and insight into the pathophysiology of depression.1 To date, numerous open and controlled clinical trials have demonstrated that high-frequency (> 1 Hz) rTMS applied over the left dorsolateral prefrontal cortex (DLPFC) and low-frequency (< 1 Hz) rTMS applied over the right DLPFC have antidepressant effects (for a review see Gershon and colleagues2). On the other hand, a detailed review of the literature reveals that the data to date are inconsistent.1,3 Because of these contradictory results, the efficacy of rTMS remains a topic of debate.

Recent positive, sham-controlled studies with more aggressive treatment parameters and large numbers of sessions demonstrated the efficacy of rTMS in the treatment of depression.4,5 Therefore it appears desirable to further assess the antidepressant effects of high-frequency rTMS over the left DLPFC,6 and one promising avenue might be to investigate characteristics of treatment responders versus non-responders. This makes it possible to search for markers of rTMS effects and to investigate possible mechanisms underlying its therapeutic efficacy. It has recently been argued that the heterogeneity of research findings might be due to a high variability in response to rTMS among depressed patients.7 Whereas previous research focused mainly on exogenous stimulation parameters, we sought to investigate endogenous features predicting outcomes of treatment.

Positron emission tomography and functional magnetic resonance imaging (fMRI) studies have established a regulative role for a specific cortico–subcortical circuit in mood disorders8 and, accordingly, have postulated that a dorsal circuit plays an important role in the interplay between emotional and attentional information processing.9 We sought to explore underlying attentional mechanisms related to the outcome of anti-depressant therapy with rTMS, starting with nonemotional stimuli. Although rTMS is currently used as a new tool for neuropsychological research in healthy individuals,10 cognitive function in studies investigating clinical populations is frequently viewed as epiphenomena.3 The most common observation of previous studies is that rTMS has no major detrimental cognitive effects after several weeks of daily rTMS in depressed patients.11 Some of these studies reported improvement among depressed patients on a number of cognitive tests, including verbal memory,12 verbal fluency13 and list recall,14 after 2 weeks of high-frequency (1–20 Hz) rTMS over the left DLPFC.

Research performed in our laboratory suggests that one session of placebo-controlled high-frequency rTMS over the left DLPFC in medication-free patients with depression has a beneficial effect on task-switching performance but not on mood (unpublished data). We administered stimulation within the parameters used in recent clinical studies that reported an antidepressant outcome. As a result, one could suggest that cognitive changes emerge before an antidepressant effect of rTMS is observed and that later changes in depressive symptoms might be a secondary effect of this treatment. Previous studies have demonstrated no correlation between improvement in mood and improvements in executive function after a single session of high-frequency rTMS,15 after 5 sessions (in healthy volunteers),16 or after anodal trans-cranial direct current stimulation17 and rTMS18 (in depressed patients), all administered over the left DLPFC. However, when exploring the interplay between attentional and emotional function, it may be that very specific components of cognitive function should be investigated because selective attention and working memory are heterogeneous constructs. Therefore, it may be that immediate effects of rTMS on attentional control and mood should be investigated using a placebo-controlled design differentiating responders and non-responders.19 In the present study, we included a homogeneous group of medication-free, therapy-resistant patients with depression to investigate the influence of a single session of rTMS over the left DLPFC compared with 2 weeks of daily sessions.

Cognitive flexibility, which is required during task-switching, is an operationalization of attentional control, a core function of the DLPFC that is impaired among depressed patients.20 We extended our previous single-session rTMS study using a task-switching paradigm (unpublished data) to measure cognitive function before and after rTMS. We tested cognitive performance before treatment, after a single session and after 2 weeks of rTMS. In light of our previous results and those in the existing literature, we hypothesized that rTMS would have a significant beneficial effect on depressive symptoms only after 2 weeks of treatment and that positive effects on task-switching performance after 1 session of high-frequency rTMS would differentiate treatment responders and nonresponders. We defined responders as those showing a 50% or more reduction in their Hamilton Depression Rating Scale (HAM-D) scores after treatment. Furthermore, we predicted that after 2 weeks of rTMS cognitive improvements would be observed only in treatment responders, whereas the cognitive performance of nonresponders would not change.

## Methods

The present study was part of a larger project investigating the influence of rTMS on different neurocognitive markers. Data on 5 of our participants were also included in a previous study on the effects of 1 session of rTMS (unpublished data).

### Participants

We recruited a homogeneous group of right-handed patients with refractory depression from in- and outpatient facilities of the University Hospital of Brussels (UZ Brussel) between January 2005 and May 2007. The local Medical Ethics Committee of UZ Brussel approved our study protocol. After a full description of the experiment, we obtained written informed consent from all participants.

