Modafinil improves information processing speed and increases energetic resources for orientation of attention in narcoleptics: Double-blind, placebo-controlled ERP studies with low-resolution brain electromagnetic tomography (LORETA)

Sleep Medicine 10 (2009) 850–858

Contents lists available at ScienceDirect

Sleep Medicine journal homepage: www.elsevier.com/locate/sleep

Original Article

Modafinil improves information processing speed and increases energetic resources for orientation of attention in narcoleptics: Double-blind, placebo-controlled ERP studies with low-resolution brain electromagnetic tomography (LORETA) Michael Saletu b,*, Peter Anderer a, Gerda Maria Saletu-Zyhlarz a, Magdalena Mandl a, Bernd Saletu a, Josef Zeitlhofer b a b

Department of Psychiatry and Psychotherapy, Medical University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria Department of Neurology, Medical University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria

a r t i c l e

i n f o

Article history: Received 9 December 2007 Received in revised form 20 December 2008 Accepted 23 December 2008 Available online 1 April 2009 Keywords: Narcolepsy Event-related potentials LORETA Cognition Attention Modafinil Prefrontal cortex

a b s t r a c t Background and purpose: Recent neuroimaging studies in narcolepsy discovered significant gray matter loss in the right prefrontal and frontomesial cortex, a critical region for executive processing. In the present study, event-related potential (ERP) low-resolution brain electromagnetic tomography (LORETA) was used to investigate cognition before and after modafinil as compared with placebo. Patients and methods: In a double-blind, placebo-controlled cross-over design, 15 patients were treated with a 3-week fixed titration scheme of modafinil and placebo. The Epworth Sleepiness Scale (ESS), Maintenance of Wakefulness Test (MWT) and auditory ERPs (odd-ball paradigm) were obtained before and after the 3 weeks of therapy. Latencies, amplitudes and LORETA sources were determined for standard (N1 and P2) and target (N2 and P300) ERP components. Results: The ESS score improved significantly from 15.4 (± 4.0) under placebo to 10.2 (± 4.1) under 400 mg modafinil (p = 0.004). In the MWT, latency to sleep increased nonsignificantly after modafinil treatment (11.9 ± 6.9 versus 13.3 ± 7.1 min). In the ERP, N2 and P300 latencies were shortened significantly. While ERP amplitudes showed only minor changes, LORETA revealed increased source strengths: for N1 in the left auditory cortex and for P300 in the medial and right dorsolateral prefrontal cortex. Conclusion: LORETA revealed that modafinil improved information processing speed and increased energetic resources in prefrontal cortical regions, which is in agreement with other neuroimaging studies. Ó 2009 Elsevier B.V. All rights reserved.

1. Introduction Neuroimaging in sleep medicine has provided an important non-invasive investigational tool of structural changes, as well as indirect functional measures based on blood flow, metabolic activities and neurotransmission [1]. However, structural neuroimaging studies with voxel-based morphometry (VBM) in narcolepsy, a rare disabling sleep disorder characterized by excessive daytime sleepiness and abnormal rapid-eye-movement (REM) sleep manifestations, yielded controversial findings [1]. Whereas some studies reported hypothalamic gray matter changes [2] as well as inferior temporal and frontal atrophy [3], no differences between patients and controls were found in another study [4]. In the most recent VBM study in narcolepsy, significant gray matter loss was observed in the right prefrontal and frontomesial cortex, indicating a dis* Corresponding author. Tel.: +43 1 40 400 3683; fax: +43 1 40 259 09. E-mail address: [email protected] (M. Saletu). 1389-9457/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.sleep.2008.12.005

ease-related atrophy pattern, which may also be responsible for the reported prefrontal dysfunctions [5]. Functional imaging studies using functional MRI have not been able to provide a clear model of alterations in regional cerebral function either [6]. In the last decade, electrophysiological neuroimaging techniques, such as EEG low-resolution brain electromagnetic tomography (LORETA), were developed in order to identify brain regions that are involved in neuropsychiatric disorders and are the targets of therapeutic drug action [7–9]. LORETA is a novel electrophysiological neuroimaging method first presented by Pascual-Marqui and co-workers in 1994, which computes a unique 3dimensional electrical source distribution by assuming that the smoothest of all possible inverse solutions is most plausible, which is consistent with the assumption that neighboring neurons are simultaneously and synchronously active [10]. Generators of electric brain activity can therefore be localized utilizing the Talairach human brain atlas [11]. In our recent EEG-tomographic studies with LORETA on vigilance differences between narcolepsy patients

