[Psychopathy in childhood and adolescence]. / Psychopathie im Kindes- und Jugendalter.

Personality disorders manifest themselves in stable deviations of personality traits that especially arise in negative interactions with others. Those typical dysfunctional interactions can be observed particularly in dissocial and antisocial personality disorders: adults with severe forms of personality disorders often showed specific symptoms in childhood and adolescence. Clinical researchers therefore demand a respective diagnosis of childhood and adolescent psychopathy. There is an extensive body of research on dissocial and borderline personality disorders in children and adolescents. To date the severe form of dissocial personality disorders, psychopathy, cannot be specifically classified. The transfer of this diagnosis to childhood and adolescence is crucial: based on the assumption of persistence and the risk of stigmatisation many clinicians refuse to diagnose psychopathy at a young age. On the other hand there are positive treatment outcomes that aim at symptom reduction from a very early age. The specification of dissocial personality disorders in childhood and adolescence may encourage the development of new treatments and the de-stigmatisation as well as qualify the dogged assumption of persistence of personality disorders.

First genome-wide association scan on neurophysiological endophenotypes points to trans-regulation effects on SLC2A3 in dyslexic children.

Dyslexia is one of the most common learning disorders affecting about 5% of all school-aged children. It has been shown that event-related potential measurements reveal differences between dyslexic children and age-matched controls. This holds particularly true for mismatch negativity (MMN), which reflects automatic speech deviance processing and is altered in dyslexic children. We performed a whole-genome association analysis in 200 dyslexic children, focusing on MMN measurements. We identified rs4234898, a marker located on chromosome 4q32.1, to be significantly associated with the late MMN component. This association could be replicated in an independent second sample of 186 dyslexic children, reaching genome-wide significance in the combined sample (P = 5.14e-08). We also found an association between the late MMN component and a two-marker haplotype of rs4234898 and rs11100040, one of its neighboring single nucleotide polymorphisms (SNPs). In the combined sample, this marker combination withstands correction for multiple testing (P = 6.71e-08). Both SNPs lie in a region devoid of any protein-coding genes; however, they both show significant association with mRNA-expression levels of SLC2A3 on chromosome 12, the predominant facilitative glucose transporter in neurons. Our results suggest a possible trans-regulation effect on SLC2A3, which might lead to glucose deficits in dyslexic children and could explain their attenuated MMN in passive listening tasks.

Association study of energy homeostasis genes and antipsychotic-induced weight gain in patients with schizophrenia.

INTRODUCTION: Marked inter-individual variation has been observed with respect to the risk of weight gain and related metabolic disturbances during antipsychotic treatment, which in part could be explained by heritability. Such adverse effects have been proposed to occur through drug-induced mechanisms involving both the central nervous system and different peripheral tissues. METHODS: We genotyped tagSNPs in several genes ( ADIPOQ, PRKAA1, PRKAA2, PRKAB1, PRKAG1, PRKAG2, PRKAG3, FTO and FABP3) that regulate lipid and energy homeostasis for their possible association to antipsychotic-induced weight gain. RESULTS: In a sample of 160 patients of German origin with schizophrenia who had been monitored with respect to body weight, we found marked association between antipsychotic-related changes in BMI and 6 markers in the adiponectin gene ( ADIPOQ). DISCUSSION: These findings support previous observations (in patients' serum) that adiponectin is involved in antipsychotic-mediated metabolic adverse effects.

Association between the insulin-induced gene 2 (INSIG2) and weight gain in a German sample of antipsychotic-treated schizophrenic patients: perturbation of SREBP-controlled lipogenesis in drug-related metabolic adverse effects?

Atypical antipsychotics are nowadays the most widely used drugs to treat schizophrenia and other psychosis. Unfortunately, some of them can cause major metabolic adverse effects, such as weight gain, dyslipidemia and type 2 diabetes. The underlying lipogenic mechanisms of the antipsychotic drugs are not known, but several studies have focused on a central effect in the hypothalamic control of appetite regulation and energy expenditure. In a functional convergent genomic approach we recently used a cellular model and demonstrated that orexigenic antipsychotics that induce weight gain activate the expression of lipid biosynthesis genes controlled by the sterol regulatory element-binding protein (SREBP) transcription factors. We therefore hypothesized that the major genes involved in the SREBP activation of fatty acids and cholesterol production (SREBF1, SREBF2, SCAP, INSIG1 and INSIG2) would be strong candidate genes for interindividual variation in drug-induced weight gain. We genotyped a total of 44 HapMap-selected tagging single nucleotide polymorphisms in a sample of 160 German patients with schizophrenia that had been monitored with respect to changes in body mass index during antipsychotic drug treatment. We found a strong association (P=0.0003-0.00007) between three markers localized within or near the INSIG2 gene (rs17587100, rs10490624 and rs17047764) and antipsychotic-related weight gain. Our finding is supported by the recent involvement of the INSIG2 gene in obesity in the general population and implicates SREBP-controlled lipogenesis in drug-induced metabolic adverse effects.