A detailed psychiatric examination along with the structured Mini International Neuropsychiatric Interview21 confirmed the diagnosis of a major depressive episode with melancholic features for all participants. We assessed treatment resistance using the Thase and Rush criteria.22

We weaned patients from their antidepressants or augmentation medications, including mood stabilizers or anti-psychotics, for at least 2 weeks before administering rTMS. If the patients were using fluoxetine, we required them to be free of antidepressant pharmacotherapy for at least 3 weeks. A trained psychiatrist (C.B.) closely monitored the patients during the weaning process. Once the patients had withdrawn, they remained on a steady dose of their somatic medication and, if necessary, on a steady dose of benzodiazepines. We used the following types and doses of benzodiazepines: tranxene (10 mg twice daily), alprazolam (1 mg twice daily), rivotril (0.5 mg 3 times daily) and zolpidem (10 mg once daily). We included only patients who did not need rescue medication or concomitant therapies during this period in our study, and we included patients only if this condition had been maintained for about 2 weeks. We allowed no change in any of their medications during the stimulation sessions. If patients required any changes in medication, we considered them to be dropouts.

All participants underwent a physical and neurologic examination (electroencephalography) and structural 3-dimensional magnetic resonance imaging (MRI) of the brain for nonstereotactic identification of the stimulation site (Brodmann area 9/46, left DLPFC). None of the patients had received rTMS before, and all met the safety criteria for rTMS.23 We included only patients whose scores on the 17-item HAM-D24 were 21 or higher. We excluded patients if they had any history of physical illness likely to affect brain physiology; head injury; comorbid psychiatric conditions, including alcohol or substance abuse; bipolar disorder; or contraindications to rTMS.

### Stimulation protocol

The rTMS stimulation parameters were well within the established safety guidelines.23 We performed magnetic stimulation using a MAGSTIM high-speed magnetic stimulator (Magstim Company Ltd.) combined with a figure-8-shaped coil.

On the first treatment trial, we established a stimulation intensity of 110% of the motor threshold of the right abductor pollicis brevis muscle using electromyography. We determined the precise DLPFC site using MRI, marked the skull and administered stimulation with the coil at a fixed position for all sessions of rTMS.

We delivered 10 sessions of high-frequency (10 Hz) rTMS daily from Monday to Friday over a period of 2 weeks. We used the following parameters for each session: 40 trains of 3.9 seconds’ duration separated by an intertrain interval of 26.1 seconds, resulting in 1560 pulses per session. The total stimulation time was about 20 minutes. At the start of their rTMS treatment, each patient also received 1 sham placebo-stimulated session. We used a crossover design to determine what type of stimulation would be administered during the first session (rTMS or sham). Sham stimulation was administered at the same location on the skull as rTMS, but the figure-8-shaped coil was held at a 90° angle, resting on the scalp with only 1 edge, in compliance with recent sham guidelines.25 During stimulation, all participants wore earplugs and blindfolds to guarantee “optimal” blinding.

### Clinical mood assessments

To evaluate antidepressant outcome, we assessed the efficacy of rTMS using the HAM-D scores at baseline and after 2 weeks. We considered patients to be responders to treatment if the HAM-D score after treatment had decreased by 50% or more from baseline.19 To evaluate immediate subjective mood changes, we asked patients to indicate their current mood states on visual analogue scales that included subscales for depression, anger, fatigue, vigour and tension. We asked participants to describe how they felt “at that moment” by indicating on horizontal 10-cm lines whether they experienced these 5 mood states on a scale from “totally not” to “very much.” We used the scales to record mood at various stages of the experiment: at baseline (Tpre), immediately after rTMS (Tpost), about 30 minutes after the first session (Tpost30) and at the end of the rTMS treatment period (Tposttreatment).

### Task-switching paradigm

Participants performed a computerized self-paced switching task. The device consisted of a board (connected to a personal computer) with 1 push button in the middle and 8 push buttons positioned in a semicircle around the central button. In addition, a speaker and a pedal were connected to the device.

The paradigm included 3 continuous task blocks in which the first 2 blocks consisted of repetitive tasks (1 with 28 visual stimuli and 1 with 28 auditory stimuli), whereas the last block consisted of switch trials. The 2 motor responses involved different stimulus modalities and, therefore, non-overlapping neural systems.

During the first block, we instructed patients to remove their fingers from the central push button and put out the illuminated push buttons. At each visual trial, 4 of the 8 push buttons could illuminate randomly, 1 at a time. For visual trials, we recorded 2 components of the reaction time: namely, initiation time and movement time. Initiation time, a central cognitive component, referred to the time required to initiate a response and corresponded to the time that lapsed between stimulus onset and the release of the middle push button. Movement time, a peripheral executive component, referred to the motor activity or the time required to complete the response.26

In the second block, we instructed participants to press a foot pedal when they heard a buzzer. In this task, only total reaction time could be recorded. We instructed participants to place their feet above the pedal and to respond to the signal as quickly and as accurately as possible.