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

and controls we observed a significant decrease in a-2 power, mainly in the frontal, temporal and parietal cortices of the right hemisphere, along with a global decrease in beta power, also accentuated over the right cortical brain areas [9], which confirms a deterioration of the fronto-temporo-parietal network of the ‘‘right hemisphere vigilance system” [12] in narcolepsy. In order to bridge the currently wide gap between the neurotransmitter and the behavioral level and to improve temporal resolution up to the millisecond level it may be useful to study the various components of event-related brain potentials (ERPs). These include the N1 and P2 components, which reflect fundamental aspects of perception such as initial orienting or attention-directing activity in the primary sensory projection areas [13,14], and N2 and P300 components, which reflect fundamental aspects of cognition such as the allocation of attentional resources for the evaluation of possible targets as well as the allocation of resources for stimulus encoding [15,16]. So far, P300 investigations in narcolepsy patients have yielded diverging results [17–19]. Modafinil is a new centrally wake-promoting psychostimulant with a low risk of CNS, cardiovascular or gastrointestinal adverse events, abuse and dependence [20]. The drug appears to have multiple effects on catecholamine systems in the brain, including DAT (dopamine transporter) and NET (norepinephrine transporter) inhibition, thereby being relatively selective for cortical over subcortical areas [21]. The efficacy, safety and tolerance of modafinil in narcolepsy patients have been demonstrated in controlled trials [22,23]. In our narcolepsy studies, modafinil (as compared with placebo) increased fast alpha-2 power in the frontotemporal and sub-lobar cortical regions and beta-1-3 power in the temporoparietal and limbic cortices, predominantly in the left hemisphere, which was less affected by the disease [9]. Cognitive dysfunction may be a particularly important treatment target of modafinil–not only in narcolepsy, but also in other neuropsychiatric disorders [21]. There is now increasing evidence that the drug is able to improve cognitive function, particularly working memory, episodic memory, and processes requiring cognitive control [21]. A multicenter randomized, double-blind, placebo-controlled 12-week study in narcolepsy found armodafinil, a single-isomer formulation of modafinil, to have several effects on cognition, as revealed by reaction time, episodic recall and recognition tasks [24]. Studies in animal models and neuroimaging in humans suggest that these effects may be related to specific actions of modafinil in the frontal cortex [21]. Unfortunately, there is a paucity of empirical studies objectively assessing and localising the cognitive effects of modafinil in narcolepsy. Thus, the aim of the present double-blind, randomized, placebocontrolled study was to investigate cognition before and after modafinil as compared with placebo, utilizing LORETA for identifying generators of event-related potential (ERP) components in cortical regions.

2. Methods 2.1. Patients, inclusion and exclusion criteria Fifteen patients (7 males and 8 females; aged 38 ± 18 years, all right-handed) with the ICD-10 diagnosis of narcolepsy (G 47.4) were included in the study. Inclusion criteria called for patients satisfying the diagnostic criteria of the International Classification of Sleep Disorders (ICSD-1) for narcolepsy [25]. Symptoms were required to have been stable for 2 weeks before the beginning of the study. All patients were suffering from EDS and recurrent daytime naps. The mean age of onset of EDS was 14 years (range 2–50). The mean

851

Epworth Sleepiness Scale score was 17.3 (±4.0), which suggested moderate to severe EDS. Ten of the 15 patients had a clear cataplexy. A statistical subanalysis was calculated for this narcolepsy-cataplexy subgroup. Twelve patients experienced hypnagogic hallucinations and 6 sleep paralysis. Screened patients complaining of EDS underwent neuropsychiatric, physical and laboratory examinations (including HLA typing for DQB1*0602), an overnight polysomnography and a Multiple Sleep Latency Test (MSLT). Polysomnography revealed a mean sleep latency of 4.6 (±3.1) min, a mean REM sleep latency of 27.0 (±41.1) min and a mean sleep efficiency index of 85% (±15.8). The mean apnea-hypopnea index was 4.2 (±4.5)/h of sleep, which excluded an obstructive sleep apnea syndrome as a possible cause of EDS. The mean arousal index was 15.9 (±10.0)/h of sleep; the PLM arousal index 1.6 (±2.8). In the MSLT, mean sleep latency was 4.3±1.9 min (range 1.33–7) and the number of sleep onset REM periods (SOREMPs) was at least 2 per patient out of 5 nap trials. HLA typing for DQB1*0602 was positive in 12 patients. The following groups were excluded from the study: patients with evidence of a medical or psychiatric disorder that might account for the primary complaint; patients with sleep apnea, restless legs syndrome (RLS) or periodic limb movement disorder; pregnant or lactating women; patients with a history of drug abuse or dependency, including alcohol; patients requiring psychoactive medication or unwilling to temporarily discontinue anticataplectic medication or any other drug that might interfere with the study assessments; patients who were unable or unwilling to comply with the protocol; patients who worked at night. The study was performed in accordance with the relevant guidelines of the Declaration of Helsinki, 1964, as amended in Tokyo, 2004. Written informed consent was obtained. The study protocol was approved by the Ethics Committee of the Medical University of Vienna and the General Hospital of the City of Vienna. 2.2. Study design The study was designed as a 3-week, double-blind, randomized, placebo-controlled cross- over trial with the two treatment periods separated by a washout phase of one week. The randomization list was generated by an individual who was operationally independent from the study personnel who executed the randomization assignment. Two glasses with medication were prepared for each patient, marked with A (for the first treatment phase) and B (for the second treatment phase). The code was broken only after the end of the study. 2.3. Drug administration One hundred mg modafinil and placebo were prepared in capsules that looked identical. According to the fixed titration schedule, patients were treated as follows: Week 1: one capsule in the morning and one capsule at midday; week 2: two capsules in the morning and one capsule at midday; week 3: two capsules in the morning and two capsules at midday. At the beginning and at the end of each 3-week treatment block, ERP and MWT were performed and the ESS was completed. 2.4. Sleep log and Epworth Sleepiness Scale In addition, to control for variable sleep-wake schedules and effects of sleep loss, patients kept a simple sleep log in which the estimated amount of sleep in the nights preceding the study evaluation days was documented. The Epworth Sleepiness Scale (ESS) [26], which covers subjective sleepiness during the last week, was completed as well at each visit.