Body weight gain induced by atypical antipsychotics: an extension of the monozygotic twin and sib pair study.

BACKGROUND AND OBJECTIVE: In our original study based on five monozygotic twin pairs and seven same-sex sib pairs, we previously showed that genetic factors contribute to body weight gain induced by the atypical antipsychotic clozapine. We aim to study this further by including patients treated with the atypical antipsychotics olanzapine or risperidone as well as opposite-sex sib pairs. METHODS: Twin and sib pairs were identified by a telephone screening. Measured data on weight and other clinical variables were obtained cross-sectionally and retrospectively from medical records. In seven monozygotic twin pairs and 12 sib pairs (total number of patients treated: n = 38, mean age 29.5 +/- 9.5, range 13.7-54.3 years), the similarity in BMI (kg/m(2)) change under these atypical antipsychotics (atypical Delta BMI) and upon additional inclusion of BMI change under prior antipsychotic medication (total Delta BMI) was explored. RESULTS: For total Delta BMI we found greater similarity in antipsychotic-induced BMI change in MZ twin pairs than in sib pairs (intrapair difference) with a heritability of h(2) = 0.6, but not for atypical Delta BMI, possibly because of a genetically influenced weight plateau achieved under antipsychotic medication. CONCLUSION: The results of the present and our previous report suggest a contribution of genetic factors in antipsychotic-induced weight gain of 60-80%.

[The assessment of therapy outcome in child and adolescent psychiatry under naturalistic conditions--conception and implementation of the Marburg System of Quality Assurance and Therapy Evaluation]. / Die Erfassung des Therapieerfolges in der Kinder- und Jugendpsychiatrie unter naturalistischen Bedingungen.

OBJECTIVES: The objective of and concept behind the Marburg System of Quality Assurance and Therapy Evaluation (MARSYS) are presented. This a modular system with several instruments that is applicable for the evaluation of therapies with psychiatrically ill children and adolescents and can be employed in everyday routine care. METHODS: The system was tested between 1999 and 2006 in the Marburg project on therapy evaluation in inpatient care. The article presents data from a complete user population with 1,321 patients. Data reporting on the practicability of the system comprise completeness of samples, dropout rates, acceptance of the evaluation among patients and parents, and the input of time and costs. RESULTS: The results show that the system is very well suited to obtain significant findings with regard to the quality of outcome of the treatments. CONCLUSIONS: The possibilities of implementing this system are discussed with regard to the matter of costs.

Schizophrenia and related disorders in children and adolescents.

This paper reviews the concept and recent studies on childhood and adolescent psychoses with special reference to schizophrenia. After a short historical introduction, the definition, classification, and epidemiology of child- and adolescent-onset psychoses are described, pointing out that some early-onset psychotic states seem to be related to schizophrenia (such as infantile catatonia) and others not (such as desintegrative disorder). The frequency of childhood schizophrenia is less than 1 in 10,000 children, but there is a remarkable increase in frequency between 13 and 18 years of age. Currently, schizophrenia is diagnosed according to ICD-10 and DSM-IV criteria. The differential diagnosis includes autism, desintegrative disorder, multiplex complex developmental disorder (MCDD) respectively multiple developmental impairment (MDI), affective psychoses, Asperger syndrome, drug-induced psychosis and psychotic states caused by organic disorders. With regard to etiology, there is strong evidence for the importance of genetic factors and for neurointegrative deficits preceding the onset of the disorder. Treatment is based upon a multimodal approach including antipsychotic medication (mainly by atypical neuroleptics), psychotherapeutic measures, family-oriented measures, and specific measures of rehabilitation applied in about 30% of the patients after completion of inpatient treatment. The long-term course of childhood- and adolescent-onset schizophrenia is worse than in adulthood schizophrenia, and the patients with manifestation of the disorder below the age of 14 have a very poor prognosis.

