Obsessive–compulsive symptoms in schizophrenia: associated clinical features, cognitive function and medication status

Abstract

Objective

To determine the prevalence and clinical significance of obsessive–compulsive (OC) symptoms among a group of stable outpatients with schizophrenia.

Methods

We studied 118 patients with schizophrenia from an urban clinic, characterized using clinical symptoms scales, including the Yale–Brown Obsessive–Compulsive Scale (Y-BOCS), and neuropsychological testing. We categorized patients into three groups according to severity of OC symptoms and used multivariate linear regression and chi-square tests to compare groups on variables of interest.

Results

Only 10 patients (8.8%) had Y-BOCS scores greater than 16, a standard criterion for OCD studies. The patient group with the most OC symptoms (Y-BOCS scores >11) scored higher on the Hamilton Depression Scale, the positive symptoms subscale of the Positive and Negative Syndromes Scale (PANSS) and its delusions item, but not on any of the neuropsychological tests compared to the other two groups. Patients with most severe compulsive symptoms (but not OC symptoms together, or obsessions alone) were more likely to be treated with olanzapine or clozapine, atypical antipsychotic medications previously reported to induce or worsen OC symptoms.

Conclusions

Our results confirm previous findings that patients with schizophrenia and comorbid OC symptoms have more positive symptoms but not the suggestion that such patients are more cognitively impaired than their counterparts without OC symptoms. We suggest possible explanations for discrepancies in the literature, including differences in patient sampling and definition of comorbid OC symptoms. Finally, our data suggest that olanzapine and clozapine may produce or worsen compulsions in some patients; prospective studies need to address this possibility.

Introduction

Obsessive–compulsive (OC) symptoms are common in patients with schizophrenia. First recognized early in the 20th century (Jahreiss, 1926, Rosen, 1957), modern interest in this topic can be traced to Fenton and McGlashan (1986) who found clinically significant OC symptoms in 12.9% of 163 hospitalized schizophrenia patients, a prevalence rate well above what would have been expected by simple chance comorbidity. Subsequent studies have reported widely varying rates of comorbid OC symptoms in schizophrenia patients (Bermanzohn et al., 2000, Fabisch et al., 2001). In addition, several studies have found that schizophrenia patients with OC symptoms may exhibit more severe psychotic symptoms and poorer cognitive function compared to other patients with schizophrenia. Indeed, it has been suggested that these patients may have a subtype of schizophrenia with distinct pathophysiology, treatment response and clinical course (Berman et al., 1999).

We summarize relevant features of the 13 published studies which have examined demographic, clinical and cognitive characteristics in schizophrenia patients with OC symptoms (Table 1; Berman et al., 1995, Berman et al., 1998, Eisen et al., 1997, Hermesh et al., 2003, Hwang et al., 2000, Kruger et al., 2000, Lysaker et al., 2002, Lysaker et al., 2000, Nechmad et al., 2003, Ohta et al., 2003, Poyurovsky et al., 1999, Poyurovsky et al., 2001, Tibbo et al., 2000). The samples range in size from 20 to 108 subjects (mean 58) and include first-episode patients (Poyurovsky et al., 1999), chronic state hospital inpatients (Hwang et al., 2000) and outpatients recruited by advertisement (Tibbo et al., 2000). Several used a Structured Clinical Interview for DSM-IV (SCID) diagnosis of OCD to define clinically significant OC symptoms, while others relied on Yale–Brown Obsessive–Compulsive Scale (Y-BOCS) scores of greater than 7, 8 or 17. There have been no large studies examining both clinical and cognitive status in stable outpatients.

Most studies found differences in clinical characteristics of those patients with schizophrenia and comorbid OC symptoms compared to patients with schizophrenia alone. However, the pattern of findings was highly variable; some reported less severe positive and negative symptoms, others reported no differences, and still others reported more severe positive and negative symptoms in this population. With regard to cognitive functioning, most studies found an association between comorbid OC symptoms and impaired performance on tests of frontal lobe function. Impaired performance on the Wisconsin Card Sort Test (WCST) was the most commonly reported finding; a higher rate of perseverative errors only was found in some studies and nonperseverative errors only in others. These inconsistencies prompted Lysaker et al. (2002) to conclude: “Thus, the data so far suggest that OC symptoms in schizophrenia spectrum disorders are linked with a distinctive clinical picture, but the exact nature of that clinical picture remains unclear”.

The relationship between OC symptoms and atypical antipsychotic medications is similarly unclear. Atypical antipsychotics may be useful adjunctive therapy in treatment-refractory OCD but clozapine, risperidone, olanzapine and quetiapine may also induce or exacerbate OC symptoms in patients with schizophrenia (Lykouras et al., 2003). Predictably, the number of reports involving each medication appears to be proportionate to the length of time the medication has been available (Lykouras et al., 2003). Ziprasidone and aripiprazole have not yet been reported to exacerbate or ameliorate OC symptoms in patients with schizophrenia. Emergence of OC symptoms during treatment with atypical agents may be due to antiserotonergic properties of these medications. Following a review of the literature, we hypothesized a priori that clozapine and olanzapine would be more likely to be associated with OC symptoms than other antipsychotic medications (see Discussion).

We set out to study the clinical and cognitive characteristics of stable schizophrenia outpatients with OC symptoms. Our sample was well characterized in terms of clinical measures, cognitive functioning and current medications. Thus, we were interested in examining the cross-sectional relationships between OC symptoms and psychosis, cognition and medication status.