852

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

2.5. Maintenance of Wakefulness Test (MWT) To evaluate the effect of modafinil on the capacity to remain awake, the Maintenance of Wakefulness Test (MWT 20) was performed using the standardized version described by Mitler et al. [27]. Sleep recordings comprised 3 EEG channels (C4-A1, Cz-O2, and C3-A2) according to the international 10/20 system, 2 electro-oculogram (EOG) channels (left/right), and submental electromyogram (EMG). For sleep staging, 30-second epochs were visually scored according to the criteria of Rechtschaffen and Kales [28]. Four runs with a duration of 20 min were performed at 10 AM, 12 AM, 2 PM and 4 PM. For each run, the patient was seated in a comfortable, slightly reclining chair in a dark room and was instructed to remain awake. When definite signs of sleep where observed for at least one epoch in the PSG, the trial was stopped. Sleep onset was defined as one 30 s epoch of stage 1 sleep or any other sleep stage. 2.6. Recording and analysis of ERPs ERP recordings were not performed on the same day as the MWT to avoid interference of the two vigilance test procedures. Subjects were instructed to sit comfortably in a reclining chair situated in a constantly lit, sound-attenuated Faraday’s cage, to open their eyes and focus on a point 2.5 m straight ahead. A background noise (white noise of 45 dB SPL) was presented via earphones. The experiment started with a 1 min resting recording with eyes open. The two-tone oddball-paradigm applied in this study was based on a design first reported by Squires et al. [29] and adapted for longitudinal studies by Semlitsch et al. [30]. Loud tone bursts (standard tone, probability 0.9, 1000 Hz, 50 ms dura-

tion, 5 ms rise/fall time, 90 dB SPL) were presented binaurally with a constant inter-stimulus interval of 1 s, randomly interrupted by soft tone bursts (target tone, probability 0.1, 1000 Hz, 50 ms duration, 5 ms rise/fall time, 70 dB SPL). The total number of tone bursts presented was 330. The subjects were asked to mentally count the soft tone bursts and report their number at the end of the experiment. The error rate was calculated as the absolute difference between the number of target tones reported by the subject and the number of target tones actually presented. Gold electrodes were attached to the scalp according to the international 10/20 system. Twenty-one channels, including 2 EOG and 19 EEG channels (Fp1, Fp2, F7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3, Pz, P4, T6, O1 and O2 to averaged mastoids), were recorded by means of a Nihon Kohden 4421 G polygraph. Due to the high number of artifacts, the frontopolar electrodes were excluded from the analysis. The vertical EOG was recorded from an electrode at the mid-forehead to the average of one electrode below the left and one below the right eye. The horizontal EOG was recorded from the outer canthi (time constant: 1.0 s; high frequency response: 70 Hz; frequency range: 0.16–70 Hz; amplification: approximately 1:20000 and 1:4000 for EEG and EOG, respectively). On-line data acquisition as well as data analyses were performed on a personal computer system using our software package [30]. The sampling frequency was 256 Hz, resulting in a time resolution of approximately 4 ms. After minimizing ocular artifacts based on regression analysis in the time domain [31,32] and automatic artifact elimination [33], averaged ERP wave shapes were computed for target and standard tones (Fig. 1). The analyzed time epoch was 1 s, including 100 ms pre-stimulus time. Waveforms were digitally filtered by means of a phase-linear, low-pass filter with a cut-off frequency of 20 Hz.

Fig. 1. Grand mean standard (blue) and target tone (red) ERP waveforms in patients with narcolepsy with N1, P2, N2 and P300 components(n: 15) . In addition to 19 EEG channels, the vertical EOG (VEOG) and the horizontal EOG (HEOG) are shown. Note that the waveforms have been corrected for EOG interference.

853

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

Latencies and amplitudes of standard tone N1 and P2 as well as target tone N2 and P300 components were determined by the following procedure: Peak latencies of the spatial average waveforms (i.e., average across 17 leads) were determined automatically and checked visually. If necessary, target tone N2 latency was determined with the aid of the difference waveform (target tone ERPstandard tone ERP). The amplitudes of the ERP components were measured at the defined latencies relative to pre-stimulus baseline (0–100 ms prior to stimulus onset). Thus, the component amplitudes measured at the different leads represent the topographic distribution at one single time point. Subsequently, on the basis of the scalp-recorded electric potential distribution, LORETA was used to estimate the 3-dimensional intracerebral current density distribution in 2394 voxels with a spatial resolution of 0.343 cm3 [7]. LORETA solves the non-unique ‘‘inverse” problem (i.e., the computation of the electric sources from surface data) by assuming that the smoothest of all possible source distributions is most plausible. This assumption is consistent with known electrophysiology showing highly correlated activity in neighboring neuronal populations. The applied LORETA version introduces, in addition to the smoothness constraint, a neuroanatomical constraint by restricting the solution space to cortical gray matter volume, as determined by the digitized Probability Atlas (Brain Imaging Centre, Montreal Neurologic Institute) in the Talairach domain [11]. A voxel was labeled as gray matter if its probability of being gray matter was higher than 33% and higher than its probability of being white matter or cerebrospinal fluid (for details see [7]). Grand mean LORETA images were inspected for local maxima. Thus, in the first step, artifact-free standard and target responses were averaged separately for each subject and condition. Thereafter, N1 and P2 peak latencies for standard ERPs and N2 and P300 peak latencies for target ERPs were determined. Next, the amplitude distribution at the ERP peak latency was subjected to the LORETA analysis, resulting in 2394 current density values for each subject, condition and component. Finally, these values underwent statistical evaluation, i.e., computation of group means and voxel-by-voxel t-tests.

subarea of variable/region combinations which appear most important with respect to the goals of the study (adjusted alpha probability = 5%). The primary target variable in the present study in narcolepsy patients was P300 latency. The other ERP variables were included in the second, descriptive part. In this part, all p-values regarding intergroup differences are reported and displayed as graphs or images (statistical parametric maps, SPMs). For p-values below an (unadjusted) alpha probability of 5% it was determined if they appeared in a ‘‘near regular” pattern associated with numerically relevant effects. To evaluate differences between modafinil and placebo, t-tests were computed for log-transformed LORETA values. These voxel-by-voxel t-values were displayed as SPMs for p < 0.05. A single null hypothesis was tested for ‘‘omnibus” significance by means of a binomial test (p < 0.05) [36]. On the basis of the Structure-Probability Maps Atlas [37], the number of significant voxels in each lobe (frontal, parietal, occipital, temporal, limbic and sub-lobar) of the left and right hemisphere was computed. Data on sleep latencies in the MWT and ESS were analyzed by Wilcoxon’s signed rank tests as they were not normally distributed. Normal distribution was tested by means of the 1-sample Kolmogorov Smirnov Test. The null hypothesis stated: there are no differences between modafinil and placebo (error probability = 0.05).