Molecular genetic aspects of attention-deficit/hyperactivity disorder.

Two genome wide scans, one of which was subsequently extended, have led to the identification of different chromosomal regions assumed to harbour genes underlying attention-deficit/hyperactivity disorder (ADHD). Some of these regions were also identified in patients with autism and/or dyslexia. The only region for which both studies detected a LOD score >1 was on chr 5p13 which is in the vicinity of the location of the candidate gene DAT1. The candidate gene approach has revealed the most robust and replicated findings for DRD4, DRD5, and DAT1 polymorphisms. Meanwhile interesting endophenotype studies have also been conducted suggesting a genetic basis for different diagnostic and therapeutic criteria. Animal studies for ADHD have investigated especially hyperactivity and have focused mainly on knockout and QTL designs. In knockout mice models the most promising results were obtained for genes of the dopaminergic pathway. QTL results in rodents suggest multiple loci underlying different forms of natural and induced hyperactivity. The molecular results mentioned above are presented and discussed in detail, thus providing both clinicians and geneticists with an overview of the current research status of this important child and adolescent psychiatric disorder.

Serum leptin, gonadotropin, and testosterone concentrations in male patients with anorexia nervosa during weight gain.

Amenorrhea in female patients with anorexia nervosa is associated with low leptin secretion, thus suggesting a causal link. In an attempt to address the hypothesis that leptin also influences the hypothalamo-pituitary-gonadal function in males, we studied three male patients with acute anorexia nervosa longitudinally. Serum levels of leptin, LH, FSH, testosterone, and SHBG were measured on a biweekly basis during weight gain. Leptin levels at low body mass index values were below the 5th percentile. During weight gain, leptin levels reached or surpassed the 95th percentile. The temporal dynamics of body mass index and fat mass were closely related to those of leptin concentrations in serum. Leptin increments were paralleled by increments of gonadotropins, testosterone, and the free androgen index (FAI). In each of the patients, serum concentrations of leptin were positively correlated with those of testosterone (P = 0.0001, P = 0.01, P = 0.07, respectively) and FAI (P = 0.0001, P = 0.0001, P = 0.09, respectively). In addition, in the combined data set of all patients changes of leptin over time were positively correlated with changes in LH (P = 0.01), FSH (P = 0.0001), testosterone (P = 0.002), and FAI (P = 0.002). In conclusion, these data suggest that leptin might also play an important role in the regulation of the hypothalamo-pituitary-gonadal axis and fertility in underweight males as has previously been shown in underweight females.

Independent confirmation of a major locus for obesity on chromosome 10.

Linkage results obtained in genome-wide scans for complex phenotypes require confirmation in independent samples. Recently, linkage of obesity to chromosome 10p12 with a maximal multipoint LOD score of 4.85 was reported upon use of an affected sib-pair approach including nuclear families in which the adult index case had a BMI > or = 40 kg/m2 and at least one further sibling had a BMI > or = 27 kg/m2 (Hager et al., 1998, Nat Genet 20:304-8). To attempt to replicate this linkage finding we genotyped 11 markers spanning approximately 23 cM from 10p13 to 10ql1 in a total of 386 individuals stemming from 93 nuclear families with two or more young obese offspring with a BMI > or = 90th age percentile. The highest multipoint maximum likelihood binomial (MLB) LOD score using the extreme concordant sib-pair approach in which one sib had a BMI > or = 95th percentile, and other sibs a BMI > or = 90th percentile was 2.32. Six markers yielded nominal p-values < 0.05, the highest two point MLB-LOD score of 2.45 (nominal p = 0.0004) was obtained for the marker TCF8. Transmission disequilibrium tests for the most frequent parental allele yielded no nominal p-value < 0.05. The linkage results confirm the presence of a major susceptibility locus for obesity in a region near the centromere on chromosome 10.

Systematic mutation screening of the estrogen receptor beta gene in probands of different weight extremes: identification of several genetic variants.

