Correspondence : Yang-Whan Jeon, MD, Department of Psychiatry, Our Lady of Mercy Hospital, College of Medicine, The Catholic University of Korea, 665 Bupyeong 6-dong, Bupyeong-gu, Incheon 403-720, Korea
Tel : +82-32-510-5800, Fax : +82-32-510-5678, E-mail : jeonleo@olmh.cuk.ac.kr

Introduction

Benzodiazepines have historically been the mainstay of treatment for sleep disturbances, yet they have many shortcomings such as causing residual daytime sedation, accumulation during multiple dosage, cognitive impairment and injuries.¹ The newer non-benzodiazepine agents such as zolpidem, zopiclone, and zaleplon are considered safer, probably because of their pharmacokinetic properties as well as their selective pharmacological activities at the omega-1 receptors. These properties may explain the limited negative influences of these agents on daytime performance. Among them, zolpidem has become one of the most commonly prescribed hypnotics that treat initial and middle insomnia. Zolpidem is rapidly absorbed (20-40 minutes), reaches a peak plasma concentration after about 1.5 hours and has a short half-life (1-2.4 hours).^2,3 Due to its short elimination half-life, zolpidem appears to have a low likelihood of producing residual sedative effects and, therefore, to be associated with a reduced risk of accidents and injuries.
A negative effect on cognitive and psychomotor function near the time of plasma peak concentration has been reported in many studies.^4,5 Greenblatt and colleagues reported that zolpidem produced benzodiazepine agonist-like pharmacokinetic effects of relatively short duration on electroencephalography (EEG).² Berlin et al. found that the effects had disappeared at 4 hours,⁵ and Rush et al. reported similar results.⁶ The time course of EEG activity indicated the duration of zolpidem's effect to be about 4 hours.⁷ However, some studies have suggested that the residual effects of zolpidem last for a longer time.^8,9 One of the most effective ways of objectively evaluating sedative effects on cognition is to measure the electrophyiologic activity through the P300 ERP. The P300 ERP is often obtained with a simple discrimination task in which two different stimuli are randomly presented with one occurring less frequently (target stimulus) than the other (nontarget stimulus).¹⁰ Changes in the amplitude and latency of P300 have been associated with cognitive processing. Since the P300 ERP is affected by the physical state of its underlying physiology such as the arousal level, it may be a useful method of detecting hypnotic effects on cognition in sleep studies.¹¹
P300 shows a decrease in amplitude and a delay in latency not only in response to benzodiazepine hypnotics, such as triazolam,¹² midazolam,¹³ and flunitrazepam,14 but also to the non-benzodiapine hypnotic, zolpidem,¹⁵ even though zolpidem is known to be safer and more tolerable. Lucchesi and colleagues investigated the acute neurophysiological effects of zolpidem at the time of its peak-plasma concentration and showed that it decreased the P300 amplitude and delayed the P300 latency.¹⁵
To the best of our knowledge, there have been very few previous studies which investigated the residual effects of zolpidem on cognitive function using ERPs.^16,17 Moreover, these previous studies measured the ERPs long after the ingestion of zolpidem, at a time which exceeded its efficacy lifetime, and thus found no residual effects. Thus, this study was designed to examine the residual cognitive effects of zolpidem 5 hours after its administration in young, healthy subjects by measuring the auditory P300.

Methods

Participants
Twelve young university students were recruited for this study (all males, mean age=23.6 years, SD=3.2) and received pecuniary remuneration for their participation. All of the subjects had normal hearing acuity and had no history of psychiatric/neurological disorders. All of the subjects were non-smokers, in order to avoid smoking effects on P300.^18,19 This protocol was approved by the Institutional Review Board in Our Lady of Mercy Hospital. Written informed consent was obtained from all subjects prior to the study.

Procedure
The baseline pre-drug auditory P300 was recorded at 6 p.m. Two weeks later, the participants were administered 10 mg of zolpidem orally at 1 p.m. and were instructed to take bed rest for 5 hours. The post-drug audi tory P300 was recorded in each subject at 6 p.m.