2.7. Statistical analyses

Standard N1 and P2 as well as target N2 and P300 latencies and amplitudes at midline electrodes after modafinil as compared with placebo are summarized in Table 1. While in the N1 and P2 components a slight latency increase (3.1 ms; p < 0.05) and no significant change ( 1.3 ms) occurred under modafinil as compared with placebo; the latencies of the N2 and P300 cognitive components were significantly shortened by 13.3 ms (p < 0.05) and -32.8 ms, respectively (p < 0.01). In the narcolepsy-cataplexy subgroup (n = 10) the latencies of the N2

Statistical analysis was based on the concept of descriptive data analysis (DDA), as proposed by Abt [34]. DDA is particularly suitable for studies with many variables at many locations and many time points [35] and has been successfully applied in various controlled clinical and pharmaco-EEG studies. It consists of a confirmatory and a descriptive part. In the confirmatory part, techniques for coping with multiplicity are applied for a predefined

3. Results All 15 narcolepsy patients were included in the analysis. In the ERP, the mean percentage of artifact-free epochs was 87.3% for target and 87.2% for standard tones. Concerning the error rates (i.e., the absolute difference between the number of target tones reported by the patients and the number of target tones actually presented), there were no significant differences between the three recording sessions (baseline: 1.67, placebo: 1.33, modafinil: 1.73). 3.1. Differences between changes in ERP latencies induced by modafinil and placebo

Table 1 Effects of modafinil on ERP latencies and amplitudes in narcoleptics (n:15) Latency (ms)

Fz-amplitude (lV)

N1

Placebo Modafinil Change

84.9 (±6.5) 88.0 (±5.7) 3.1*

5.0 (±2.7) 5.0 (±2.5) 0.0

P2

Placebo Modafinil Change

186.2 (±31.7) 184.9 (±34.7) 1.3

3.4 (±1.6) 2.9 (±1.6) 0.4

N2

Placebo Modafinil Change

242.7 (±30.6) 229.4 (±25.8) 13.3*

P300

Placebo Modafinil Change

408.9 (±39.8) 376.0 (±53.4) 32.8 **

Paired-samples t-test. p < 0.05. ** p < 0.01). *

3.3 (±4.6) 3.4 (±4.8) 0.1 1.9 (±4.5) 2.3 (±2.9) 0.4

Cz-amplitude (lV) 5.2 (±2.9) 5.5 (±3.0) 0.3 3.9 (±1.8) 4.0 (±1.7) 0.0 2.8 (±5.0) 3.7 (±5.9) 0.9 5.7 (±4.0) 5.7 (±3.9) 0.0

Pz-amplitude (lV) 3.3 (±2.6) 3.8 (±3.0) 0.5 2.4 (±1.6) 2.7 (±1.6) 0.3 0.1 (±3.4) 1.8 (±4.5) 1.7 9.2 (±3.8) 8.2 (±3.9) 1.0

854

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

and P300 cognitive components were still significantly shortened by 16.8 ms (p < 0.05) and 28.5 ms, respectively (p < 0.05). 3.2. Differences between changes in ERP amplitudes induced by modafinil and placebo The N1 amplitude decreased significantly (p < 0.05) left temporally (T3), while the P2 amplitude increased significantly (p < 0.05) right occipitotemporally (T6) under modafinil as compared with placebo. The N2 amplitude tended to increase left occipitally (O1). The P3 amplitude decreased significantly left occipitotemporally at T6 (p < 0.01) and left occipitally at O1 (p < 0.05) and tended to decrease left temporally (T3). 3.3. Differences between changes in standard N1 and P2 as well as target N2 and P300 LORETA source strength induced by modafinil and placebo LORETA revealed increased source strength for N1 in the left auditory cortex (p < 0.05) under modafinil as compared with placebo (Fig. 2). Additionally, in unspecific regions an increase in the left posterior cingulate gyrus as well as a decrease in the paracentral and superior/middle frontal gyri were observed. In P2 no significant (p < 0.05) changes occurred in regions specific for P2 generation (superior temporal gyrus and precuneus), while an increase was seen in the left temporal and parahippocampal regions. In N2, source strength decreased in the left supramarginal and occipital region. Regarding P300, source strength decreased in the left superior temporal gyrus and increased in the medial frontal regions and

the right dorsolateral prefrontal cortex under modafinil as compared with placebo (Fig. 2). Details on the regional differences in P300 LORETA source strength between changes induced by modafinil and placebo are shown in 17 horizontal slices in Fig. 3. In the narcolepsy-cataplexy subgroup LORETA source strength increased in the medial frontal regions and in the right dorsolateral prefrontal cortex as was seen in the total group, but the findings did not reach the level of statistical significance. 3.4. Differences between modafinil and placebo in the MWT and the ESS The ESS score improved significantly from 15.4 (± 4.0) under placebo to 10.2 (±4.1) under 400 mg modafinil (p = 0.004). In the MWT, latency to sleep increased nonsignificantly from 11.9 ± 6.9 min under placebo to 13.3 ± 7.1 min under modafinil. In the narcolepsy-cataplexy subgroup the ESS score improved significantly like in the total group from 15.7 (± 3.7) under placebo to 11.1 (+ 3.9) under 400 mg modafinil (p = 0.017). In the MWT, latency to sleep increased nonsignificantly after modafinil treatment (12.3 + 7.0 as compared with 10.5 ± 6.7 min under placebo).

