روابط بین ابعاد علائم و بی قراری در اسکیزوفرنی

Previous studies have suggested that qualitatively distinct aspects of dysphoria (anxiety and depression) are related to specific dimension of schizophrenia symptomatology. Most of these studies used simple dimensions and dysphoria models, although finer distinctions could help defining specific relationships. This study examined the relationships of distinctive aspects of depression and anxiety (both state and trait) with symptom dimensions. Forty patients with a DSM-IV diagnosis of schizophrenia were assessed for symptoms (SAPS-SANS), trait and state anxiety (STAI) and depression (CDS). Symptoms ratings were summarized as dimensional scores according to a two-, three- or five-dimensional models proposed in the literature. The correlation analysis replicates previous observations that distinct aspects of dysphoria are associated with specific dimensions of schizophrenia, with the exception of disorganization. Moreover, controlling for intercorrelated variables revealed that schizophrenia and dysphoric symptoms might act in combination and/or through indirect links to contribute to illness expression. Our data further suggested that these associations may be best understood in terms of interactions between various processing biases alluded in the most recent cognitive accounts of schizophrenia symptoms.

Heterogeneity of symptoms has appeared as a major fact since the early descriptions of schizophrenia by Bleuler (1911) and Kraepelin (1919). In the 1970s, dichotomic concepts differentiating positive and negative forms of schizophrenia have developed based on anatomo-clinical correlations (Crow, 1985 and Andreasen and Olsen, 1982). For several decades, rating scales have been developed to explore the clinical and neural correlates of these two symptom dimensions. More recently, factor analytic techniques applied to rating scales examined the interrelationships between positive and negative symptoms. Review of these studies indicates that three rather than two dimensions better account for the symptomatology seen in schizophrenia (Liddle, 1987, Peralta et al., 1992 and Andreasen et al., 1995). For instance, the ‘Psychomotor Poverty’ dimension (i.e., poverty of speech decreased spontaneous movements and blunting of affect) grossly corresponds to the original concept of negative symptoms, while the positive symptoms subdivides into two dimensions: a ‘Disorganization’ dimension (including thought disorders, inappropriate affect, poverty of content of speech and disorganized behavior) and a ‘Psychotic’ or ‘Reality Distortion’ dimension (hallucinations, delusions and thought disorders). More recent models emphasizing the heterogeneous nature of the psychotic dimension further dissociate an ‘Auditory Hallucinations’ or ‘Schneiderian’ dimension from ‘Bizarre Delusions’ or paranoid dimension (Vasquez-Barquero et al., 1996, Toomey et al., 1997 and Peralta and Cuesta, 1999). This distinction is supported by the observation that hallucinations and delusions can occur independently in schizophrenia as well as in other psychiatric and organic disorders. Furthermore, the heterogeneity of the negative dimension has been also addressed. More specifically, symptoms of ‘Disordered Relatings’ long thought to be independent of other negative symptoms (Strauss et al., 1974) have been identified as an independent dimension in many factor-analytic studies (Keefe et al., 1992, Peralta et al., 1994, Toomey et al., 1997, Nakaya et al., 1999 and Marengo et al., 2000) (Appendix B). Although the physiopathological validity of the main dimensions received strong support from studies showing that each dimension has distinct cognitive (Liddle and Morris, 1991, Cuesta and Peralta, 1995, Norman et al., 1997 and Guillem et al., 2001), structural (Chua et al., 1997 and McIntosh et al., 2001), metabolic (Liddle et al., 1992 and Kaplan et al., 1993) and neurophysiological correlates (Harris et al., 1999 and Williams et al., 2000), the number of relevant dimensions remains a matter of debate (Smith et al., 1998 and Stuart et al., 1999). Apart from methodological differences, the discrepancies between studies can be associated with dimensional variations according to the patient's status (first episode vs. chronic schizophrenia) (McGorry et al., 1998 and Vasquez-Barquero et al., 1996), phase of the illness (remitted vs. acute phase) (Mellers et al., 1996, Nakaya et al., 1999 and Lançon et al., 2000), age (Sauer et al., 1999, Schultz et al., 1997 and Marengo et al., 2000), illness onset and duration (early vs. late onset) (Häfner et al., 1998 and Mojtabai, 1999) and antipsychotic medication status (Molina-Rodriguez et al., 1998 and Meagher et al., 2001). In summary, there is accumulating evidence that nonspecific temporal factors and/or state-dependent factors interact with the symptoms of schizophrenia to determine the expression of the illness. It is possible that dysphoric states play an important role as factors in determining the expression of schizophrenia. Dysphoria