ERP recording
The stimuli were 200, 75 dB tones, with a 2 s interstimulus interval. Eighty percent of the tones were 'non-targets' with a frequency of 1,000 Hz and the remaining 20% were 'targets' with a frequency of 1,000 Hz and these tones were presented in a quasi-random order. The subjects were instructed to press a button in response to the targets only. The reaction times (ms) to press the button in response to the ERP target tones were measured.
The EEG activity was recorded from 25 electrodes that included Fz, F3/4, F7/8 (frontal row), FCz, FC3/4, FC7/8 (fronto-central row), Cz, C3/4, T7/8 (central row), CPz, CP3/4 (centro-parietal row), Pz, P3/4, P7/8 (parietal row), M1/2 (mastoid), referred to nose, with forehead ground and impedance set at 5 kΩ or less. Four additional electrodes were placed at both lateral canthi, below and above the left eye, to measure the electro-ocular (EOG) activity with bipolar recordings. The bandpass was 0.05-30 Hz, and the EEG was digitized at 500 Hz for 1,000 ms, with a 100 ms prestimulus baseline. Waveforms were averaged off-line, such that those trials on which the EEG or EOG exceeded ±100μV were rejected. Single trial data were also subjected to an EOG correction procedure to remove any remaining artifacts.²⁰ The P300 component was defined as the largest positive-going peak occurring within 300-800 ms at each electrode. 

Data analyses
Drug effects were assessed with a 3-factor (2 Drug conditions (pre- vs. post-drug)×3 anterior-to-posterior electrode locations ×5 coronal electrode locations) analysis of variance (ANOVA) applied to the amplitude and latency data of the target stimulus P300. Greenhouse-Geisser corrections to the degree of freedom were used to adjust for violations of the sphericity assumption for repeated measures factors containing more than two levels.

Results

Task performance
The response accuracy was not different between the pre- and post-drug conditions (100% vs. 98%). The reaction time of the post-drug condition (712±114 ms) was significantly delayed compared to that of the predrug condition (375±125 ms)(t=6.9, df=22, p<0.001).

P300 amplitude and latency
Figure 1 illustrates the grand averaged ERPs from the target stimuli for the pre- and post-drug conditions and different electrodes. Figure 2 illustrates the mean P300 amplitude and latency data from the target stimuli as a function of the coronal electrode site for the frontal, central, and parietal arrays.
Table 1 summarizes the results of the statistical analyses.
There was a significant main effect of drug condition (pre- and post-drug) on the P300 amplitude [F(1,11) =5.6, p<0.05]. The P300 amplitude of the post-drug condition was decreased compared to that of the pre-drug condition across all electrodes, but the P300 latency was not delayed. There was a significant main effect of the 3 anterior to posterior electrode locations [F(2,22)=38.1, p<0.001] and 5 coronal electrode locations [F(4,44)=12.9, p<0.001] on the P300 amplitude. There was also a significant interaction between the 3 anterior to posterior electrode locations and the 5 coronal electrode locations [F(8,88)=10.6, p<0.001].