4. Discussion 4.1. Modafinil normalizes cognitive ERP latencies Our ERP study revealed significantly shortened N2 and P300 latencies under modafinil as compared with placebo, which reflects an improvement of information processing speed. In a recent study we found prolonged latencies and reduced amplitudes in N2 and

Fig. 2. Differences between modafinil- and placebo-induced changes in standard N1 and P2 as well as target N2 and P300 LORETA source strength in 15 narcoleptic patients. Statistical differences based on t-values are projected to the left and right lateral and medial cortical surfaces (t > 2.06: p<0.05). Red colors represent significant increases, blue colors significant decreases after modafinil as compared to placebo. The yellow numbers on the right side of figure correspond to physiological standard N1 and P2 as well as target N2 and P300 generators. LORETA revealed increased source strength for N1 in the left auditory cortex. P300 source strength decreased in the left superior temporal gyrus and increased in the medial frontal regions and in the right dorsolateral prefrontal cortex.

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

855

Fig. 3. Regional differences in P300 LORETA source strenght between modafinil-induced and placebo-induced changes in narcolepsy depicted in 17 horizontal slices including all 2,394 voxels (n=15). In the color key, t values are represented: blue colors indicate a significant decrease, red colors a significant increase. Modafinil increases source strength in the right dorsolateral prefrontal cortex for target P300 as compared with placebo (P <0.05, t >2.05).

P300 components in the same drug-free patients as compared with normal controls but no intergroup differences in N1 and P2 components [38]. This suggests impaired cognitive information processing in narcolepsy, while perception remains largely unaffected. In the present study, modafinil was shown to improve stimulus evaluation time, thereby tending to normalize cognitive information processing in narcoleptics, which confirms a ‘‘key-lock principle” in the diagnosis and therapy of impaired cognition in these patients. Visual P300 latency (VL) was used to predict treatment response to modafinil in narcolepsy: Non-responders (as measured by sleep latency in the MWT) had a longer VL than responders [39]. This was also shown in patients with multiple sclerosis related fatigue [40]. Patients with a shorter auditory P300 latency at baseline were more likely to benefit from modafinil treatment. The difference in P300 latencies and amplitudes recorded before and during treatment did not show any significant correlations with the change in the VAS score before and during treatment. P300 latency therefore not only seems to be a good treatment monitor as shown in our study, but can predict treatment response in various patient groups as well. 4.2. LORETA identifies the medial and right dorsolateral prefrontal cortex as a target region of modafinil in regard to cognitive information processing In order to identify cerebral target structures of modafinil that were involved in cognitive information processing we subjected

our data to further analyses by means of LORETA. LORETA revealed increased source strength for N1 in the left auditory cortex and for P300 in the medial and right dorsolateral prefrontal cortex. Our baseline studies in untreated narcoleptics had shown decreased energetic resources for cognitive processing, specifically in brain areas of the executive attention network [38]. As our study was planned before 2005, we did not primarily differentiate between narcolepsy with and without cataplexy. However, we carefully ruled out other reasons for excessive daytime sleepiness than narcolepsy. In a subgroup analysis there was only a trend for an increase in the above mentioned hot spot regions, which did not reach the level of statistical significance due to the small sample size (n = 10). Varied findings suggest that modafinil may potentiate both DA and NE neurotransmission [21]. The drug has been demonstrated to directly bind to the DAT and NET, which it inhibits at modest potency [21]. These effects are particularly prominent in the neocortex and generally less potent or minimal in various subcortical areas [21]. Parenteral administration of modafinil does lead to extracellular DA levels along with NE levels (measured by microdialysis) that are increased significantly in the prefrontal cortex of rats (PFC) [41]. In post-mortem human brains, DAT is found not only in the striatum, but also throughout the neocortex, including the PFC, albeit at relatively lower concentrations [42]. In healthy humans (with or without undergoing sleep deprivation), working memory, recognition memory, sustained attention, and other tasks dependent on cognitive control are enhanced with modafinil [21,43]. Among adult psychiatric patients, there is evi-

856

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

dence that modafinil improves several prefrontal-cortex-dependent cognitive functions in schizophrenia, major depression, adult ADHD and detoxified alcohol-dependent patients [21,44]. A recent VBM study in narcolepsy reported a significant gray matter loss in the right prefrontal and frontomesial cortex of patients, indicating a disease-related atrophy pattern, which may also be responsible for reported prefrontal dysfunctions [5]. Comparing these findings with our present study again reflects a ‘‘key-lock principle” in the diagnosis and therapy of prefrontal dysfunction. 4.3. Cognitive dysfunction in narcolepsy Concentration and learning difficulties were reported by a significant proportion of narcolepsy sufferers [45,46]. Rieger et al. also found deficits in divided and flexible attention in narcolepsy [47]. Nauman et al. described deficits in some attention and all executive function tests, whereas memory and routine alertness tasks were only mildly impaired in narcolepsy, indicating a reduced capacity to maintain a sufficient level of alertness across longer periods of time when concentrating on a demanding task [48]. In a multicenter randomized, double-blind, placebo-controlled 12-week study in narcolepsy armodafinil improved reaction time and episodic recall and recognition [24]. In our own recent study we determined the effects of modafinil on cognition and mood in narcoleptics and investigated whether neurophysiological vigilance changes correlated with cognitive and subjective vigilance alterations at the behavioral level [49]. Cognitive performance measured in an arithmetic task (Pauli Test) was significantly better after modafinil than after placebo. Correlation analyses revealed that a decrease in prefrontal delta, theta and alpha-1 power correlated with an improvement in cognitive performance. The general pattern of cognitive dysfunction in narcolepsy is consistent with a reduction or limitation of cognitive processing resources in prefrontal areas. The hypothesized mechanism may be related to changes in the hypocretin system [50,51]. A disruption of this system is associated with a deficient regulation of cortical activity and deficient vigilance [52,53]. 4.4. Functional neuroimaging in narcolepsy Functional neuroimaging studies of executive functioning have consistently demonstrated activation in an interconnected and distributed network of cortical areas. This network includes the dorsolateral prefrontal cortex, the cognitive division of the anterior cingulate, and the posterior parietal cortex [54]. There is evidence of vulnerability of prefrontal cortical mechanisms in states of inadequate or disrupted sleep [55,56]. In order to show pharmacological drug effects in narcoleptic subjects as compared with normal controls, functional magnetic resonance imaging (fMRI) has been used to detect regional brain responses to changes in sensory stimuli [57]. Mean cortical activation levels during the presentation of periodic visual and auditory stimulation showed no appreciable differences with either age or sex. Modafinil caused an increase in self-reported levels of alertness in 7 of 8 narcoleptic subjects, but there was no significant difference between mean pre- and post-treatment activation levels, as determined by fMRI for either normal or narcoleptic syndrome subjects receiving modafinil. The negative results might be due to the sample size. In another fMRI study with three drug-free narcoleptics, cerebral activation was mapped during the performance of a 2-back verbal working memory task [58]. In all subjects during the first scan a bilateral and widespread activation in known nodes of the executive network was seen, including the lateral prefrontal, posterior parietal and anterior cingulate cortex. There was a reduction in cerebral activation, especially but not exclusively in the prefrontal cortex, associated with a slowing of performance from the

