This study included consecutive 15 patients with Asperger’s disorder (DSM-IV-TR) who visited the outpatient clinic of the Department of Neuropsychiatry at Fukushima Medical University Hospital; they were diagnosed as having ADHD symptoms according to the DSM-IV-TR criteria except for criterion E (Table 1). The lead author of this report was solely responsible for diagnosing. The subjects consisted of 7 men and 8 women with the mean age ± standard deviation age of 30.40 ± 7.63 years (range, 19 - 49 years). Written informed consent was obtained from all subjects prior to participation.

Milnacipran was prescribed within the dose ranges used in common clinical practice. Dose of methylphenidate administered prior to milnacipran remained unchanged during milnacipran administration.

GAF scores were determined before and 12 weeks after milnacipran. The symptoms were assessed with Conners’ Adult ADHD Rating Scales-Objective-Screening Version (CAARS-O-SV) [32] . One of the coauthors (SI) purchased CAARS-O-SV from the commercial source.

ANOVA was performed, being GAF scores/CAARS scores as the dependent variables and milnacipran/me-

GAF: Global Assessment of Functioning; CAARS-O-SV: Conner’s Adult ADHD Rating Scales ScreeningVersion; MIL: milnacipran; MPH: methylphenidate.

thylphenidate as the factors. In order to determine the relationship between the changes in GAF scores and changes in clinical symptoms, multiple regression analysis was performed using “changes in GAF scores (difference from baseline)” (N = 15) as the dependent variable. Multiple regression analysis involved symptom assessment focusing on “inattention”, “hyperactivity”, “impulsivity” and “ADHD index”.

The changes in social functioning as assessed by GAF scores with treatment were examined. The GAF score (mean ± SD) in the 15 patients improved significantly from 44.00 ± 8.49 before therapy (N = 15) to 65.00 ± 9.26 after milnacipran (F = 33.89, df = 1, p < 0.001).

Analysis of variance (ANOVA) using GAF scores and CAARS-O-SV scores as the independent variables was performed employing “before/after milnacipran” and “with/without methylphenidate” as the factors. As a result, significant effects of “before/after milnacipran” were observed on GAF, inattention, inattention + hyperactivity/ impulsivity and ADHD index (F = 33.89, df = 1, p < 0.001; F = 21.20, df = 1, p < 0.001; F = 18.25, df = 1, p < 0.001; F = 70.64, df = 16, p < 0.001; respectively). No significant effects of “before/after milnacipran” were observed on hyperactivity/impulsivity. Moreover, no significant effects of “with/without methylphenidate” were

GAF: Global Assessment of Functioning; CAARS-O-SV: Conners’ Adult ADHD Rating Scales Objective Screening Version; MIL: milnacipran; *p < 0.001 (refer to Table 3).

GAF: Global Assessment of Functioning; CAARS-O-SV: Conners’ Adult ADHD Rating Scales Screening Version; MIL: milnacipran; MPH: methylphenidate.

observed on GAF, inattention, hyperactivity/impulsivity, inattention + hyperactivity/impulsivity and ADHD index. Furthermore, no significant interactions were observed between “before/after milnacipran” and “with/ without methylphenidate”.

In order to determine the relationship between changes in social functioning and changes in clinical symptoms, multiple regression analysis was performed using changes in social functioning (i.e., GAF scores) (N = 15) as the dependent variable. “Changes in inattention”, “change in hyperactivity/impulsivity”, “change in inattention + hyperactivity/impulsivity” and “change in ADHD index” were chosen as the independent variables. The regression was not significant.

In the present study, we prescribed milnacipran to 15 adult Asperger’s disorder patients with ADHD symptoms and evaluated the results of milnacipran. To our knowledge, this is the first report of clinical experience with milnacipran in adult Asperger’s disorder with ADHD symptoms. The results of our study suggested that milnacipran produces improvement in social functioning in patients having ADHD symptoms associated with Asperger’s disorder, particularly through improvement of inattention.