Discussion

A hypnotic that is used in the management of insomnia should rapidly induce sleep and be free of adverse residual effects on cognitive and psychomotor function. Zolpidem has a short half-life and thus has a short duration of action. Though several studies have suggested that zolpidem does not induce any residual next-morning negative effect on cognitive function, their results are inconsistent because of the diverse experimental methods that were employed.
The residual sedating effect of zolpidem on cognitive function can be evaluated by means of behavioral measures, EEG and ERPs. The P300 is elicited by using the "oddball" paradigm and is considered to reflect cognitive processing and attention.²⁰ The P300 amplitude is thought to index brain activity that is proportional to the level of attention and memory.²¹ The P300 latency is considered to be a measure of the stimulus classification speed during memory-updating.²²
The present study showed that zolpidem led to changes in the behavioral measures and ERPs when assessed 5 hours after intake. In terms of the behavioral measures, the reaction time on the task performance was much delayed. Some studies showed negative effects on cognitive and psychomotor function when measured up to 5 hours after the administration of zolpidem.^8,9 Other studies reported a lack of any negative residual effects on cognitive and psychomotor function from 6 hours after zolpidem administration.^5,23 Further studies using behavioral measures are needed to confirm the residual effects around these time periods. 
In this study, the P300 amplitude after zolpidem intake was decreased compared to that of the pre-drug condition across all electrodes. This finding suggests that zolpidem may have residual effects on cognitive function measured by ERPs. Although few studies have investigated the residual cognitive effects of zolpidem, it is well known that benzodiazepines produce residual effects on cognition. Triazolam, a benzodiazepine derivative having a similar half-life to that of zolpidem, caused cognitive dysfunction that was most prominent at 6 hours after administration, although its elimination half-life is very short.²⁴ This finding supports those of the present study that hypnotics may cause residual effects on cognition after the elimination half-life. 
In this study, the P300 latency was not delayed, although the reaction time on the task performance was delayed. This finding suggests that zolpidem does not affect the stimulus discrimination speed, but does affect the motor response.
Another interesting finding is the smaller P300 amplitude in the right frontal area in the post-drug condition, although the interaction between the drug conditions (pre- and post-drug) and anterior-to-posterior electrodes and coronal electrodes did not reach statistical significance (D×C×A/P, F=1.7, p=0.1; Figure 2). Discrimination between target and standard stimuli in an odd-ball paradigm is hypothesized to initiate frontal lobe activation that reflects the attentional focus required by task performance.²⁵ Actually, the initial neural activation during the oddball task appears to originate from the right frontal cortex,²⁶ and the P300 amplitude is reported to be larger over the right frontal/central areas as compared to that over the left frontal/central areas, as shown for the pre-drug condition in this study (Figure 2).²⁷ Previous studies reported a reduction of the P300 amplitude in the frontal area after acute zolpidem treatment¹⁵.htm','','width=600,height=100')">15 and during sleep onset,²⁸ which is consistent with the results in this study. Thus, the findings of this study suggest that zolpidem may lead to the impairment of the discrimination between relevant and irrelevant information or more generally to an attentional deficit.²⁰
The lack of a placebo-controlled group and the small sample size limits the interpretation of the findings of this study. In some studies, the experiment was conducted during the nighttime to create a similar situation to that experienced in everyday life. However, the experimental procedure used in this study was performed during daytime, which limits the generalization of the findings to everyday life. Further studies are needed to confirm the results of this study with more samples and various time periods.
In conclusion, this study suggests that the residual effects of zolpidem on cognitive function, such as task performance and attention, may persist after its elimination half-life, although zolpidem is known to be essentially free of residual cognitive impairment when taken at night.

1. Ray WA, Griffin MR, Downey W. Benzodiazepines of long and short elimination half-life and the risk of hip fracture. JAMA 1989;262:3303-3307.

2. Greenblatt DJ, Harmatz JS, von Moltke LL, Wright CE, Durol AL, Harrel-Joseph LM, et al. Comparative kinetics and response to the benzodiazepine agonists triazolam and zolpidem: evaluation of sex-dependent differences. J Pharmacol Exp Ther 2000;293:435-443.

3. Salva P, Costa J. Clinical pharmacokinetics and pharmacodynamics of zolpidem. Therapeutic implications. Clin Pharmacokinet 1995;29:142-153.

4. Wesensten NJ, Balkin TJ, Davis HQ, Belenky GL. Reversal of triazolam- and zolpidem-induced memory impairment by flumazenil. Psychopharmacology (Berl) 1995;121:242-249.

5. Berlin I, Warot D, Hergueta T, Molinier P, Bagot C, Puech AJ. Comparison of the effects of zolpidem and triazolam on memory functions, psychomotor performances, and postural sway in healthy subjects. J Clin Psychopharmacol 1993;13:100-106.

6. Rush CR, Griffiths RR. Zolpidem, triazolam, and temazepam: behavioral and subject-rated effects in normal volunteers. J Clin Psychopharmacol 1996;16:146-157.

7. Patat A, Trocherie S, Thebault JJ, Rosenzweig P, Dubruc C, Bianchetti G, et al. EEG profile of intravenous zolpidem in healthy volunteers. Psychopharmacology (Berl) 1994;114:138-146.

8. Danjou P, Paty I, Fruncillo R, Worthington P, Unruh M, Cevallos W, et al. A comparison of the residual effects of zaleplon and zolpidem following administration 5 to 2 h before awakening. Br J Clin Pharmacol 1999;48:367-374.