first to the last tolerated scan. On stimulants, subjective alertness, activation and objective performance were readily maintained. In a recent SPECT study before and after modafinil or placebo in narcolepsy, chronic modafinil treatment for 4 weeks increased regional cerebral blood flow (rCBF) in the right dorsolateral and anterior cingulate cortices, which confirms our results of this hot spot region for attention [59]. The same group of scientists studied the effects of modafinil on rCBF in normal subjects and showed that a single dose of modafinil (400 mg at once) increased rCBF mainly in the thalamus and pons as well as prefrontal and cingulate cortices in healthy volunteers [60]. Functional imaging studies with sleep deprived normal subjects, as compared with narcolepsy patients, could clarify the question if the increase in the medial and right dorsolateral prefrontal cortex is specific for narcolepsy.

5. Conclusion 1. Our ERP studies demonstrated that modafinil shortened N2 and P300 latency, which reflects accelerated cognitive information processing. As untreated narcoleptics exhibited prolonged N2 and P300 latencies, these drug-induced changes confirm a ‘‘key-lock principle” in the diagnosis and treatment of narcolepsy with modafinil. 2. Our electrophysiological neuroimaging studies with LORETA identified the medial and right prefrontal cortex as a target region of modafinil in regard to cognitive information processing, as increased source density reflected increased energetic resources. Baseline studies in untreated narcoleptics by means of LORETA had shown decreased energetic resources for cognitive processing, specifically in brain areas of the executive attention network. Thus, the ‘‘key-lock principle” also applies to the source density data. 3. The prefrontal dysfunction, in which modafinil has been shown to be of therapeutic benefit, has been found to be crucial in a number of neuropsychiatric disorders. Our study showed that modafinil works as a cognitive enhancer in narcolepsy, too. 4. Finally, our study suggests that the same brain regions that have already been described as pathogenetically of interest in narcolepsy by morphological and functional MRI neuroimaging techniques may also be identified by electrophysiological neuroimaging. Acknowledgment This study was supported by an independent research Grant by Cephalon GmbH, Germany to Bernd Saletu, MD, professor of psychiatry, Section of Sleep Research and Pharmacopsychiatry, Department of Psychiatry and Psychotherapy, Medical University of Vienna. Cephalon was not involved in the study design, the collection, analysis and interpretation of data, writing of the report, and the decision to submit the paper for publication. A portion of the data contained in this manuscript was presented at the APSS meeting, June 2007 in Minneapolis, USA. Authors’ competing financial interests. P.A, M.M, J.Z. have no competing financial interests to declare. B.S. has received research grants, honoraria for consulting and lectures from most of the pharmaceutical companies. G.M.S has received research grants from GlaxoSmithKline, Pfizer and Lundbeck. M.S has received travel grants from Cephalon, Pfizer, Roche Austria, GlaxoSmithKline and Boehringer Ingelheim. The authors would further like to express their thanks to Mag Elisabeth Grätzhofer for editorial assistance, and finally to the entire staff of the Section of Sleep Research, Department of Psychiatry and Neurology, Medical University of Vienna for their cooperative assistance in this project.

