Decreased serum levels of transforming growth factor-beta1 in patients with autism

BACKGROUND: The neurobiological basis for autism remains poorly understood. Given the key role of transforming growth factor-beta1 (TGF-beta1) in brain development, we hypothesized that TGF-beta1 plays a role in the pathophysiology of autism. In this study, we studied whether serum levels of TGF-beta1 are altered in patients with autism. METHODS: We measured serum levels of TGF-beta1 in 19 male adult patients with autism and 21 age-matched male healthy subjects using enzyme-linked immunosorbent assay (ELISA). RESULTS: The serum levels (7.34+/-5.21 ng/mL (mean+/-S.D.)) of TGF-beta1 in the patients with autism were significantly (z=-5.106, p<0.001) lower than those (14.48+/-1.64 ng/mL (mean+/-S.D.)) of normal controls. However, there were no marked or significant correlations between serum TGF-beta1 levels and other clinical variables, including Autism Diagnostic Interview-Revised (ADI-R) scores, Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), aggression, Theory of Mind, and Intellectual Quotient (IQ) in patients. CONCLUSIONS: These findings suggest that decreased levels of TGF-beta1 may be implicated in the pathophysiology of autism.     

Decreased serum levels of hepatocyte growth factor in male adults with high-functioning autism

BACKGROUND: The mechanisms underlying the pathophysiology of autism are currently unknown. Given the role of hepatocyte growth factor (HGF) in brain development, we hypothesized that HGF plays a role in the pathophysiology of autism. In this study, we studied whether serum HGF levels are altered in subjects with high-functioning autism. METHODS: Using an enzyme-linked immunosorbent assay (ELISA), we measured serum levels of HGF in 17 male adults with high-functioning autism and age-matched 18 male healthy subjects. RESULTS: The serum levels (503.5+/-160.5 pg/mL (mean+/-SD)) of HGF in the subjects with high-functioning autism were significantly (Mann-Whitney U=34.0, p<0.001) lower than those (817.6+/-232.4 pg/mL (mean+/-SD)) of control subjects. However, there were no correlations between serum HGF levels and clinical variables in the patients. CONCLUSIONS: This study suggests that reduced HGF levels may play a role in the pathophysiology of high-functioning autism.     

Decreased serum levels of epidermal growth factor in adult subjects with high-functioning autism.

BACKGROUND: The neurobiological basis for autism remains poorly understood. Given the role of growth factors in brain development, we hypothesized that epidermal growth factor (EGF) may play a role in the pathophysiology of autism. In this study, we examined whether serum levels of EGF are altered in adult subjects with high-functioning autism. METHODS: We measured serum levels of EGF in the 17 male subjects with high-functioning autism and 18 age-matched healthy male subjects. RESULTS: The serum levels of EGF in the subjects with high-functioning autism (72.4 +/- 102.8 pg/mL [mean +/- SD]) were significantly lower (Mann-Whitney U = 22.0, p < .001) than those of normal control subjects (322.3 +/- 122.0 pg/mL [mean +/- SD]). However, there were no correlations between serum EGF levels and clinical variables in the subjects with autism. CONCLUSIONS: This study suggests that decreased levels of EGF might be implicated in the pathophysiology of high-functioning autism.     

Increased serum levels of glutamate in adult patients with autism

BACKGROUND: Precise mechanisms underlying the pathophysiology of autism are currently unknown. Given the major role of glutamate in brain development, we have hypothesized that glutamatergic neurotransmission plays a role in the pathophysiology of autism. In this study, we studied whether amino acids (glutamate, glutamine, glycine, D-serine, and L-serine) related to glutamatergic neurotransmission are altered in serum of adult patients with autism. METHODS: We measured serum levels of amino acids in 18 male adult patients with autism and age-matched 19 male healthy subjects using high-performance liquid chromatography. RESULTS: Serum levels (mean = 89.2 microM, S.D. = 21.5) of glutamate in the patients with autism were significantly (t = -4.48, df = 35, p < 0.001) higher than those (mean = 61.1 microM, S.D. = 16.5) of normal controls. In contrast, serum levels of other amino acids (glutamine, glycine, d-serine, l-serine) in the patients with autism did not differ from those of normal controls. There was a positive correlation (r = 0.523, p = 0.026) between serum glutamate levels and Autism Diagnostic Interview-Revised (ADI-R) social scores in patients. CONCLUSIONS: The present study suggests that an abnormality in glutamatergic neurotransmission may play a role in the pathophysiology of autism.     

Investigation of serum BDNF levels in drug-naive patients with schizophrenia

The role of brain-derived neurotrophic factor (BDNF) is to promote and modulate the neuronal responses across neurotransmitter systems in the brain. Therefore, abnormal BDNF signaling may be associated with the pathophysiology of schizophrenia. Decreased BDNF levels in the brain and the serum of patients with psychotic disorders have been reported. In the present study, we assessed serum BDNF levels in a group of 14 drug-naive first-episode patients with schizophrenia (FEP), compared to 15 healthy controls. The serum BDNF levels in the sample of FEP patients was significantly reduced compared to normal controls (23.92+/-5.99 ng/ml vs. 30.0+/-8.43 ng/ml, F=5.01, df=1, p=.034). Negative correlations were shown between serum BDNF levels of the patients and the PANSS Positive and Negative subscale scores. Our findings indicate that BDNF levels at the onset of schizophrenia may reflect associated pathophysiological processes as well as the severity of positive and negative psychotic symptoms.     

