The observed responsivity of primary OCD to serotonergic antidepressants largely contributed to the serotonin hypothesis of OCD (Bandelow et al. 2008; Barr et al. 1992; Linden 2006; Nakao et al. 2005; Saxena et al. 1999, 2009). In consequence, SSRIs and SNRIs (serotonin and noradrenaline reuptake inhibitor) have become an integral part of international treatment guidelines for OCD (Koran et al. 2007; Kordon et al. 2011). Based on the assumption of a pathophysiological overlap between primary OCD and secondary comorbid OCS observed in other psychiatric conditions, it seems plausible to modulate the serotonergic neurotransmission in order to attain symptom relief. The cumulative evidence for this approach (Hwang et al. 2009) within schizophrenia-spectrum disorders over the past decades is summarized in Table 11.1. Overall, augmentation strategies with antidepressants have been evaluated in N = 132 patients with schizophrenia and co-occurring OCS. However, the comparability is restricted by methodological reasons: While there are numerous single and multiple case reports, some trials also employed active (Reznik and Sirota 2000b; Sayeed Khan et al. 2004) or placebo (Berman et al. 1995) controlled conditions. In a three-armed trial, adding anti-obsessive antidepressants to FGA treatment was superior to the treatment with FGAs or SGAs alone (Sayeed Khan et al. 2004). In detail, the tricyclic antidepressant clomipramine (Bark and Lindenmayer 1992; Berman et al. 1995; Kurokawa and Tanino 1997; Margetic et al. 2008; Poyurovsky and Weizman 1998; Zohar et al. 1993) and the following SSRIs have been examined: fluvoxamine (Gahr et al. 2014; Poyurovsky et al. 1999; Reznik and Sirota 2000a, b), sertraline (Hoehns et al. 2001; Poyurovsky et al. 2003), escitalopram (Stryjer et al. 2012) and fluoxetine (Kulkarni et al. 2012). Recently, beneficial effects of the SNRI milnacipran were reported (Hung and Hung 2014). The results of these trials were not entirely homogeneous but showed some overall positive tendencies despite the possibility of a publication bias. Several descriptive observations (de Haan et al. 1999) are also considered.

Adding antidepressants in bipolar disorder (BD) (see also Sect. 11.2.3.2) bears the risk of inducing manic switches. In contrast, the risk of aggravating the primary psychotic symptoms in schizophrenia seems negligible: To date, there are only two publications reporting deteriorations of psychotic symptoms after the add-on of clomipramine (Margetic et al. 2008; Zohar et al. 1993). This is in line with the current perception that antidepressants and even pro-dopaminergic substances such as bupropion (Englisch et al. 2010a, b, 2013) may be applied for the treatment of NS in schizophrenia without jeopardizing the patients’ mental state, given they are on stable antipsychotic treatment.

Specific non-psychiatric safety concerns that need to be considered: One patient died from sudden cardiac arrest during polypharmacy with CLZ, RISP, sertraline and lorazepam (Hoehns et al. 2001). Adding pro-serotonergic antidepressants for OCS in schizophrenia renders some additional issues: (1) Clomipramine extends the anticholinergic profile of agents such as CLZ or OLZ and therefore may increase the risk of cardiovascular AEs. (2) SSRIs such as fluoxetine or fluvoxamine bear the potential of marked pharmacokinetic interactions within the cytochrome P450 system. This may result in a rise of antipsychotic serum levels which, in turn, might increase the antiserotonergic effects of CLZ and OLZ in particular and thus aggravate OCS.

Except for milnacipran, for SNRIs such as venlafaxine, duloxetine or vortioxetine, no systematic evaluations have been performed to date. As these agents bear the additional potential of enhancing neurocognition while the risk of pharmacokinetic interactions appears to be low, it will be important to systematically assess these multimodal antidepressants in order to improve the pharmacological treatment of OCS in schizophrenia.

Mood stabilizers (MS) such as lithium, valproic acid, carbamazepine, lamotrigine and topiramate are the treatment of choice for bipolar disorder and were also effective in reducing OCS secondary to BD (Abdel-Ahad and Kazour 2013). As recently summarized, they have also been used in primary OCD, both as a monotherapy or as an augmentation of non-selective and selective SRIs. The underlying neurobiological mechanisms remain speculative; however, they reach far beyond the serotonergic system. Lamotrigine, for instance, exerts its effects by modulating the glutamatergic neurotransmission (Poyurovsky et al. 2010).

Within the schizophrenia spectrum, the available data on MS is less conclusive (Citrome 2009; Tiihonen et al. 2009b; Zink et al. 2010): Recent meta-analyses on lithium (Leucht et al. 2007), valproic acid (Schwarz et al. 2008), carbamazepine (Leucht et al. 2014) and lamotrigine (Tiihonen et al. 2009b) were unable to confirm their therapeutic effects in schizophrenia. While they may be suited to treat specific conditions such as agitation, aggression or tardive dyskinesia, they are generally considered an off-label treatment for psychotic disorders. Further research is required in order to define their therapeutic effects in schizophrenia on a syndromal level which might also embrace comorbid OCS.

While there are no reports on pregabalin, lithium, topiramate and carbamazepine, descriptions of a small number of patients (N = 18) with low level of evidence suggest a relief of OCS in schizophrenia when SGAs are augmented with valproic acid (Aukst-Margetic et al. 2011; Bisol and Lara 2009; Canas et al. 2012; Suppes and Rush 1996; Zink et al. 2007) or lamotrigine (Poyurovsky et al. 2010; Rodriguez et al. 2010) (see Table 11.2). In addition to the small number of published cases, their narrative design and a putative publication bias pose further limitations. To date, no controlled trials have been performed to evaluate the efficacy and tolerability of augmentation strategies with mood stabilizers for comorbid OCS in schizophrenia. In addition to investigating changes in psychopathological features, further research should also focus on pharmacokinetic interactions as well as additive AEs which may arise from polypharmacy.

While there have been recent efforts to implement CLZ therapy earlier in the course of schizophrenia (Leucht et al. 2012; Remington et al. 2013), it is still considered a substance of last resort when other treatment options fail. Thus, patients receiving CLZ are usually the most severely affected, and they also carry the highest risk of developing second-onset OCS (see Chap.​ 10).

CLZ-treated patients have been in the focus of extensive research, and the evidence for an optimized CLZ treatment employing means of polypharmacy has been summarized in several recent reviews (Kane 2011; Nielsen et al. 2011; Zink et al. 2010). While switching to other SGAs in monotherapy could not be corroborated in clinical trials, it has been shown that CLZ-treated patients with second-onset OCS may benefit from combination strategies with other SGAs (see also Sect. 11.2.1).

As CLZ is a substance with weak antidopaminergic but prominent antiserotonergic properties, it has been proposed to add agents with mainly dopaminergic modes of action in order to counteract pro-obsessive effects inherent to CLZ. In particular, there is some evidence supporting the add-on of APZ (Chang et al. 2008; Englisch et al. 2009; Englisch and Zink 2008; Eryilmaz et al. 2013; Glick et al. 2009; Peters and de Haan 2009; Rocha and Hara 2006; Villari et al. 2011; Zink et al. 2006) and ZIPR (Krause et al. 2013). Based on theoretical considerations, the combination with AMS might also be beneficial (Kim et al. 2008) (see Table 11.3). With respect to these substances, the highest level of evidence exists for APZ, where positive effects have been reported in a number of cases and in one RCT (Chang et al. 2008). However, in general and as previously mentioned (compare sections “Antidepressive substances” and “Mood stabilizers”), the number of cases is limited and may be subject to publication biases. Therefore, conclusions should be drawn with caution.