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

References [1] Dang-Vu TT, Desseilles M, Petit D, Mazza S, Montplaisir J, Maquet P. Neuroimaging in sleep medicine. Sleep med 2007;8(4):349–72. [2] Draganski B, Geisler P, Hajak G, Schuierer G, Bogdahn U, Winkler J, et al. Hypothalamic gray matter changes in narcoleptic patients. Nature medicine 2002;8(11):1186–8. [3] Kaufmann C, Schuld A, Pollmacher T, Auer DP. Reduced cortical gray matter in narcolepsy: preliminary findings with voxel-based morphometry. Neurology 2002;58(12):1852–5. [4] Overeem S, Steens SC, Good CD, Ferrari MD, Mignot E, Frackowiak RS, et al. Voxel-based morphometry in hypocretin-deficient narcolepsy. Sleep 2003;26(1):44–6. [5] Brenneis C, Brandauer E, Frauscher B, Schocke M, Trieb T, Poewe W, et al. Voxel-based morphometry in narcolepsy. Sleep Medicine 2005;6(6):531–6. [6] Nofzinger EA. Neuroimaging the sleep disordered patient: a commentary. Sleep Med 2005;6(5):383–4. [7] Pascual-Marqui RD, Lehmann D, Koenig T, Kochi K, Merlo MC, Hell D, et al. Low resolution brain electromagnetic tomography (LORETA) functional imaging in acute, neuroleptic-naive, first-episode, productive schizophrenia. Psychiatry Res 1999;90(3):169–79. [8] Saletu B, Anderer P, Saletu-Zyhlarz GM, Arnold O, Pascual-Marqui RD. Classification and evaluation of the pharmacodynamics of psychotropic drugs by single-lead pharmaco-EEG, EEG mapping and tomography (LORETA). Methods Find Exp Clin Pharmacol 2002;24(Suppl. C):97–120. [9] Saletu M, Anderer P, Saletu-Zyhlarz GM, Mandl M, Arnold O, Zeitlhofer J, et al. EEG-tomographic studies with LORETA on vigilance differences between narcolepsy patients and controls and subsequent double-blind, placebocontrolled studies with modafinil. J Neurol 2004;251(11):1354–63. [10] Pascual-Marqui RD, Michel CM, Lehmann D. Low resolution electromagnetic tomography: a new method for localizing electrical activity in the brain. Int J Psychophysiol 1994;18(1):49–65. [11] Talairach J, Tournoux P. Co-Planar Stereotaxic Atlas of the Human Brain. Stuttgart: Thieme 1988. [12] Arruda JE, Walker KA, Weiler MD, Valentino DA. Validation of a right hemisphere vigilance system as measured by principal component and factor analyzed quantitative electroencephalogram. Int J Psychophysiol 1999;32(2):119–28. [13] Näätänen R. Implications of event-related potentials data for psychological theories of attention. In: Renault MKM, Coles MGH, Gailland AGM editors. Event-related potentials investigations of cognition. Amsterdam: North Holland; 1989, p. 117–64. [14] Picton T, Hillyard S. Endogenous event-related potentials. In: Picton T, editor. Human event-related potentials Handbook of electroencephalography and clinical neurophysiology. Amsterdam: Elsevier; 1988. p. 361–426. [15] Polich J. P300 clinical utility and control of variability. J Clin Neurophysiol 1998;15(1):14–33. [16] Halgren E, Marinkovic K, Chauvel P. Generators of the late cognitive potentials in auditory and visual oddball tasks. Electroencephalogr Clin Neurophysiol 1998;106(2):156–64. [17] Sangal RB, Sangal JM. P300 latency: abnormal in sleep apnea with somnolence and idiopathic hypersomnia, but normal in narcolepsy. Clin Electroencephalogr 1995;26(3):146–53. [18] Sangal RB, Sangal JM, Belisle C. Longer auditory and visual P300 latencies in patients with narcolepsy. Clin Electroencephalogr 1999;30(1):28–32. [19] Naumann A, Bierbrauer J, Przuntek H, Daum I. Attentive and preattentive processing in narcolepsy as revealed by event-related potentials (ERPs). Neuroreport 2001;12(13):2807–11. [20] Mitler MM, Harsh J, Hirshkowitz M, Guilleminault C. Long-term efficacy and safety of modafinil (PROVIGIL((R))) for the treatment of excessive daytime sleepiness associated with narcolepsy. Sleep Med 2000;1(3): 231–43. [21] Minzenberg MJ, Carter CS. Modafinil: A Review of Neurochemical Actions and Effects on Cognition. Neuropsychopharmacology 2008;33(7):1477–502. [22] US Modafinil in Narcolepsy Multicenter Study Group. Randomized trial of modafinil for the treatment of pathological somnolence in narcolepsy. Ann Neurol 1998;43(1):88–97. [23] US Modafinil in Narcolepsy Multicenter Study Group. Randomized trial of modafinil as a treatment for the excessive daytime somnolence of narcolepsy. Neurology 2000;54(5):1166–75. [24] Harsh JR, Hayduk R, Rosenberg R, Wesnes KA, Walsh JK, Arora S, et al. The efficacy and safety of armodafinil as treatment for adults with excessive sleepiness associated with narcolepsy. Current medical research and opinion 2006;22(4):761–74. [25] American Sleep Disorders Association. The International Classification of Sleep. Disorders, revised Diagnostic and Coding Manual. ed. Rochester; 1997 [26] Johns M. A new method for measuring daytime sleepiness: The Epworth Sleepiness Scale. Sleep 1991;14(6):540–5. [27] Mitler MM, Gujavarty KS, Browman CP. Maintenance of Wakefulness Test: a polysomnographic technique for evaluation treatment efficacy in patients with excessive somnolence. Electroencephalogr Clin Neurophysiol 1982;53(6):658–61. [28] Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Brain Information Service, Los Angeles, University of California, 1968.

