Controlled trials in patients with bulimia nervosa have demonstrated efficacy of antidepressant medications with serotonergic function (e.g. fluoxetine) as well as noradrenergic function (e.g. desipramine). Sixteen out-patients with bulimia nervosa according to DSM-IV criteria were treated in a drug surveillance with 100 mg of milnacipran, a specific serotonin and noradrenaline reuptake inhibitor (SNRI). Ten patients completed the 8-week observation period. The reasons for premature attrition were improvement in one patient (no. 12), a generalized exanthema in one patient (no. 7), severe nausea in one patient (no. 8) and non-compliance due to non-drug-related reasons in three patients (no. 1, 2, and 16). An intent-to-treat analysis exhibited a significant reduction in weekly binge eating and vomiting frequency from baseline to the end of treatment. Three patients stopped binge eating and purging completely during the last week of treatment. Furthermore, there was a concomitant decrease of depression ratings (HAMD, BDI). Our preliminary data give rise to the notion that milnacipran may be promising in the treatment of bulimia nervosa.
From a neurobiological perspective, research suggests that the serotonergic system may play an important role in the pathophysiology and perhaps etiopathogenesis of eating disorders. Disturbances of brain serotonin activity have been described in acutely ill as well as long-term recovered patients with bulimia nervosa (Goldbloom et al., 1990, Brewerton et al., 1992, Levitan et al., 1997, Jimerson et al., 1997, McBride et al., 1991, Monteleone et al., 1998 and Smith et al., 1999, Tauscher et al., 2001). There is evidence for an impaired serotonergic responsiveness which largely but not entirely seems to normalize following remission (Wolfe et al., 2000; Kaye et al., 1998; Weltzin et al., 1994 and Weltzin et al., 1995; Oldman et al., 1995).
However strong the link between eating disorders and serotonergic activity seems to be, it is unlikely that a specific dysfunction of one neurotransmitter system can fully explain the pathogenesis of bulimia nervosa. There is also evidence, although less strong, for a disturbance of noradrenergic function in normal-weight bulimic subjects (Kaye et al., 1990; Pirke, 1996).
There is a substantial body of literature demonstrating that MAO inhibitors, tricyclic antidepressants, and selective serotonin reuptake inhibitors (SSRIs) are effective in treating patients with bulimia nervosa (BN) (Lennkh et al., 1997, Fichter, 1993 and Walsh et al., 1997). It seems possible that antidepressants correct a disturbance of monoaminergic function in patients with eating disorders. Among the SSRIs, fluoxetine has been investigated most extensively and has demonstrated a significant superiority over placebo in reducing binge eating and vomiting frequency. In addition, fluoxetine has a favorable side-effect profile (Fluoxetine Bulimia Nervosa Collaborate Study Group, 1992 and Goldstein et al., 1995). Other SSRIs like fluvoxamine, paroxetine, citalopram, and sertraline seem to have a similar efficacy in patients with eating disorders, however have not been studied to the same extent as fluoxetine.
Even though SSRIs are currently recommended as the first-line pharmacological treatment in BN, antidepressants with noradrenergic activity also are effective in BN. Imipramine (Agras et al., 1987) and desipramine were found to be significantly superior to placebo in reducing binge eating and purging frequencies and in improving the associated psychopathology in short-term treatment studies. Desipramine is a tricyclic secondary amine antidepressant with preferential affinity for the noradrenergic reuptake sites. In double-blind, placebo-controlled studies with desipramine, the reduction in binge eating frequencies ranged from 40 to 90% (Hughes et al., 1986; Barlow et al., 1988; Blouin et al., 1988; Walsh et al., 1991). The tricyclic antidepressant desipramine has a number of pharmacodynamic properties additionally to the noradrenergic one, which are linked to burdensome side effects, specifically for the necessary long-term outcome. In this line, high drop out rates—up to 50%—have been reported in studies employing desipramine, which were mainly due to side effects (Leitenberg et al., 1994; Barlow et al., 1988; Walsh et al., 1991; Agras et al., 1992).
We recently published the results of a drug surveillance using 8 mg of reboxetine, a novel selective noradrenaline reuptake inhibitor in seven patients with bulimia nervosa (El-Giamal et al., 2000). The monthly binge eating frequency showed a reduction of 73% and the frequency of vomiting episodes per month decreased by 67%. The only clinically relevant adverse effect was constipation which led to early attrition in two patients.
The present drug surveillance intended to measure the efficacy and tolerability of milnacipran in the short-term treatment of patients with BN. Milnacipran is a dual acting antidepressant which inhibits the reuptake of both serotonin and noradrenaline with approximately equal potency with no effect on the reuptake of dopamine and no action at any pre- or postsynaptic receptors (Montgomery et al., 1996). It is rapidly absorbed when taken orally, has an intermediate half-life of about 8 h and no active metabolite. It does not modify, and is not affected by, the cytochrome P450 system (for review see Briley, 1998). Data from placebo-controlled trials and the results of comparator studies involving TCRs and SSRIs have confirmed that milnacipran is an effective and well-tolerated antidepressant, particularly useful in patients with severe depression (Kasper, 1997; Kasper et al., 1996).