9. Hindmarch I, Patat A, Stanley N, Paty I, Rigney U. Residual effects of zaleplon and zolpidem following middle of the night administration five hours to one hour before awakening. Hum Psychopharmacol 2001;16:159-167.

10. Picton TW. The P300 wave of the human event-related potential. J Clin Neurophysiol 1992;9:456-479.

11. Engelhardt W, Friess K, Hartung E, Sold M, Dierks T. EEG and auditory evoked potential P300 compared with psychometric tests in assessing vigilance after benzodiazepine sedation and antagonism. Br J Anaesth 1992;69:75-80.

12. Urata J, Uchiyama M, Iyo M, Enomoto T, Hayakawa T, Tomiyama M, et al. Effects of a small dose of triazolam on P300 and resting EEG. Psychopharmacology (Berl) 1996;125:179-184.

13. Engelhardt W, Friess K, Hartung E, Sold M, Dierks T. EEG and auditory evoked potential P300 compared with psychometric tests in assessing vigilance after benzodiazepine sedation and antagonism. Br J Anaesth 1992;69:75-80.

14. Lucchesi LM, Pompeia S, Manzano GM, Kohn AF, Galduroz JC, Bueno OF, et al. Flunitrazepam-induced changes in neurophysiological, behavioural, and subjective measures used to assess sedation. Prog Neuropsychopharmacol Biol Psychiatry 2003;27:525-533.

15. Lucchesi LM, Braga NI, Manzano GM, Pompeia S, Tufik S. Acute neurophysiological effects of the hypnotic zolpidem in healthy volunteers. Prog Neuropsychopharmacol Biol Psychiatry 2005;29:557-564.

16. Nakagome K, Ichikawa I, Kanno O, Iyo R, Suzuki M, Takazawa S, et al. Effects of zolpidem and triazolam on sleep and daytime activities in normal young volunteers: II. Eventrelated potentials (ERP) study. Jpn J Neuropsychopharmacol 1993; 15:603-615 (abstract).

17. Nakajima T, Takazawa S, Hayashida S, Nakagome K, Sasaki T, Kanno O. Effects of zolpidem and zopiclone on cognitive and attentional function in young healthy volunteers: an event-related potential study. Psychiatry Clin Neurosci 2000;54:37-40.

18. Polich J, Ochoa CJ. Alcoholism risk, tobacco smoking, and P300 event-related potential. Clin Neurophysiol 2004;115:1374-1383.

19. 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;23:695-703.

20. Coull JT. Neural correlates of attention and arousal: insights from electrophysiology, functional neuroimaging and psychopharmacology. Prog Neurobiol 1998;55:343-361.

21. Siddle DA. Orienting, habituation, and resource allocation: an associative analysis. Psychophysiology 1991;28:245-259.

22. Verleger R. On the utility of P3 latency as an index of mental chronometry. Psychophysiology 1997;34:131-156.

23. Wesensten NJ, Balkin TJ, Belenky GL. Effects of daytime administration of zolpidem versus triazolam on memory. Eur J Clin Pharmacol 1995;48:115-122.

24. Hayakawa T, Uchiyama M, Urata J, Enomoto T, Okubo J, Okawa M. Effects of a small dose of triazolam on P300. Psychiatry Clin Neurosci 1999;53:185-187.

25. Pardo JV, Fox PT, Raichle ME. Localization of a human system for sustained attention by positron emission tomography. Nature 1991;349:61-64.

26. Polich J, Alexander JE, Bauer LO, Kuperman S, Morzorati S, O'Connor SJ, et al. P300 topography of amplitude/latency correlations. Brain Topogr 1997;9:275-282.

27. Alexander JE, Porjesz B, Bauer LO, Kuperman S, Morzorati S, O'Connor SJ, et al. P300 hemispheric amplitude asymmetries from a visual oddball task. Psychophysiology 1995;32:467-475.

28. Cote KA, De Lugt DR, Campbell KB. Changes in the scalp topography of event-related potentials and behavioral responses during the sleep onset period. Psychophysiology 2002;39:29-37.