The P300 component of the auditory evoked potential (ERP) can be derived from the EEG in oddball type experiments when rare task-relevant stimuli are interspersed into frequently presented non-task related stimuli [1]. The P300 component of the auditory evoked potential emerges between 260 ms to 500 ms following the onset of the target stimulus. The P300 is one of the most robust components within ERP research and the amplitude of the P300 component rises with lower objective probability and lower subjective certainty about the occurrence of the target stimulus [2]. The P300 reflects attention dependent target processing. A recent model on the neuropsychological basis of the P300 by Polich [3] suggests that a target stimulus initiates frontal lobe activity in the allocation of attention resources needed to perform the task associated with the target stimulus and the exchange of working memory content. Further processes involve the temporal lobe and the parietal cortex with stimulus representation maintenance mechanisms and subsequent memory storage and updating processes. Therefore, a simple task to press a button upon the rare occurrence of a deviant auditory stimulus evokes the initiation of a set of cognitive functions and their related areas in the brain [4].

Methadone substitution is a treatment for subjects addicted to opiates, mainly to heroin. Methadone is a synthetic opiate agonist at cerebral μ- and κ-opiate receptors. If given in a sufficient dose, methadone prevents opiate addicts from withdrawal symptoms after cessation of heroin use [5]. Methadone is taken orally and has an half-life of about 24 hours, so that usually one single dose per day is sufficient. When given in therapeutic doses to people adapted to opiates, methadone exhibits no noticeable psychotropic effects like sedation or euphoria. Nevertheless, the question emerged whether methadone has deteriorating effects on cognitive functioning. Previous studies found impairments in methadone substituted patients relative to healthy controls with respect to attention, psychomotor speed, working memory and information processing [6–9]. It is difficult, though, to differentiate the effects of methadone from other factors, like long term effects of opiate abuse, effects of former or concomitant use of other drugs like alcohol, cannabis or cocaine, or impairments following breath suppression after heroin overdoses [6, 10]. Current studies on cognitive abilities indicate that methadone-treated opiate addicts perform better than untreated opiate users, but worse than abstinent former opiate addicts [8, 11, 12].

The auditory P300 has been assessed in a sample of detoxified opiate addicts with and without methadone substitution [13]. Compared to controls, both patient groups showed decreased amplitudes and increased latencies of the P300 component. While in this study the P300 component in methadone substituted the group was less impaired, this result points to the hypothesis of methadone substitution acts as a stabilizing factor in attention related target processing. In a study by Kouri et al. [14] opiate addicts were detoxified and then treated with and without buprenorphine, a partial κ-antagonist and μ-agonist. While the P300 component decreased after detoxification in placebo treated subjects, buprenorphine reversed this decrement to the level of non-dependent control subjects. While both of these studies used small sample sizes of seven and ten subjects per group, they indicate that opiate addicts may show reduced P300 amplitudes with or without substitution [13], and that substitution treatment may improve auditory target processing in detoxified opiate addicts [14]. The clinical relevance of these neurophysiological studies is to gain insight into whether opiate addicted patients under substitution treatment show deficits which may impact their ability to return to or to attain more stable socio-cultural and occupational settings.

While nicotine use is highly prevalent in methadone substituted patients, with rates of comorbidity reaching 90%and more [15, 16], the effects of nicotine on the P300 component may be confounded with those of opiate addiction. Chronic nicotine consumption has been found to be associated with P300 amplitude reduction using a visual oddball paradigm in a large sample of current smokers when compared to subjects who never smoked or those who had terminated smoking [17]. More recently, Neuhaus et al. (2006), using an auditory oddball paradigm, reported persistent P300 amplitude reduction in smokers and furthermore, even in former smokers [18]. Results from these studies indicate that chronic nicotine consumption is associated with P300 amplitude reduction, which may confound results when comparing P300 results of nicotine dependent opiate addicts to those of non-smoking control subjects.