The third block, the double task condition, was an alternating switch block with 29 auditory and 28 visual stimuli that were randomly mixed. During the switch block, 24 of the 29 auditory signals and 25 of the 28 visual signals were switch trials (we analyzed only the switch trials). Because of a technical problem, the first trials were lost and only 22 auditory and 22 visual switch trials could be recorded. We instructed patients to focus their attention on the visual stimuli and to switch attention when the auditory stimuli appeared. After each response to a visual trial, patients had to return their fingers to the central push button as quickly as possible, triggering stimulus onset asynchrony for the next trial. After every auditory signal, they had to take their feet off the pedal to trigger stimulus onset asynchrony for the next trial. In each of the 3 tasks, stimulus onset asynchrony differed randomly between 3000 ms and 6000 ms. We used the same trial sequence for all participants. If errors occurred, we replaced stimuli to obtain the same amount of correctly performed reactions for every participant. We removed movement times greater than 3000 ms (considered to be a delayed reaction time) and less than 200 ms (considered to be an anticipatory reaction time) from our analyses. We considered delayed or anticipatory reaction times for initiation time and auditory reaction times to be errors, and these were replaced by a new stimulus.

### Statistical analysis

We conducted a multivariate analysis of variance (MANOVA) to analyze short-term mood changes after a single session of high-frequency rTMS for mood scores, evaluated with 5 visual analogue subscales as dependent variables. We conducted a 2 × 3 × 2 mixed MANOVA with stimulation (rTMS, sham) and time (Tpre, Tpost, Tpost30) as within-subject factors and treatment response (responder/nonresponder) as a between-subject factor.

For task-switching analyses, we used 3 dependent variables for each analysis of variance (ANOVA): namely, the mean reaction time in milliseconds on visual (initiation time and movement time) and auditory (total reaction time) switch trials, corrected for individual processing speed. More specifically, we used the reaction time on visual and auditory switch trials from block 3 minus the reaction time on repetitive trials from block 2 and block 1, respectively.

For the expected acute effects of rTMS on task-switching results, we performed 3 sets of repeated-measures ANOVA to analyze the immediate effects of rTMS on cognition for all participants. The basic design was a 2 × 2 × 2 × 2 design with stimulation condition (rTMS, sham) and time (Tpre, Tpost30) as within-subject factors and treatment response (responders/nonresponders) and order (first sham/first rTMS) as between-subject factors.

Moreover, we performed 3 sets of mixed ANOVA to analyze the expected treatment effects of 2 weeks of rTMS on measures of cognitive processing. We used a 2 × 2 factorial design, including time (pre- to posttreatment) as a within-subject factor and treatment response (responders/non-responders) as a between-subject factor. For each participant, we based premeasures of cognitive processing on the first task-switching performance, which occurred before either rTMS or sham stimulation.

To further explore changes in cognitive function within groups of responders and nonresponders separately, we used Wilcoxon signed-rank nonparametric tests as conservatively as possible within the context of our small group sizes. Statistical significance was set at *p* < 0.05 for all analyses. We conducted statistical analyses using SPSS software, version 12.0 (SPSS Inc.).

## Results

### Participants

We included 15 patients with refractory depression in our study. The mean age of patients was 45.6 (range 22–61, standard deviation [SD] 5.9) years. All patients were right-handed, according to the van Strien Hand Preference scale.27 An overview of the demographic and clinical characteristics of all participants at baseline is presented in Table 1.

View this table:
[Table 1](http://jpn.ca/content/34/2/119/T1)

Table 1 
Demographic and clinical patient characteristics at baseline

In 10 patients, the depressive episode had been diagnosed for the first time more than 3 years before our study and was considered to be at least stage III treatment resistant: these patients had a minimum of 2 unsuccessful trials with selective serotonin or serotonin–norepinephrine reuptake inhibitor therapy and 1 failed trial with tricyclic antidepressants, as described by Thase and Rush.22 Five patients were considered to be stage V treatment resistant: they had unsuccessful trials with a monoamine oxidase inhibitor and a failed course of bilateral electroconvulsive therapy.

All patients tolerated the experimental procedure well; 2 participants reported a mild headache just after stimulation. We observed no difference in the Thase and Rush22 treatment-resistance score (*t* < 1) or in the types of anti-depressants used (χ23 = 1.473, *p* = 0.69) among rTMS responders and nonresponders.

### Effects of rTMS on depressive symptoms (treatment response)

Eight out of 15 patients (53%) reported a 50% reduction in their scores on the HAM-D after 2 weeks of rTMS (treatment responders). At baseline, the HAM-D scores of responders and nonresponders did not differ significantly (*z* = 0.005, *p* = 0.98). An overview of the group data is presented in Table 2.