857

[29] Squires NK, Squires KC, Hillyard SA. Two varieties of long-latency positive waves evoked by unpredictable auditory stimuli in man. Electroencephalogr Clin Neurophysiol 1975;38(4):387–401. [30] Semlitsch HV, Anderer P, Saletu B, Resch F, Presslich O, Schuster P. Psychophysiological research in psychiatry and neuropsychopharmacology. I. Methodological aspects of the Viennese Psychophysiological Test-System (VPTS). Methods Find Exp Clin Pharmacol 1989;11(1):25–41. [31] Anderer P, Semlitsch HV, Saletu B, Barbanoj MJ. Artifact processing in topographic mapping of electroencephalographic activity in neuropsychopharmacology. Psychiatry Res 1992;45(2):79–93. [32] Semlitsch HV, Anderer P, Schuster P, Presslich O. A solution for reliable and valid reduction of ocular artifacts, applied to the P300 ERP. Psychophysiology 1986;23(6):695–703. [33] Anderer P, Saletu B, Semlitsch HV, Pascual-Marqui RD. Perceptual and cognitive event-related potentials in neuropsychopharmacology: methodological aspects and clinical applications (pharmaco-ERP topography and tomography). Methods Find Exp Clin Pharmacol. 2002;24(Suppl. C):121–37. [34] Abt K. Planning controlled clinical trials on the basis of descriptive data analysis. Statistics in medicine 1991;10(5):777–95. [35] Ferber G, Abt K, Fichte K, Luthringer R. IPEG guideline on statistical design and analysis for pharmacodynamic trials. International Pharmaco-EEG Group. Neuropsychobiology 1999;39(2):92–100. [36] Cross EM, Chaffin WW. Use of the binomial theorem in interpreting results of multiple tests of significance. Educational and Psychological Measurement. 1982;42:25–34. [37] Lancester JL, Rainey LH, Summerlin JL, Freitas CS, Fox PT, Evans AC, et al. Automated labeling of the human brain a preliminary report on the development and evaluation of a forward transform method. Human Brain Mapping 1997;5:238–42. [38] Saletu M, Anderer P, Saletu-Zyhlarz GM, Mandl M, Zeitlhofer J, Saletu B. Eventrelated potentials low-resolution brain electromagnetic tomography (ERPLORETA) suggests decreased energetic resources for cognitive processing in narcolepsy. Clin Neurophysiol 2008;119(8):1782–94. [39] Sangal RB, Sangal JM, Belisle C. Visual P300 latency predicts treatment response to modafinil in patients with narcolepsy. Clin Neurophysiol 1999;110(6):1041–7. [40] Nagels G, D’hooghe MB, Vleugels L, Kos D, Despontin M, De Deyn PP. P300 and treatment effect of modafinil on fatigue in multiple sclerosis. J Clin Neuroscience 2007;14:33–40. [41] De Saint-Hilaire Z, Orosco M, Rouch C, Blanc G, Nicolaidis S. Variations in extracellular monoamines in the prefrontal cortex and medial hypothalamus after modafinil administration: a microdialysis study in rats. Neuroreport 2001;12(16):3533–7. [42] Ciliax BJ, Drash GW, Staley JK, Haber S, Mobley CJ, Miller GW, et al. Immunocytochemical localization of the dopamine transporter in human brain. The Journal of comparative neurology 1999;409(1):38–56. [43] Saletu B, Grünberger J, Linzmayer L, Stöhr H. Pharmaco-EEG, psychometric and plasma level studies with two novel alpha-adrenergic stimulants CRL 40476 and 40028 (Adrafinil) in elderlies. New Trends in Experimental Clinical Psychiatry 1986(2):5–31. [44] Saletu B, Saletu M, Grünberger J, Frey R, Anderer P, Mader R. Treatment of the alcoholic organic brain syndrome: double-blind, placebo-controlled clinical, psychometric and electroencephalographic mapping studies with modafinil. Neuropsychobiology 1993;27(1):26–39. [45] Schulz H, Wilde-Frenz J, Grabietz-Kurfurst U. Cognitive deficits in patients with daytime sleepiness. Acta neurologica Belgica 1997;97(2):108–12. [46] Smith KM, Merrit SL, Cohen FL. Can we predict cognitive impairment in persons with narcolepsy. Loss Grief Care 1992(5):103–13. [47] Rieger M, Mayer G, Gauggel S. Attention deficits in patients with narcolepsy. Sleep 2003;26(1):36–43. [48] Naumann A, Bellebaum C, Daum I. Cognitive deficits in narcolepsy. J Sleep Res 2006;15(3):329–38. [49] Saletu M, Anderer P, Semlitsch HV, Saletu-Zyhlarz GM, Mandl M, Zeitlhofer J, et al. Low-resolution brain electromagnetic tomography (LORETA) identifies brain regions linked to psychometric performance under modafinil in narcolepsy. Psychiatry Res 2007;154(1):69–84. [50] Peyron C, Faraco J, Rogers W, Ripley B, Overeem S, Charnay Y, et al. A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nature medicine 2000;6(9):991–7. [51] Thannickal TC, Moore RY, Nienhuis R, Ramanathan L, Gulyani S, Aldrich M, et al. Reduced number of hypocretin neurons in human narcolepsy. Neuron 2000;27(3):469–74. [52] Rogers AE, Rosenberg RS. Tests of memory in narcoleptics. Sleep 1990;13(1):42–52. [53] Valley V, Broughton R. Daytime performance deficits and physiological vigilance in untreated patients with narcolepsy-cataplexy compared to controls. Revue d’electroencephalographie et de neurophysiologie clinique 1981;11(1):133–9. [54] Duncan J, Owen AM. Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends in neurosciences 2000;23(10):475–83. [55] Muzur A, Pace-Schott EF, Hobson JA. The prefrontal cortex in sleep. Trends Cogn Sci 2002;6(11):475–81. [56] Harrison Y, Horne JA, Rothwell A. Prefrontal neuropsychological effects of sleep deprivation in young adults–a model for healthy aging? Sleep 2000;23(8):1067–73.

858

M. Saletu et al. / Sleep Medicine 10 (2009) 850–858

[57] Ellis CM, Monk C, Simmons A, Lemmens G, Williams SC, Brammer M, et al. Functional magnetic resonance imaging neuroactivation studies in normal subjects and subjects with the narcoleptic syndrome. Actions of modafinil. J Sleep Res 1999;8(2):85–93. [58] Thomas RJ. Fatigue in the executive cortical network demonstrated in narcoleptics using functional magnetic resonance imaging–a preliminary study. Sleep Med 2005;6(5):399–406.

[59] Joo EY; Seo DW; Tae WS; Hong SB. Effect of modafinil on cerebral blood flow in narcolepsy patients. Sleep 2008;1,31(8):868–73. [60] Joo EY, Tae WS, Jung KY, Hong SB. Cerebral blood flow changes in man by wake-promoting drug, modafinil: a randomized double blind study. J Sleep Res 2008;17(1):82–8.