کارکردهای اجرایی در نوجوانان مبتلا به اختلال شخصیت

Abstract Adolescents meeting diagnostic criteria for schizotypal personality disorder (SPD) are presumed to be at risk for developing schizophrenia in adulthood, making them an important group for exploring the developmental trajectory of the disease. Deficits in executive functioning have been documented in schizophrenia patients and adults with SPD. The present study examined executive functions in adolescents with SPD. It was predicted that the SPD group would score below comparison groups (normals and adolescents with other disorders) on measures of executive function, and that those with greater ‘negative’ signs of SPD would show more pronounced performance deficits. Analyses revealed that the performance of the SPD subjects was impaired relative to the other groups on the modified Wisconsin Card Sorting Test (MCST), but not on the Tower of London or the Controlled Oral Word Association Test. Consistent with prediction, regression analyses indicated that MCST deficits were associated with greater negative signs of SPD, but not positive signs.

Introduction The introduction of schizotypal personality disorder (SPD) as a diagnostic category in DSM-III rapidly led to the study of SPD as a strategy for shedding light on the etiology of schizophrenia. Further, it has been suggested that children and adolescents with SPD can serve as a high-risk group for research on the developmental precursors of schizophrenia. Retrospective studies demonstrate an increased incidence of schizotypal symptoms in the premorbid histories of adult schizophrenia patients (Ambleas, 1992, Fish, 1987, Kendler et al., 1981 and Schulsinger et al., 1987). Prospective studies of SPD indicate a poor prognosis, although the rate of eventual schizophrenia has not been established (Nagy and Szatmari, 1986, Stone, 1993 and Wolff, 1991). There is extensive evidence that SPD is part of the schizophrenia spectrum, with both genetic links and similarities in biological and psychological correlates (Siever et al., 1993). For example, youths with SPD show cognitive and biological abnormalities similar to those observed in adult schizophrenia patients. These include formal thought disorder (Caplan et al., 1990a and Caplan et al., 1990b), communication deficits (Caplan and Guthrie, 1992), impairments in abstract thinking (Caplan et al., 1990a and Caplan et al., 1990b), memory deficits (Bergman et al., 1998 and Park and McTigue, 1997), and reduced P300 responses (Erwin et al., 1986). The hypothesis that frontal lobe dysfunction characterizes schizophrenia spectrum disorders has received considerable support. Several studies have demonstrated increases in ventricular volume among schizophrenia (e.g. Andreasen et al., 1982 and Klausner et al., 1992) and SPD patients (Siever et al., 1995) suggestive of a loss of frontal tissue. Neuroimaging studies have also provided direct evidence of structural and functional abnormalities of the frontal lobes in adult schizophrenia patients (e.g. Andreasen et al., 1992, Andreasen et al., 1994, Cleghorn et al., 1989, Ebmeier et al., 1993, Ingvar and Franzen, 1974, Jernigan et al., 1991, Raine et al., 1992a, Raine et al., 1992b, Schroder et al., 1995, Weinberger et al., 1992 and Wolkin et al., 1992) and schizotypal subjects (Brooks et al., 1998, Buschbaum et al., 1997, Raine et al., 1992a, Raine et al., 1992b and Silverman et al., 1992). Paralleling the findings from neuroimaging studies, numerous investigations have shown that schizophrenia is associated with deficits on neuropsychological measures of frontal or ‘executive’ functions, especially the Wisconsin Card Sorting Task (WCST), a widely utilized measure of frontal integrity [for review, see Randolph et al. (1993)]. Schizophrenia patients also show deficits on tests of verbal fluency, a function presumed to be subserved by the frontal cortex (Gruezelier et al., 1988, Hoff