View this table:
[Table 2](http://jpn.ca/content/34/2/119/T2)

Table 2 
HAM-D and BDI scores before and after rTMS treatment among responders and nonresponders

### Short-term effects of rTMS on mood and depressive symptoms

An overview of the visual analogue scale measures before (Tpre), immediately after (Tpost) and 30 minutes after (Tpost30) rTMS or sham stimulation are presented in Table 3. Because of some missing values on the mood scales, we removed 2 participants from the analysis.

View this table:
[Table 3](http://jpn.ca/content/34/2/119/T3)

Table 3 
Visual analogue scale measures before, immediately after and 30 minutes after rTMS or sham stimulation

The main effect for treatment response did not reach significance (*F*5,7 = 1.986, *p* = 0.18; stimulation *F*5,7 = 0.748, *p* = 0.61 and time *F*5,7 = 3.463, *p* = 0.17). We observed no 2-way interaction effects (*F* < 1.61), and the interaction effect between stimulation, time and treatment response yielded no significant effect: *F*5,7 = 0.304, *p* = 0.93. Thus there were no short-term mood changes from baseline caused by high-frequency rTMS applied over the left DLPFC when comparing ratings immediately after or 30 minutes after stimulation for responders and nonresponders.

### Effects of rTMS on task-switching: treatment response

Data on the effects of rTMS on task-switching are outlined in Table 4 and Table 5. As predicted, the 2-way interaction between time and treatment response was significant *F*1,13 = 6.191, *p* = 0.027 for initiation time in visual trials. The main effect of time was not significant (*F* < 1, *p* = 0.38), whereas the main effect of treatment response was significant (*F*1,13 = 9.134, *p* = 0.010). For movement time in visual trials, neither the main nor the interaction effect reached statistical significance (*F* < 1, *p* > 0.47).

View this table:
[Table 4](http://jpn.ca/content/34/2/119/T4)

Table 4 
Reaction time latencies in switch trials during task-switching with rTMS or sham stimulation in responders and nonresponders

View this table:
[Table 5](http://jpn.ca/content/34/2/119/T5)

Table 5 
Reaction time latencies in repetitive trials during task-switching with rTMS or sham stimulation in responders and nonresponders

As expected, the 2-way interaction between time and treatment response in auditory reaction trials was significant (*F*1,13 = 14.088, *p* = 0.002). The 2 main effects were not significant (*F* < 2.61, *p* > 0.51).

Among our group of responders, 2 weeks of rTMS significantly decreased initiation time (*z* = 2.240, *p* = 0.025) in visual trials and reaction time (*z* = 2.521, *p* = 0.012) in auditory trials. In contrast, among nonresponders cognitive function did not change for either component of visual and auditory reaction times (*z* < 1.014, *p* > 0.26). Most importantly within the context of this design, reaction times of responders and non-responders did not significantly differ during their first task-switching performance (initiation time *z* = 1.154, *p* = 0.25; movement time *z* = 0.694, *p* = 0.49; auditory reaction time *z* = 0.581, *p* = 0.33). Moreover, after rTMS there was a significant difference between responders and nonresponders in reaction times on initiation time (*z* = 2.199, *p* = 0.029) and auditory reaction times (*z* = 2.315, *p* = 0.021), but not on movement time (*z* = 0.347, *p* = 0.78).

We tested the same factorial design for repetitive trials. However, no main or interaction effects reached statistical significance (*F* < 1.2, *p* > 0.29).

### Short-term effects of rTMS on task-switching

Because the factor order was not implicated in any main or interaction effect (*F* < 1.4), we left this factor out of all subsequent analyses.

As predicted, the key 3-way interaction among treatment response, stimulation condition and time was significant for short-term effects of rTMS on initiation time in visual reaction times (*F*1,13 = 5.369, *p* = 0.039). The main effect of time (*F*1,13 = 6.105, *p* = 0.029) and the 2-way interaction between stimulation condition and time (*F*1,13 = 4.976, *p* = 0.046) also reached statistical significance. The other main or 2-way interactions yielded no significant effects (*F* < 2.98, *p* > 0.11).

Our ANOVA of movement time in visual trials yielded no main or interaction effects (*F* < 1.18). The 3-way interaction (*F*1,13 = 0.003, *p* = 0.96), indicating no effects of rTMS or sham stimulation on this component of visual reaction time among responders and nonresponders alike.

For auditory reaction times, the key 3-way interaction (*F*1,13 = 3.308, *p* = 0.09) yielded no significant effects. Moreover, the main or interaction effects were not significant (*F* < 1, *p* > 0.48), indicating no effects of rTMS or sham stimulation on this component.