Here we aimed to assess whether methadone substituted opiate addicts differ from control subjects with regard to the auditory P300 component in a two-tone oddball design. Previous studies on the P300 in opiate addicts used small sample sizes and it remains unclear whether the P300 is unaffected in a larger group of methadone substituted opiate addicts when compared to controls. A second aim was to control for the potentially confounding effect of nicotine consumption in control subjects. Therefore, we compared the auditory P300 component in a group of methadone substituted opiate addicts to control subjects with and without current nicotine consumption.

We assessed 47 opiate addicted outpatients under current methadone substitution and 65 control subjects matched for age and gender. The study was conducted in compliance with the ethical principles for medical research involving human subjects according to the Declaration of Helsinki and was approved by the local ethics committee of the medical department of the University of Duisburg-Essen. Patients and control subjects gave written informed consent before study inclusion.

The comparison of methadone substituted opiate addicts with a group of matched control subjects revealed no significant group differences with regard to P300 amplitudes and latencies indicating no overall differences between patients and controls. In the analysis of subgroups, nicotine use in control subjects was associated with reduced P300 amplitudes and patients P300 amplitudes were in between those of smoking and non-smoking controls. Additional non-opioid drug use in patients had no significant impact upon P300 amplitudes or latencies. In the comparisons of controls with and without nicotine with patients with or without additional non-opioid drug use, only the comparison of patients with non-opioid drug use with smoking controls revealed significant differences with regard to P300 amplitudes. Here, patients with additional non-opioid drug use had significantly higher P300 amplitudes when compared to smoking controls. Latencies did not differ between groups and subgroups. With regard to the first aim of our study we found no attenuation of P300 amplitudes or latencies between patients under methadone substitution treatment and control subjects. With regard to the second aim of our study we found a significant effect of nicotine in control subjects.

Detoxification in opiate addicts has been associated with lowered P300 amplitudes when compared to controls and substitution treatment has been shown to increase P300 amplitudes [13, 14]. Regarding cognitive performance, patients under methadone treatment tend to show lower performance in measures of attention, memory and executive function but not in simple reaction time tasks when compared to controls [8, 23]. However, methadone treatment has been shown to improve memory and psychomotor speed [11]. In our study, patient's P300 amplitudes were in between those of the two control groups. While all patients in our study were smokers, it may be suggested that methadone treatment may have heightened P300 amplitudes to some degree over those of nicotine using control subjects.

With regard to control subjects, we found significant P300 amplitude differences in the comparison of healthy subjects with and without nicotine use. Nicotine users showed largely reduced P300 amplitudes at the three electrode sites under investigation. Methadone substituted patients did not differ with regard to concurrent non-opiate drug use and did not differ significantly from control groups. Patients P300 amplitudes were in between controls with and without nicotine use. Latencies of the P300 component did not differ between groups. With regard to behavioral data, healthy subjects without nicotine use showed a low intra-subject reaction-time variability.

Acute nicotine effects have been shown to have positive effects in experimental studies on learning and memory functions in animals as well as in humans [24]. With regard to the visual P300, amplitude increases [25] and latency decreases [26] have been reported with acute nicotine use. In a study on the auditory P300, amplitudes have been shown to be decreased in smokers after several hours of abstinence and to be normalized after smoking [27]. In the Houlihan et al. (1996) study however, neither amplitudes nor latencies were affected with smoking in subjects under short time nicotine abstinence [26].

In contrast to acute effects, chronic nicotine use has been shown to be associated with lower cognitive performance. Cognitive performance worsened with nicotine abstinence especially in those subject to maternal nicotine use during pregnancy [28, 29]. The detrimental effects of chronic smoking on the P300 have been established in two large scale studies demonstrating amplitude reductions with regard to the auditory [18] and the visual P300 [17]. The sample size in the Anokhin study was large and the authors controlled their results for a couple of additional confounding factors. Acute nicotine use remained a predictor of lowered P300 amplitudes and only family density of alcohol dependency emerged as another independent predictor of P300 amplitude. The effects of current alcoholism and drug dependence effects vanished after controlling for nicotine use [17]. While the effect of sustained nicotine use on the P300 is not new, our study demonstrates, that nicotine use has to be controlled for in studies on cognitive function in patient samples. A higher proportion of nicotine use in patient groups as compared to controls may act as a confound, indicating differences between patient and control samples which in fact may be due to differences in nicotine use.