includes both anxiety and depression (Stedman, 1995 and Taber, 1997). Similar to the symptom or dimensions of schizophrenia, it is well known that both aspects of dysphoria vary from time to time (i.e., mood states) as well as across the life span (Krasucki et al., 1998, Henderson et al., 1998, Karel, 1997 and Form, 2000). Research in schizophrenia has also typically associated anxiety with positive symptoms (Ciompi, 1994 and Kingdon and Turkington, 1994) and depression with negative symptoms (Knights and Hirsch, 1981 and Zubin, 1985). In spite of the growing importance of the dimensional approach of schizophrenia, only a few studies have attempted to elucidate the actual relationships between the symptom dimensions and the various aspects of dysphoria. In their ‘pyramidical model of schizophrenia’, Kay and Sevy (1990) initially inferred a ‘depressive’ dimension including both the anxious and depressive aspects of dysphoria as a key element contributing to the paranoid expression when associated to positive symptoms and to the catatonic expression when associated with negative symptoms. However, the first study to examine directly the correlations between dysphoria and positive vs. negative symptoms was that of Norman and Malla (1991) who found dysphoria to be significantly related to positive but not negative symptoms. Subsequently, in a longitudinal study, these same authors (Norman and Malla, 1994a and Norman and Malla, 1994b) reported a greater likelihood of positive correlations in changes over time between dysphoria and positive rather than negative symptoms. This association has been replicated later by others (Lysaker et al., 1995 and Nakaya et al., 1997) and by Sax et al. (1996) who reported evidence of a more specific correlation between depression and positive symptoms. In a more recent study, Norman et al. (1998) examined further these relationships using a three-dimensional model of schizophrenia symptomatology while separating the anxious and depressive aspects of dysphoria. The results of Norman et al. showed a significant correlation between anxiety correlates and the Psychotic (or Reality Distortion) dimension, whereas depression was found to correlate with the negative dimension. On the other hand, they failed to replicate the previously described association between depression and positive symptoms when controlling for anxiety. The authors did not exclude the possibility that part of their results were due to extrapyramidal side effects of antipsychotic medication, a likely possibility since the instruments used (Hamilton's and Beck's depression and anxiety rating scales) were not specifically designed for schizophrenia. As suggested by Lançon et al. (2000), the evaluation of depressive symptoms should be conducted with more specific instruments that are less contaminated by extrapyramidal symptoms such as the Calgary Depression Scale (CDS) (Addington et al., 1993 and Reine et al., 2000). Conversely, Norman et al. indicated that it is unclear whether anxiety results from rather than causes psychotic symptoms (i.e., hallucination and delusions), or if, as suggested in other studies, the influence occurs in the opposite direction (Ciompi, 1994, Norman and Malla, 1994a and Norman and Malla, 1994b). Perhaps, introducing the distinction between state and trait anxiety, as assessed by the State-Trait Anxiety Inventory (STAI) (Spielberger, 1966 and Spielberger et al., 1970), could prove useful to disentangle these possibilities. For instance, trait anxiety, defined as the subject's proneness to develop anxious reaction, could well have a causal effect, whereas state anxiety, which is a transitory response to stressful experience, could better be viewed as a resulting effect of symptoms. These also raise the possibility that complex dimensional models may conceivably be more accurate than the classical three-dimensional view. For instance, given the transitory nature of hallucinations compared to that more continuous of delusions, models dissociating these symptoms (e.g., Toomey et al., 1997) could prove particularly relevant with respect to the state vs. trait aspects of the illness. The distinction between the Diminished Expression dimension that includes motor symptoms likely affected by depression and/or side effects of medication and the Disordered Relatings dimension concerned with social adaptation symptoms often present before the onset of the illness (Baum and Walker, 1995 and Cuesta et al., 1999) could be also relevant in the same respect. The current study was designed to take into account these different aspects. We have investigated the correlations of unique aspects of depression, state and trait anxiety as measured by the CDS and STAI to symptom dimensions of schizophrenia. The symptom dimensions have been defined according to different models including the classical negative/positive dichotomy, an exemplar of the three-dimensional model (Andreasen et al., 1995), and the more complex five-dimensional model proposed by Toomey et al. (1997).