We used Wilcoxon signed-rank nonparametric tests to further explore the interaction effects among responders. We found that, in contrast to sham stimulation (*z* = 0.420, *p* = 0.67), reaction times for initiation time significantly decreased after rTMS (*z* = 2.308, *p* = 0.017). We observed no changes among nonresponders (*z* < 0.943, *p* > 0.34). Moreover, we found a significant difference in initiation time between changes after rTMS compared with sham stimulation among responders, with the change after rTMS being greater (z = 2.100, *p* = 0.036). Among nonresponders, the differences in changes before and after rTMS compared with sham stimulation did not reach significance (*z* = 1.840, *p* = 0.06). To sort out a nonspecific influence of rTMS on cognitive function, we tested the same factorial design for repetitive trials. However, no main or interaction effects reached statistical significance (*F* < 1.3, *p* > 0.58).

## Discussion

Evidence suggesting that rTMS might be a promising treatment procedure with rapid onset of action has aroused growing interest. However, inconsistent research findings on rTMS treatment outcomes indicate that further investigation of the possible underlying mechanisms and the characteristics of treatment responders is warranted.29

We therefore performed rTMS in a homogeneous research group of therapy-resistant medication-free patients with depression. We administered rTMS over the left DLPFC, which is a typical stimulation target in mood disorders, and measured depressed mood symptoms and attentional control before rTMS, after 1 session and after 10 sessions.

Subjective mood reports did not change after a single session of high-frequency rTMS. We measured positive treatment outcomes using the HAM-D. Although our sample was small, more then half (53%) of our therapy-resistant population experienced a decrease of at least 50% in their HAM-D scores after 2 weeks of daily high-frequency rTMS over the left DLPFC.

Among responders, we observed a significant improvement in attentional control after 1 double-blind placebo-controlled rTMS session. We demonstrated that a specific component of visual switch trials (i.e., initiation time) had improved after high-frequency rTMS. Movement time in visual switch trials and auditory switch trials (i.e., the motor component of switching) was not influenced by rTMS. Although it is known that both reaction and movement times are prolonged among patients with severe depression, it appears that 2 weeks of rTMS does not influence motor aspects (movement time) of attentional control. These current findings are in accordance with the previous research conducted in our laboratory, indicating that a single session of rTMS does not affect movement time in healthy volunteers30,31 or in depressed patients (unpublished data).

The enhancement of attentional control after a single session of placebo-controlled high-frequency rTMS over the left DLPFC is in line with the involvement of the DLPFC in the representation of goal-directed behaviour. The DLPFC has been shown to play a key role in sustaining task-relevant representations of nonemotional and emotional stimuli to accomplish the task goals.9

The acute effect of rTMS on cognition in depressed patients that we observed is in line with the results of 1 previous study,7 but contrasts those of another study.32 Compared with the latter experimental study, we used a relatively intensive treatment protocol (10 Hz, 110% of motor threshold, 1560 pulses per session), which may account for the rapid attentional control improvements. On the other hand, because the attentional control changes emerged in the group of treatment responders only when no mood effects had as yet occurred, cognitive effects are specific and interpretable, holding intriguing implications for clinical practice. Although recent sham-controlled studies used more aggressive treatment parameters (e.g., ≥ 110% motor threshold; > 10 sessions4,5) than we did, we also observed effects in a comparable depressed population after 2 weeks of treatment. Nevertheless, because in the present study we did not include a control group that received 2 weeks of sham stimulation, it might be that patients who showed cognitive changes after 1 session of rTMS are participants with the highest placebo response after 2 weeks of treatment. Although such a placebo response is not likely to occur among treatment-resistant patients, the only way to rule out the possibility that cognitive changes might be related to a placebo effect is to have a control group receiving 10 days of sham stimulation. Moreover, it has been shown recently that 1–4 additional weeks of rTMS seems to increase the response rate,33 which might decrease the predictive value of cognitive tasks.

Most importantly, the current results highlight a link between higher attentional and emotional information processing, these 2 processing pathways being part of the same dorsolateral cortico–subcortical network. Neuroimaging studies are gradually beginning to unravel the underlying dynamics of these affective processing pathways. Reduced left DLPFC function within a fronto–subcortical network is thought to play a pivotal role in the pathophysiology of depression and to connect emotional and cognitive information processing.34 Importantly, neuroimaging research faces the critical problem of understanding cause and effect, whereas rTMS can demonstrate causal relations in information processes. However, in psychiatry, rTMS has been studied mainly as a potential antidepressant treatment. Only limited research has been reported on cognitive effects of rTMS immediately after termination of stimulation, probably because cognitive effects have been investigated mainly to evaluate aspects of treatment safety.15 To our knowledge, no other controlled studies have demonstrated the role of specific attentional control processes on treatment outcome (unpublished data). However, prior studies have investigated acute changes after rTMS that may trigger an antidepressant outcome.