Estrogens are known to have an inhibitory effect on food intake in rodents and primates. Decreased estrogen levels that are found for instance in menopausal woman and in ovarectomized rodents result in body weight gain. Estrogen can act both in the periphery and in the central nervous system via at least two different estrogen receptors (alpha and beta). We systematically screened the coding region and part of the 5' and 3'regions of the estrogen receptor beta gene (ER beta) in 96 extremely obese children and adolescents, 50 patients with anorexia nervosa (AN), 28 patients with bulimia nervosa (BN), and 25 healthy underweight individuals. We detected five different sequence variants in the ER beta: a) A 21 bp deletion (codons 238 to 244) was detected in two obese probands and an underweight individual. b) An 846G-->A transition leading to a nonconservative amino acid substitution (G-250-S) was found in two obese male probands. Both a) and b) were located within the flexible hinge region between DNA and ligand binding domain. c) For a 1082G-->A polymorphism we found suggestive evidence for an association between the more common 1082G-allele and anorexia nervosa (nominal p=0.04). d) One silent mutation (1421T-->C) was found solely in two obese probands. e) A common variant is located in the 3' nontranslated region at position 1730(A-->G). We did not detect association of this polymorphism to any of the analyzed phenotypes. We conclude that the ER beta harbors several different mutations and polymorphisms, none of which can readily be associated with the phenotypes under study.

Systematic mutation screening of the pro-opiomelanocortin gene: identification of several genetic variants including three different insertions, one nonsense and two missense point mutations in probands of different weight extremes.

Pro-opiomelanocortin (POMC) is the precursor of melanocortins (adrenocorticotropin: ACTH, beta-endorphin, beta-lipotropin: beta-LPH, corticotropin like intermediate peptide, alpha-, beta- and gamma-melanocyte-stimulating hormone: alpha-, beta- and gamma-MSH) some of which act in the brain to reduce food intake and are potential mediators of leptin action. Recently, three different mutations in the POMC gene (POMC) were identified in two unrelated children that lead to early-onset extreme obesity, adrenal insufficiency, and red hair pigmentation. In the present study we systematically screened the coding region of POMC in 96 extremely obese children and adolescents, 60 healthy underweight individuals and 46 patients with anorexia nervosa (AN) and identified several variants. a) A 9 and an 18 base pair insertion (9bp and 18bp: AGC AGC GGC and AGC AGC GGC AGC AGC GGC, respectively, between codon 73 and 74; 1,2). These in-frame variants lead to the insertion of three or six amino acids (Ser-Ser-Gly; Ser-Ser-Gly-Ser-Ser-Gly) carboxy-terminal to gamma-MSH. Frequencies of the 9bp insertion allele varied between 3 and 5% among the different study groups (Pearson's chi2 P>0.5). b) Both an out-of-frame 6 bp insertion (within codon 176: GGG CCC) leading to the insertion of two amino acids (Arg-Ala) and a premature stop-codon (G-7316-T: Glu-180-Stop) within the gamma-LPH sequence were maternally inherited in an obese female proband. This proband inherited another missense mutation from her father (Glu-188-Gly). c) A missense mutation (G-7016-A; Asp-80-Asn) was observed in a single patient with AN who also harboured the 9bp insertion on a paternally derived haplotype. d) The allelic co-occurence of two silent mutations (C-6982-T and C-7285-T) was detected in two obese subjects. e) Two further silent mutations (C-3832-T; C-7111-G) were detected in an underweight and an obese subject, respectively. We conclude that the POMC gene harbors several different polymorphisms and mutations, none of which can readily be associated with the phenotypes under study.

Several mutations in the melanocortin-4 receptor gene including a nonsense and a frameshift mutation associated with dominantly inherited obesity in humans.

