Background
Treatment of patients with acute mania remains a considerable medical challenge, amongst other reasons because manic patients often lack insight into their illness and into the necessity being treated. Therefore, there is a need for rapidly acting antimanic substances since initial willingness being treated may change and then seriously compromise treatment outcome. Yet, onset of action of medication commonly used in the treatment of mania is delayed for at least several days (e. g. haloperidol, lorazepam, olanzapine, risperidone, valproate) or more (e. g. lithium, carbamazepine) [1-3]. Faster onset of action of antimanic medication would not only shorten the burden of patients but also significantly decrease cost due to shorter duration of hospital stay [1].
While above mentioned substances are sedating, we intend to use a rapidly acting, stimulating substance in this study. This plan is based on the ‘vigilance regulation model of mania’. It takes into account that the vigilance level –vigilance defined as brain arousal– does not only influence behaviour but that in turn, behaviour can also affect the level of vigilance; i.e. a more/less stimulating environment can be actively created in order to increase/reduce vigilance levels. For example, overtired children often develop a hyperactive, talkative and sensation seeking behaviour which can be interpreted as an autoregulatory attempt to stabilise vigilance by increasing external stimulation. We postulate that this physiological autoregulatory mechanism may result in a pathological behavioural syndrome, namely mania, in vulnerable subjects [4-6]. The autoregulatory mechanism might override the physiological tendency to seek sleep, thus aggravating sleep deficits, worsening vigilance instability and thereby starting a vicious circle ending up in full-blown mania (Figure 1) [4].
Figure 1  The vigilance regulation model of mania: Unstable vigilance induces a pathogenic circle with vigilance stabilisation syndrome leading to full-blown mania. Figure taken from: Hegerl U., et al. Are psychostimulants a treatment option in mania? Pharmacopsychiatry 2009; 42:169-174. Reproduced with kind permission of Georg Thieme Verlag KG”.   The outlined model has already been proposed earlier [7,8] and is related to personality theories about extraversion [9] and sensation seeking [10] which comparably explain these traits as an attempt to compensate for low central nervous system arousal.
The ‘vigilance regulation model of mania’ provides an explanation for several seemingly paradoxical aspects and is supported by several lines of evidence: 
• In contrast to the clinical observation of hyperactivity, there is robust evidence from EEG recordings that vigilance is unstable in manic patients: under quiet rest, manic patients show rapid decline to low vigilance stages often already within the first seconds of EEG recording [4]. In line with this, micro sleeps (abrupt intrusion of sleep spindles) were observed shortly after starting recordings [11,12].
• Sleep deficits can trigger or worsen manic behaviour [13,14] and life events disturbing sleep-wake-regulation can trigger or aggravate (hypo)manic syndromes [15]. Accordingly, stabilisation of sleep-wake rhythm is used in behavioural therapies for bipolar disorder [13,16].
• Withdrawal of vigilance-stabilising drugs, such as nicotine, can trigger mania [17,18] and a high smoking prevalence has been reported in bipolar disorder [19]. This may indicate that these patients may benefit from the vigilance-stabilising properties of nicotine. Consistently, smoking and coffee drinking were not associated with higher incidence of mania in bipolar disorder, when confounding factors were controlled for [20].
• In ADHD, a disease with high comorbidity and a broad symptom overlap with mania [21,22], psychostimulants effectively reduce attention deficits, sensation seeking behaviour and hyperactivity, symptoms which are also present in mania [23,24]. In addition, methylphenidate has been shown to improve sleep in children [25,26] and adult patients [27] with ADHD. Furthermore, randomised controlled trials in pediatric patients with ADHD and additional manic symptoms showed that psychostimulants are effective in reducing both ADHD and manic symptoms [28,29].
• There is increasing evidence that psychostimulants are effective in treating mania and that this effect is related to stabilisation of vigilance. A variety of case reports describe a pronounced and rapid improvement of manic symptoms after administration of psychostimulants in adult manic patients with bipolar disorder [4,30-33]. Bschor et al. [34] report on a manic patient with decreased vigilance who showed both a rapid (within 2 hours) and strong improvement of manic symptoms and a stabilisation of vigilance as assed in the EEG after administration of methylphenidate.
• Good response to the psychostimulant modafinil was recently described in a case report. Clinical improvement was observed already after one hour and it was associated with a stabilisation of vigilance [35]. In line with our model, psychostimulants were mostly not efficacious in the treatment of typical depression, except for apathy [36,37].
While psychostimulants have occasionally been linked to induction of mania [38] switches into mania occur rarely: A large evaluation by the U.S. Food and Drug Administration found a psychotic or manic-like reaction only in about 1 of 400 treated patients with ADHD and in the majority of cases (> 90%) the symptoms resolved within 2 days. Thus, the risk of an aggravation of manic symptoms appears to be low [39].
Onset of action of methylphenidate is fast, i.e. faster than 1 hour and maximum effects will be reached after approximately 2 hours [40]. Thus, methylphenidate is likely to provide a much faster effect on manic symptoms than other drugs commonly used in mania.
Taken together, there is increasing evidence from both case studies in adult manic patients and randomised controlled trials in pediatric patients with ADHD and concomitant manic symptoms that psychostimulants such as methylphenidate are effective in the rapid treatment of mania.
Aim of this study, the ‘Methylphenidate in Mania project’ (MEMAP), is to test for the first time the hypothesis that the psychostimulant methylphenidate has antimanic properties after 2.5 days of treatment using a double-blind placebo-controlled design.

Research objectives

Primary objective
The primary objective of the study is to test the hypothesis that methylphenidate immediate release given twice daily (BID) is significantly superior to placebo in the treatment of manic symptoms in patients with bipolar disorder after 2.5 days of treatment as assessed by the Young Mania Rating Scale (YMRS) [41].

Secondary objectives
Secondary objectives of this study comprise to evaluate 
• Whether methylphenidate immediate release given BID is significantly superior to placebo in the treatment of manic symptoms in patients with bipolar disorder after 2 hours of treatment as assessed by the YMRS and the Positive and Negative Syndrome scale - excited component (PANSS-EC) [42].
• The change from baseline to endpoint (after 2.5 days of treatment) on the Clinical Global Impression-Bipolar Scale CGI-BP [43] and the PANSS-EC.
• Whether 2.5 days of treatment with methylphenidate but not with placebo stabilise vigilance regulation as assessed by the ‘Vigilance Algorithm Leipzig’ (VIGALL) [44].
Whether instability of vigilance regulation as assessed by the VIGALL predicts response to methylphenidate.
• Whether methylphenidate immediate release given BID is associated with significantly less movements over the study period than placebo as assessed by actigraphy.
• Whether methylphenidate is significantly superior to placebo in improving cognitive performance as assessed by the “Screen for Cognitive Impairment in Psychiatry” (SCIP) [45].