Wagner and coworkers15 investigated the effects of 1 session of rTMS on healthy volunteers using a divided attention task. They proposed that the acute effects of rTMS on cognition may be produced by identical neurochemical changes that underlie successful antidepressant treatments. However, the fact that the study was based on a group of healthy volunteers made it difficult to link attentional changes during the hours after an rTMS session with the neurochemical changes underlying the antidepressant treatment.

Shajahan and colleagues35 demonstrated an increased perfusion in the anterior cingulate cortex after the first rTMS session in depressed patients, but reported no improvements in verbal fluency. However, verbal fluency might not be the best choice to represent processes associated with the DLPFC and the left dorsolateral circuit. In line with our results, a study by Möller and coleagues36 found that the P300, a major endogenous brain event-related potential component that has been found to be reduced in patients with depression, significantly increased in amplitude after rTMS over the left prefrontal cortex compared with sham stimulation. An increase in P300 amplitude is indicative of improved attentiveness.36,37 Most importantly, no significant antidepressant effects were observed in this study after 5 days of stimulation. Finally, Bermpohl and colleagues7 suggested that the affective go/no-go task, may be employed for its ability to predict an rTMS treatment response. Given the linkage between emotional and cognitive functions, the latter affective/cognitive task might serve as a rough indicator for clinical treatment response. In the present study, we used a nonemotional task focusing on attentional goal representations, and our results suggest that task-switching performance might be a promising predictor for the response of patients to multiple sessions of rTMS therapy. Moreover, higher attentional control appears to play an important role in the progress of mood disorders.

### Limitations

A number of important shortcomings deserve to be highlighted. Our results should be interpreted as preliminary because our sample was rather small. In addition, the absence of a control condition must be taken into account when interpreting the effects of rTMS that we observed. Another limitation of our study was the sham stimulation condition: despite the sham stimulation being performed at a 90° angle for minimal stimulation of the DLPFC, it is possible that the stimulation was partially active.25 In addition, the cognitive task that we used has not been tested for reliability and validity. Finally, benzodiazepines may impair an attentional task, but they may improve scores on some subscales of the HAM-D. Therefore, we do not believe that this interfered with our results. Moreover, if necessary, patients received a steady dose of their benzodiazepines for a reasonable time before entering the study, and we permitted no temporary changes during the stimulation protocol.

In sum, our results suggest that rTMS primarily influences higher attentional function. However, future research combining rTMS with functional brain imaging is necessary to provide further evidence of these cognitive changes as a marker of antidepressant effects.

## Acknowledgements

This research was supported by a grant from the Scientific Fund W. Gepts UZ-Brussels and funded by Grant BOF-01J08107 from Ghent University. The authors would like to thank Mark Moens and Marc Vercauteren for the development of the computer version of the task-switching paradigm ([www.subliemservice.be](http://www.subliemservice.be)). The authors also wish to acknowledge the contribution of Prof. Hugo D’haenen for his assistance with the design of the study and the recruitment of the sample of depressive patients and to express our sorrow at his death.

## Footnotes

*   **Competing interests:** None declared.

*   **Contributors:** All authors contributed to the study design and review of the article and all gave final approval for publication. Drs. Vanderhasselt, Leyman and Baeken acquired the data. Drs. Vanderhasselt and De Raedt analyzed the data and wrote the article.

*   Received November 7, 2007.
*   Revision received March 26, 2008.
*   Revision received June 9, 2008.
*   Revision received June 11, 2008.
*   Accepted June 12, 2008.

## References

1.  Loo CK, Mitchell PB. A review of the efficacy of transcranial magnetic stimulation (TMS) treatment for depression, and current and future strategies to optimize efficacy. J Affect Disord 2005;88:255–67.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.jad.2005.08.001&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16139895&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000233225400002&link_type=ISI) 

2.  Gershon AA, Dannon PN, Grunhaus L. Transcranial magnetic stimulation in the treatment of depression. Am J Psychiatry 2003; 160:835–45.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1176/appi.ajp.160.5.835&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12727683&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000182610400004&link_type=ISI) 

3.  Martin JLR, Barbanoj MJ, Schlaepfer TE, et al. Repetitive transcranial magnetic stimulation for the treatment of depression: systematic review and meta-analysis. Br J Psychiatry 2003;182:480–91.
    