The melanocortin-4 receptor gene (MC4-R) has been implicated in weight regulation. Recently, two independent groups reported frameshift mutations associated with a dominant form of obesity (1, 2). We screened the coding region of the MC4-R in 306 extremely obese children and adolescents (mean body mass index: BMI 34.4 +/- 6.6 kg/m2), 25 healthy underweight students (mean BMI 17.1 +/- 0.8 kg/m2), 52 normal weight individuals (mean BMI 22.0 +/- 1.0 kg/m2), 51 inpatients with anorexia nervosa (AN, DSM IV criteria, mean BMI 14.3 +/- 1.5 kg/m2) and 27 patients with bulimia nervosa (BN, DSM IV criteria, mean BMI 21.7 +/- 5.8 kg/m2) by single strand conformation polymorphism analysis (SSCP). Several mutations were identified, including the frameshift mutation described (1). The mutations were as follows: a) The deletion of 4 bp (delta of CTCT at codon 211) results in a frameshift, thus rendering a truncated protein. This mutation has been assumed to be associated with dominantly-inherited morbid obesity in humans (1). Both the index patient (BMI 42.06 kg/m2, height 171 cm, age 19.6 years) and her mother (BMI 37.55 kg/m2, height 164 cm, age 42.5 years) were heterozygous for the deletion. b) A nonsense mutation at position 35 of the MC4-R was detected in two obese probands (BMI 31.29 kg/m2 and BMI 45.91 kg/m2). This mutation leads to a truncated protein that encompasses the N-terminal extracellular domain. Both carriers additionally showed (c) a missense mutation (Asp-37-Val). In both of these cases Tyr-35-Stop and Asp-37-Val were maternally transmitted, thus these variations form a haplotype. d) e) A male obese proband harbored two missense mutations (Ser-30-Phe, Gly-252-Ser). f)-i) Four different missense mutations (Pro-78-Leu, Thr-112-Met, Arg-165-Trp, Ile-317-Thr) were detected in four different male probands, respectively. All of these mutations (a to i) were found solely in extremely obese individuals whose BMIs were all above the 99th percentile. j) A silent mutation (C-579-T, Val-193-Val) was detected in a male underweight individual. k) A previously described polymorphism (Val-103-Ile; 3) was detected with similar frequencies in all different study groups. 1) We identified a novel polymorphism (Ile-251-Leu) with similar allele frequencies in all groups under study. In conclusion, our data indicate that mutations in the MC4-R are not uncommon. Whereas our data support the evidence for dominantly inherited obesity as revealed by the three obese probands with haplo-insufficiency, the functional significance of the missense mutations remains to be determined.

Prediction of low body weight at long-term follow-up in acute anorexia nervosa by low body weight at referral.

OBJECTIVE: The authors investigated the hypothesis that in acute anorexia nervosa a low body weight predicts a poor weight prognosis for the future. METHOD: The body mass indexes at referral of 272 female patients were examined in relation to the body mass indexes of these patients after a mean follow-up of 9.5 years. RESULTS: The overall correlation between body mass indexes at referral and at follow-up was r = 0.33. Despite this low correlation, the 100 patients with body mass indexes less than 13 kg/m2 at referral had low weights at long-term follow-up. Eleven of the 12 deceased patients were among these 100 patients, as were 24 of the 46 surviving patients whose body mass indexes were 17.5 kg/m2 or less at follow-up. CONCLUSIONS: For patients with anorexia nervosa, a body mass index less than 13 kg/m2 at referral indicates a substantial risk for chronic anorexia nervosa and death related to emaciation.

Partial trisomy 16p in an adolescent with autistic disorder and Tourette's syndrome.

A partial trisomy 16p was identified in a 14-year-old male adolescent with autistic disorder. He additionally showed complex motor and vocal phenomena, including some simple tics which had first appeared in childhood. Whereas these simple tics were of subclinical significance, an additional diagnosis of Tourette's syndrome (TS) appears justified. The case report illustrates the diagnostic difficulties in assessing psychiatric symptomatology associated with both disorders, especially complex motor and vocal phenomena. The cytogenetic finding is discussed critically in the light of other chromosome abnormalities reported in both TS and autistic disorder. Chromosome 16p should be considered as a candidate region especially for autistic disorder.

Tourette syndrome and the norepinephrine transporter gene: results of a systematic mutation screening.

Tourette syndrome (TS) is a complex inherited neuropsychiatric disorder characterized by multiple motor and phonic tics. Involvement of central norepinephrine mechanisms is suggested by central norepinephrinic hyperactivity in patients with TS and by the therapeutic effects of the presynaptic alpha2-adrenergic agonist clonidine. The norepinephrine transporter gene (NET) was systematically screened by single-strand conformation analysis for genetic variants, including the whole coding region and adjacent exon-intron boundaries in 43 patients with TS and 46 healthy controls. We detected 12 DNA sequence variants, among them four missense mutations (Val69Ile, Thr99Ile, Va1245Ile, and Gly478Ser). The observed missense mutations may alter conformational rearrangements during gating of the transporter, assembly of subunits, and norepinephrine-specific uptake affinity. Allele frequency and genotype distribution of the genetic variants showed no differences between TS patients and controls. No mutation of likely functional significance was found that distinguished TS patients from healthy controls, indicating that genetic variants of the NET gene are not causally related to Tourette syndrome.