et al., 1992 and Morrison-Stewart et al., 1992). The same holds true for the Tower of London (TOL), a test that taps strategic planning (Andreasen et al., 1992 and Goldberg et al., 1990). Like schizophrenia patients, individuals with SPD also score below normal comparison groups on the WCST (Condray and Steinhauer, 1992, Lenzenweger and Korfine, 1994, Lyons et al., 1991, Obiols et al., 1997, Raine et al., 1992a, Raine et al., 1992b, Spaulding et al., 1989 and Suhr, 1997). However, adult SPD subjects do not show performance deficits on some other purported measures of frontal function, such as verbal fluency tests or the TOL (Suhr, 1997 and Trestman et al., 1995). Similarly, some studies of adults with SPD have found no evidence of deficits on measures of general cognitive function, despite significant impairment on the WCST (Trestman et al., 1995 and Voglmaier et al., 1997). Thus, adults with SPD show less pervasive cognitive deficits than schizophrenia patients. Of course, given that the modal age at onset of schizophrenia is early adulthood, the older the adult with SPD, the lower the risk rate for eventual schizophrenia. For most schizophrenia patients, the developmental course involves a progression from schizotypal signs in adolescence to clinical schizophrenia in early adulthood. Thus, adolescents with SPD may be at greater risk for schizophrenia than adults with SPD. Among schizophrenia patients, impairment on measures of executive functioning is differentially associated with symptomatology. Ratings of the severity of ‘negative’ symptoms (e.g. blunted affect and social withdrawal), but not ‘positive’ symptoms (e.g. hallucinations and delusions) (Andreason and Olsen, 1982), are correlated with structural brain abnormalities (Buchanan et al., 1993, Kemali et al., 1987, Seidman et al., 1994 and Siever et al., 1993), ‘hypofrontality’ (Andreasen et al., 1992, Schroder et al., 1995, Weinberger et al., 1992 and Wolkin et al., 1992), and deficits on tests of executive functioning (Bilder et al., 1985, Wagman et al., 1987 and Wolkin et al., 1992). To date, we are aware of only one published report addressing the relation between clinical symptoms in SPD subjects and measures of frontal function (Siever et al., 1993). The investigators found that WCST deficits were associated with ratings of physical anhedonia, social deficits and odd speech, but not perceptual aberration or ‘psychotic-like’ symptoms (e g. magical ideation, ideas of reference, illusions, suspiciousness) in adults with SPD. These above findings suggest that the ‘negative’ features of schizophrenia-spectrum disorders are the clinical expression of ‘frontal’ dysfunction (Levin, 1984, Liddle, 1987 and Siever et al., 1993). It has been proposed that a genetic diathesis to a neurodevelopmental ‘lesion’ of the frontal cortex is involved in both schizophrenia (Weinberger, 1987) and SPD (Siever et al., 1993). More specifically, it has been suggested that the behavioral consequences of this lesion are first apparent during adolescence, when the frontal lobes attain functional maturity, and adult performance levels on measures of executive functioning are achieved (Levin et al., 1991). If this is the case, then the selective relation between frontal function and negative signs should be apparent in adolescence. In the present study, we test this hypothesis by examining the association between clinical symptoms and performance on measures of executive function in adolescents with SPD. Diagnostic group differences are also predicted, such that the SPD adolescents are expected to score below normal controls and subjects with other Axis II disorders on measures of executive function. The battery of executive measures includes the modified WCST (MCST), as well as two purported measures of executive function, the TOL and the Controlled Oral Word Association Test (COWAT), that have not revealed deficits in adults with SPD.