    [Abstract/FREE Full Text](http://jpn.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MTA6ImJqcHJjcHN5Y2giO3M6NToicmVzaWQiO3M6OToiMTgyLzYvNDgwIjtzOjQ6ImF0b20iO3M6MTg6Ii9qcG4vMzQvMi8xMTkuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

4.  Avery DH, Holtzheimer PE, Fawaz W, et al. A controlled study of repetitive transcranial magnetic stimulation in medication-resistant major depression. Biol Psychiatry 2006;59:187–94.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.biopsych.2005.07.003&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16139808&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000234871600012&link_type=ISI) 

5.  O’Reardon JP, Solvason HB, Janicak PG, et al. Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial. Biol Psychiatry 2007;62:1208–16.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.biopsych.2007.01.018&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17573044&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000251119600003&link_type=ISI) 

6.  Burt T, Lisanby SH, Sackeim HA. Neuropsychiatric applications of transcranial magnetic stimulation: a meta analysis. Int J Neuropsychopharmacol 2002;5:73–103.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1017/S1461145702002791&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12057034&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

7.  Bermpohl F, Fregni F, Boggio PS, et al. Effect of low-frequency transcranial magnetic stimulation on an affective go/no-go task in patients with major depression: role of stimulation site and depression severity. Psychiatry Res 2006;141:1–13.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.psychres.2005.07.018&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16352348&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000234931600001&link_type=ISI) 

8.  Mayberg HS, Liotti M, Brannan SK, et al. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 1999;156:675–82.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1176/ajp.156.5.675&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10327898&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000080095400003&link_type=ISI) 

9.  Taylor JG, Fragopanagos NF. The interaction of attention and emotion. Neural Netw 2005;18:353–69.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.neunet.2005.03.005&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=15921888&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000230710900003&link_type=ISI) 

10. Vanderhasselt MA, De Raedt R, Baeken C, et al. The influence of rTMS over the right dorsolateral prefrontal cortex on top-down attentional processes. Brain Res 2007;1137:111–6.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.brainres.2006.12.050&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17229406&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000244849300013&link_type=ISI) 

11. Januel D, Dumortier G, Verdon CM, et al. A double-blind sham controlled study of right prefrontal repetitive transcranial magnetic stimulation (rTMS): therapeutic and cognitive effect in medication free unipolar depression during 4 weeks. Prog Neuropsychopharmacol Biol Psychiatry 2006;30:126–30.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.pnpbp.2005.08.016&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16242826&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

12. Padberg F, Zwanzger P, Thoma H, et al. Repetitive transcranial magnetic stimulation (rTMS) in pharmacotherapy-refractory major depression: comparative study of fast, slow and sham rTMSI. Psychiatry Res 1999;88:163–71.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0165-1781(99)00092-X&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10622338&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000084152400001&link_type=ISI) 

13. Triggs WJ, McCoy KJM, Greer R, et al. Effects of left frontal trans-cranial magnetic stimulation on depressed mood, cognition, and corticomotor threshold. Biol Psychiatry 1999;45:1440–6.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0006-3223(99)00031-1&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10356626&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000080504600007&link_type=ISI) 

14. Little JT, Kimbrell TA, Wassermann EM, et al. Cognitive effects of 1-and 20-Hertz repetitive transcranial magnetic stimulation in depression: preliminary report. Neuropsychiatry Neuropsychol Behav Neurol 2000;13:119–24.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10780630&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000167304300006&link_type=ISI) 

15. Wagner M, Rihs TA, Mosimann UP, et al. Repetitive transcranial magnetic stimulation of the dorsolateral prefrontal cortex affects divided attention immediately after cessation of stimulation. J Psychiatr Res 2006;40:315–21.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.jpsychires.2005.06.001&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16055155&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000238110200005&link_type=ISI) 

16. Moser DJ, Jorge RE, Manes F, et al. Improved executive functioning following repetitive transcranial magnetic stimulation. Neurology 2002;58:1288–90.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1212/WNL.58.8.1288&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=11971103&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

17. Fregni F, Marcolin MA, Myczkowski M, et al. Predictors of anti-depressant response in clinical trials of transcranial magnetic stimulation. Int J Neuropsychopharmacol 2006;9:641–54.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1017/S1461145705006280&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16939662&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

18. Boggio PS, Fregni F, Bermpohl F, et al. Effect of repetitive TMS and fluoxetine on cognitive function in patients with Parkinson’s disease and concurrent depression. Mov Disord 2005;20:1178–84.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1002/mds.20508&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=15895421&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

19. Brakemeier EL, Luborzewski A, Danker-Hopfe H, et al. Positive predictors for antidepressive response to prefrontal repetitive transcranial magnetic stimulation (rTMS). J Psychiatr Res 2007;41:395–403.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.jpsychires.2006.01.013&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16554071&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000244583900003&link_type=ISI) 

20. Austin MP, Mitchell P, Goodwin GM. Cognitive deficits in depression — Possible implications for functional neuropathology. Br J Psychiatry 2001;178:200–6.
    