No evidence for involvement of polymorphisms of the dopamine D4 receptor gene in anorexia nervosa, underweight, and obesity.

Family and twin studies suggest a genetic contribution to the etiology of anorexia nervosa (AN) and obesity. Genes involved in weight regulation can be considered as candidate genes for AN. The dopaminergic system has been implicated in weight regulation; previous results had suggested a possible involvement of the dopamine D4 receptor gene (DRD4). We screened for alleles of two different polymorphisms (13-bp deletion, 48-bp repeat) in the DRD4. For association tests, allele frequencies were compared between 109 inpatients with AN, 82 underweight students, and 327 extremely obese children and adolescents. For application of transmission disequlibrium tests (TDT) we additionally genotyped 57 and 137 trios comprising a patient with AN or an extremely obese child or adolescent, respectively, and both parents. All genotyping was performed with polymerase chain reaction fragment length polymorphism analyses. None of the association tests or TDT rendered nominal P values below 0.1. An influence of alleles of the DRD4 on the development of AN, underweight, or extreme early onset obesity was not detected. Am. J. Med. Genet. (Neuropsychiatr. Genet.) 88:594-597, 1999.

Auditory processing and dyslexia: evidence for a specific speech processing deficit.

In order to investigate the relationship between dyslexia and central auditory processing, 19 children with spelling disability and 15 controls at grades 5 and 6 were examined using a passive oddball paradigm. Mismatch negativity (MMN) was determined for tone and speech stimuli. While there were no group differences for the tone stimuli, we found a significantly attenuated MMN in the dyslexic group for the speech stimuli. This finding leads to the conclusion that dyslexics have a specific speech processing deficit at the sensory level which could be used to identify children at risk at an early age.

Attenuated hemispheric lateralization in dyslexia: evidence of a visual processing deficit.

There is controversial evidence that deficits in the processing of low contrast and low spatial frequency stimuli are of importance in the pathogenesis of dyslexia. Fifteen adult dyslexics and 19 controls were examined using visual evoked potentials (VEP) at varying spatial frequencies (2 and 11.33 cpd) and contrasts (0.2, 0.4, 0.6, 0.8). Our results show that the amplitude of VEPs following different spatial frequencies and contrasts did not differentiate between dyslexics and controls. Further, we found significantly higher amplitudes of the P1 and P2 over the right occipital cortex. For the P2, this hemispheric asymmetry was not found in the dyslexic group suggesting a specific low level visual processing deficit in the right occipital region in dyslexia.

Biogenic amines as predictors of response to clozapine treatment in early-onset schizophrenia.

The purpose of the present study was to assess the relationship between plasma monoamine levels and changes in psychopathology in a sample of children and adolescents with schizophrenia (DSM-III-R criteria) during conventional neuroleptic therapy and during short-term treatment with clozapine. After failing on conventional neuroleptics in open-labeled clinical trials lasting a mean of 1.6 years, 15 inpatients (aged 11-20 years) received clozapine. Weekly ratings of psychopathological symptoms using standard rating scales were performed in parallel to blood samplings for measurements of biogenic amines and serum levels of clozapine. These measures were obtained for 6 weeks during conventional neuroleptic treatment and for 6 weeks during the open-label clozapine trial. Serum levels of serotonin and plasma norepinephrine levels were significantly higher during treatment with clozapine than during pretreatment with typical neuroleptics. A comparison of plasma epinephrine levels in responders (n = 7) and non-responders (n = 8) to clozapine revealed that response to clozapine can be predicted by epinephrine levels prior to initiation of treatment with clozapine (responders ranging from 32.2-90.3 pg/ml; non-responders ranging from 92.5-473.5 pg/ml). Additionally, subjects who responded to clozapine showed increased mean plasma concentrations of methoxyhydroxyphenylglycol (MHPG) and epinephrine during treatment with this drug in comparison to the levels measured during pretreatment with typical neuroleptic medication. In conclusion, our results demonstrate that plasma epinephrine levels prior to initiation of clozapine therapy predict response to this atypical neuroleptic.