Recently, Gadow et al. [31] compared the clinical characteristics of ADHD associated and not associated with PDD, and reported that the ADHD symptoms was identical in both the groups. Their finding supports the existence of the ADHD symptoms in patients with PDD. As described earlier, previous reports by other authors demonstrated that methylphenidate exhibited efficacy for ADHD symptoms in PDD patients, and that PDD patients with ADHD symptoms were more prone to show adverse effects to methylphenidate as compared to ADHD patients without PDD symptoms. Although it is not clear why adverse effects of methylphenidate are more likely to occur in PDD patients than in ADHD patients, possible explanations include a difference in the pathophysiology between ADHD and PDD. A study on ADHD symptoms associated with Asperger’s disorder by the Research Units on Pediatric Psychopharmacology (RUPP) has been performed [5] .

Methylphenidate exerts its actions by inhibiting dopamine and noradrenaline uptake. In contrast, milnacipran exerts its effects by inhibiting noradrenaline and serotonin reuptake. Thus, while milnacipran, unlike methylphenidate, does not inhibit dopamine reuptake, it inhibits serotonin reuptake, which might be critical in the treatment of PDD. Thus, inhibition of serotonin reuptake, an action not exerted by methylphenidate, is one of the great advantages of milnacipran. Milnacipran with its action of inhibiting serotonin reuptake may be expected to be more useful than methylphenidate for improving the ADHD symptoms in patients of PDD.

Duloxetine, milnacipran, nefazodone and venlafaxine are clinically available SNRIs in the world. Although these SNRIs differ with respect to their potentials, they share inhibiting abilities of serotonin and noradrenaline reuptake. While clinical experiences of the use of venlafaxine for the treatment of ADHD and Asperger’s disorder has been frequently reported, information on the clinical use of duloxetine, milnacipran and nefazodone for ADHD and Asperger’s disorder is limited among the existing literature. Nefazodone is not marketed, and duloxetine commenced to be marketed in 2010 in Japan. We do not have enough experience of duloxetine in patients with developmental disorders. Therefore, we examined the effects of milnacipran for the treatment of ADHD symptoms among Asperger’s disorder.

Methylphenidate, which has long been used for ADHD symptoms enhances the activity of not only dopamine but also of noradrenaline. This is consistent with the multiple lines of evidence implicating the central DA and NE systems in the pathophysiology of ADHD [33] . On the other hand, methylphenidate is speculated to improve ADHD symptoms (particularly inattention) via increasing the activity of noradrenaline in the brain [2] -[5] . In addition, the wide involvement of noradrenaline in the pathophysiology of ADHD such as inattention, hyperactivity and impulsivity is supported by the fact that atmoxetine, a selective noradrenaline re-uptake inhibitor, is appreciated as an effective agent for ADHD. Thus, the potential inhibiting ability of noradrenaline re-uptake of milnacipran is thought to underpin its effects for ADHD symptoms. Comparison of the symptoms most likely to show improvement after venlafaxine and milnacipran treatment remains to be addressed in future studies.

On the other hand, in Asperger’s disorder, the improvement of the autistic symptoms might be produced via increasing the serotonin activity [18] . Serotonin is known to play critical roles in wide variety of psychic functions, namely aggression, anxiety, mood, impulsivity, sleep, eating behavior, reward system, and psychosis [34] . In particular, a growing body of evidence links autism spectrum disorders to abnormalities in serotonin function, and the selective serotonin reuptake inhibitors (SSRIs) have been utilized to various symptoms of the disorders. Therefore, the SNRI milnacipran, which enhances the activity of both serotonin and noradrenaline, may be expected to improve both ADHD symptoms and autistic symptoms in patients of Asperger’s disorder [34] .

Furthermore, in the present study, the multiple regression analysis of factors influencing the improvement of social functioning revealed that both “improvement of inattention” and “improvement of difficulties in interpersonal relationships” were contributory factors. This result is thought to clearly lend support to our above presumption.

Milnacipran is associated with a lower incidence of adverse effects, and is thus tolerated better than tricyclic antidepressants. The summation of adverse effects from double-blind controlled studies shows that the incidence of discontinuation of therapy due to adverse effects was 7.6% for milnacipran (1871 cases) and 14.8% for tricyclic antidepressants (940 cases) [35] . In the present study, the incidence of dropout due to adverse effects of milnacipran was similar, that is, 6.25% (1 case). This figure is lower than the dropout rate for methylphenidate therapy (18%) reported in the literature [5] .