    [Abstract/FREE Full Text](http://jpn.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MTA6ImJqcHJjcHN5Y2giO3M6NToicmVzaWQiO3M6OToiMTc4LzMvMjAwIjtzOjQ6ImF0b20iO3M6MTg6Ii9qcG4vMzQvMi8xMTkuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

21. Sheehan DV, Lecrubier Y, Sheehan KH, et al. (1998). The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59 (Suppl 20):22–33;quiz 34–57. M.I.N.I. [Dutch version by Van Vliet, Leroy and Van Megen, 2000].
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0924-9338(99)80239-9&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=9881538&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

22. 1.  Bloom FE, 
    2.  Kupfer DJ
    
    Thase ME, Rush AJ. Treatment resistant depression. In: Bloom FE, Kupfer DJ, editors. Psychopharmacology: the fourth generation of progress. New York (NY): Raven Press; 1995. p. 1081–97.
    
    

23. Wassermann EM. Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines from the international workshop on the safety of repetitive transcranial magnetic stimulation, June 5–7, 1996. Electroencephalogr Clin Neurophysiol 1998;108:1–16.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0168-5597(97)00096-8&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=9474057&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

24. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56–62.
    
    [FREE Full Text](http://jpn.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6MzoiUERGIjtzOjExOiJqb3VybmFsQ29kZSI7czo0OiJqbm5wIjtzOjU6InJlc2lkIjtzOjc6IjIzLzEvNTYiO3M6NDoiYXRvbSI7czoxODoiL2pwbi8zNC8yLzExOS5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

25. Loo CK, Taylor JL, Gandevia SC, et al. Transcranial magnetic stimulation (TMS) in controlled treatment studies: Are some “sham” forms active? Biol Psychiatry 2000;47:325–31.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0006-3223(99)00285-1&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10686267&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000085409700007&link_type=ISI) 

26. Gorus E, De Raedt R, Lambert M, et al. Attentional processes discriminate between patients with mild Alzheimer’s disease and cognitively healthy elderly. Int Psychogeriatr 2006;18:539–49.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1017/S1041610205002723&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16472408&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

27. van Strien JW. Handvoorkeur en taaldominantie. Neuropraxis 2001; 2:10–5.
    
    

28. Beck AT, Ward CH, Mendelson M, et al. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:53–63.
    
    

29. Kimbrell TA, Little JT, Dunn RT, et al. Frequency dependence of antidepressant response to left prefrontal repetitive transcranial magnetic stimulation (rTMS) as a function of baseline cerebral glucose metabolism. Biol Psychiatry 1999;46:1603–13.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0006-3223(99)00195-X&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10624541&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000084335400002&link_type=ISI) 

30. Vanderhasselt MA, De Raedt R, Baeken C, et al. The influence of rTMS over the left dorsolateral prefrontal cortex on Stroop task performance. Exp Brain Res 2006;169:279–82.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1007/s00221-005-0344-z&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16418843&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000235249100016&link_type=ISI) 

31. Vanderhasselt MA, De Raedt R, Baeken C, et al. A single session of rTMS over the left dorsolateral prefrontal cortex influences attentional control in depressed patients. World J Biol Psychiatry 2007 Dec 17:1–9 [Epub ahead of print].
    
    

32. Loo C, Mitchell P, Sachdev P, et al. Double-blind controlled investigation of transcranial magnetic stimulation for the treatment of resistant major depression. Am J Psychiatry 1999;156:946–8.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1176/ajp.156.6.946&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10360138&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000080698500023&link_type=ISI) 

33. Avery DH, Isenberg KE, Sampson SM, et al. Transcranial magnetic stimulation in the acute treatment of major depressive disorder: clinical response in an open-label extension trial. J Clin Psychiatry 2008;69:441–51.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.4088/JCP.v69n0315&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=18294022&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

34. Drevets WC, Bogers W, Raichle ME. Functional anatomical correlates of antidepressant drug treatment assessed using PET measures of regional glucose metabolism. Eur Neuropsychopharmacol 2002;12:527–44.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0924-977X(02)00102-5&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12468016&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000179919200005&link_type=ISI) 

35. Shajahan PM, Glabus MF, Steele JD, et al. 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 2002;26:945–54.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0278-5846(02)00210-5&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12369271&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

36. Möller AL, Hjaltason O, Ivarsson O, et al. The effects of repetitive transcranial magnetic stimulation on depressive symptoms and the P-300 event-related potential. Nord J Psychiatry 2006;60:282–5.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1080/08039480600790119&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16923636&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 

37. Picton TW. The P300 wave of the human event-related potential. J Clin Neurophysiol 1992;9:456–79.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1097/00004691-199210000-00002&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=1464675&link_type=MED&atom=%2Fjpn%2F34%2F2%2F119.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=A1992KD90100002&link_type=ISI)