While in our study P300 amplitudes of substituted, smoking opiate addicts did not differ from smoking or non-smoking controls, nicotine had a substantial effect on the P300 in controls. However it remains unclear, whether chronic nicotine reduces P300 amplitudes or whether subjects with lower P300 amplitudes are more prone to nicotine and use it as some form of self medication. Together with the hypothesis, that opiate and methadone use may increase P300 amplitudes, our data support the notion that effects of nicotine, opiates and effects of additional predisposing factors interact in some complex form [17]. While the interaction of these factors can not be modeled within our data, further studies, assessing nicotine and opiate withdrawal as an experimental variable in smoking opiate dependent patients, will be needed to investigate the differential contributions of nicotine and opiate use in these patients.

Taken together, the results of our study suggest that the effect of nicotine among healthy controls exceeds the effect of concurrent non-opioid drug use among methadone substituted opiate addicts. Low P300 amplitudes and high reaction time variability point to decreases in auditory target processing in subjects with current nicotine use. The fact that we found no electrode × group interactions indicates that the effects of nicotine may be a more general one and not be related specifically to frontal or parietal sources of the P300.

When comparing our data to those of previous studies, we find that our P300 amplitudes are considerably high [30, 31]. As could be expected in an auditory oddball paradigm with an active response to target stimuli, amplitudes increased from electrode Fz to Pz. A number of factors may have contributed to high P300 amplitudes in our study: A low target probability of 14%, a long inter-stimulus interval of about 3000 ms, the assessment of subjects during morning hours and the assessment of a group of young adults may have added in their effect upon amplitudes in our study [30, 31].

The clinical relevance of altered P300 amplitudes follows the relevance of the P300 as an indicator of neuronal activity related to cognitive processing [32–34]. Starting with McCarthy et al. (1997) a number of studies assessed the equivalent of the P300 component with functional MRI [35] and showed that even a simple task which requires subjects to press a button or to count the occurrence of a rare target stimulus activates a complex network of neuronal generators, with many of related to attention processing. Functional imaging studies will have to further assess the effects of nicotine and nicotine deprivation on cognitive function which up to now indicate effects on dorsolateral prefrontal cortex [36, 37], which has been identified as part of the generators of the P300 in fMRI experiments [4, 35].

The clinical conclusion of our study is that, as far as the P300 indexes relevant parts of basal cognitive processes related to attention, context updating and memory processes, patients under methadone substitution treatment are not impaired. Some of the patients, those with additional non-opioid drug use, even show larger P300 amplitudes than smoking controls. This result adds to the evidence that methadone substitution treatment may be a reasonable strategy to give these patients a basis to return to or to attain more stable socio-cultural and occupational settings.

In summary, the results of our study revealed that P300 latencies and amplitudes in methadone substituted opiate addicts are within the range of control subjects variation. With regard to reaction times we only found an effect of improved intra-subject reaction time variability in controls without nicotine use. Therefore our data do not support the notion of a lowered auditory target processing in methadone substituted opiate addicts, especially when compared to smoking control subjects. A surprising result of our study was that the effect of nicotine in healthy controls exceeded the effect of additional non-opiate use among patients and the patient/control subject group effects. With regard to the P300 as an indicator of attention dependent information processing, our results indirectly indicate that nicotine may have more detrimental effects than a history of opiate addiction when under methadone treatment. Therefore nicotine use has to be taken into account in further studies on cognitive performance and information processing in psychiatric research. Further studies are needed in order to disentangle the differential effects of nicotine and opiates in patients with substance dependency.