Results Age was used as a covariate in all analyses examining diagnostic group differences. Analyses were conducted both with and without estimated mean IQ as a covariate. 3.1. Intercorrelations among neuropsychological indices Intercorrelations among the neuropsychological measures were examined to determine the feasibility of creating a composite score of executive functioning. The zero-order Pearson correlations are presented in Table 2. As shown, the neuropsychological measures were only weakly to moderately intercorrelated. Estimated IQ was found to be moderately and significantly correlated with the MCST measures, and COWAT. Because of the generally weak correlations among the measures of executive function, a composite index of executive functioning was not derived. Therefore, the measures of executive performance were analyzed separately. Table 2. Intercorrelations among neuropsychological measures for total samplea MCST TOL COWAT Errors PerErrors Categories %Success Words Errors – PerErrors 0.80** – Categories −0.68** −0.68** – %Success −0.26* −0.13 0.24* – Words −0.22* −0.18 0.20 −0.14 – IQ −0.38** −0.44** 0.40** 0.00 0.35** a Errors: MCST errors; PerErrors: MCST total perseverative errors; Categories: MCST categories achieved; %Success: TOL percentage of successful trials; IQ: Estimated WISC-R IQ. * p≤0.05, one-tailed. ** p≤0.01, one-tailed. Table options 3.2. Diagnostic group differences on neuropsychological measures Group means and standard deviations for the neuropsychological measures and effect sizes for group differences are presented in Table 3. All variables derived from the MCST were transformed using the natural log in order to normalize the distribution of the scores. One-tailed tests, using age as a covariate, were employed to test diagnostic group differences. Table 3. Means and standard deviations (in parentheses) for executive measures by diagnostic groupa Measure Diagnostic group Significant differences [effect size]b 1 2 3 SPD OD NC MCST Errors 14.17 6.16 5.33 1>2, 3*** (10.88) (5.54) (6.56) [d=1.35] PerErrors 5.44 1.37 1.33 1>2, 3*** (8.08) (1.67) (2.26) [d=1.82] Categories 4.44 5.53 5.54 1<2, 3** (1.85) (0.77) (1.06) [d=−1.04] TOL %success 0.58 0.66 0.62 n.s. (0.14) (0.16) (0.18) COWAT TotWord 29.17 33.68 31.29 1<2, 3* (9.31) (9.78) (8.40) [d=−0.25] * p<0.10, one-tailed. ** p<0.05, one-tailed. *** p<0.01, one-tailed. a MCST: Modified Card Sort Task; TOL: Tower of London, %success=percentage of successful trials; COWAT: Controlled Oral Word Association Test, TotWord=total number of words on the COWAT. b Effects sizes (d=Mspd−Mnc/SDnc) are presented for differences between the SPD and NC groups. Table options The SPD group performed significantly worse on all measures of the MCST than both the NC and OD groups, which did not differ from each other. Controlling for both age and estimated IQ when comparing the SPD and NC groups did not change the pattern of findings. There were no significant diagnostic group differences for percentage of successful trials on the TOL. Controlling for age and estimated IQ did not change this result. There was a nonsignificant trend (p<0.10) toward lower word production on the COWAT for the SPD group when compared with both the NC and the OD groups. The NC and OD groups did not differ from each other. Subsequent analyses with both age and estimated IQ as covariates did not reveal any significant group differences on the COWAT. 3.3. Association between symptoms and neuropsychological performance Correlational and multiple regression analyses were conducted to determine whether the neuropsychological measures were associated with symptom ratings. Negative symptom ratings were correlated with the score from the COWAT (r=−0.24, p<0.05), as well as the scores from the MCST; errors (r=0.32, p<0.01), perseverative errors (r=0.33, p<0.01) and categories achieved (r=−0.24, p<0.05). The score from the TOL was not related with negative symptoms. Further, none of the performance indices was significantly correlated with ratings of positive symptoms. The same pattern of significant relations was obtained when these analyses were conducted using only the SPD group. Regression analyses were conducted to determine which, if any, of the performance measures made an independent contribution to negative symptom ratings. In the regression analyses, predictor variables were the error, perseverative error and category scores from the MCST, and the number of words generated from the COWAT. (The score from the TOL was not included because the test did not differentiate the SPD group from the OD and NC groups.) Dependent variables were the positive and negative symptom scores. As expected, the equation for the prediction of the negative symptom score reached significance (F=7.26, p<0.05), with MCST perseverative errors accounting for a significant independent portion (t=2.69, p<0.01) of the variance in negative symptom scores.