No change between the serum brain-derived neurotrophic factor in female patients with anorexia nervosa before and after partial weight recovery

BACKGROUND: Brain-derived neurotrophic factor (BDNF) is a member of the nerve growth factor family. Several lines of evidence indicate that BDNF plays a role in the pathophysiology of eating disorders (ED). We found that serum BDNF levels in patients with ED were significantly lower than in normal controls. The aim of this longitudinal study was to compare serum BDNF levels in patients with anorexia nervosa (AN) before (n=13, initial mean body mass index (BMI)=14.2 kg/m(2)+/-0.7) and after partial recovery (mean BMI=16.2 kg/m(2)+/-1.7, mean weight gain 5.0 kg+/-2.0), with age-matched normal control subjects (n=17, mean BMI=20.4 kg/m(2)+/-1.5). METHODS: Eating related psychopathology and depressive symptoms were evaluated before and after partial weight recovery, using the Eating Disorder Inventory-2 (EDI-2) and the 17-item Hamilton Depression Rating Scale (HDRS). Serum BDNF levels were measured using a sandwich enzyme-linked immunosorbent assay. RESULTS: Serum BDNF levels with the AN patients (14.5+/-4.4 ng/ml) were significantly (p<0.01) lower than in control subjects (22.1+/-8.9 ng/ml). There were no significant differences in serum BDNF levels between the patients with AN before (14.5+/-4.4 ng/ml) and after (17.2+/-6.9 ng/ml) partial weight recovery. In all subjects, there was a positive correlation (r=0.5, p<0.01) between the baseline BDNF levels and the EDI-2 scores (n=30). Furthermore, in all subjects there was a positive correlation (r=0.4, p<0.05) between the BDNF levels and the BMI. CONCLUSIONS: Serum BDNF levels did not change after partial weight recovery in AN patients. Our results suggest that an alteration of the serum BDNF in AN patients is not due to changes in body weight. Thus, other possible mechanisms that could be related to low serum BDNF levels in AN patients should be evaluated.     

The effect of chronic antidepressant treatment on serum brain-derived neurotrophic factor levels in depressed patients: a preliminary study

Recent studies suggested a role of brain-derived neurotrophic factor (BDNF) in depression. While BDNF levels are lower in depressed patients, antidepressant treatment increases serum BDNF levels of depressed patients. Our study aims to test the effect of chronic venlafaxine treatment on serum BDNF levels in patients with a major depressive disorder. Ten patients diagnosed as major depressive disorder according to DSM-IV are included in the study. Two of the patients had their first episode and were drug-naive, the other eight patients were drug-free for at least 4 weeks. The severity of depression was assessed with Hamilton Depression Rating Scale (HDRS). The control group consisted of ten age- and sex-matched subjects without any psychiatric disorder. Blood samples were collected at the baseline and after 12 weeks of antidepressant treatment (during remission). At the baseline the mean serum BDNF level was 17.9+/-9.1 ng/ml and the mean HDRS score was 23.2+/-4.6. Serum BDNF levels of the study group were significantly lower than in the control group (31.6+/-8.6 ng/ml). At the end of the study, the mean serum BDNF level was 34.6+/-7.1 ng/ml whereas the mean HDRS score was 8.2+/-3.9. From the baseline to the remission after 12 weeks of treatment, the increase in serum BDNF level and the decrease in HDRS score were statistically significant, respectively. When we compared the serum BDNF level of depressed patients at remission to that of the controls, there was no statistically significant difference. This study shows that venlafaxine treatment of depression improves serum BDNF level which may be considered as a nonspecific peripheral marker of depression.     

Serum brain-derived neurotrophic factor levels in conversion disorder: Comparative study with depression

The aim of the present study was to compare serum brain-derived neurotrophic factor (BDNF) levels of patients with major depressive disorder (MDD) and conversion disorder (CD). Serum BDNF levels were measured in the following three groups: 15 CD patients without any comorbid diagnosis of psychiatric disorder, 24 patients with MDD, and 26 healthy subjects without any psychiatric diagnosis or psychiatric treatment. The serum BDNF level of the healthy control group (31.4 +/- 8.8 ng/mL) was statistically higher than the level of the MDD group (21.2 +/- 11.3 ng/mL) and the CD group (24.3 +/- 9.0 ng/mL; P = 0.008). This suggests that BDNF level may play a similar role in the pathophysiology of MDD and CD.     

Effect of treatment on serum brain–derived neurotrophic factor levels in depressed patients

Researchers have reported that serum brain–derived neurotrophic factor (sBDNF) of drug–free depressed patients are lower than those of healthy controls and proposed that low sBDNF levels might reflect failure of neuronal plasticity in depression. In this study, we compared sBDNF levels of depressed patients (n = 28) before and after 8 weeks of antidepressant treatment, with those of healthy controls (n = 18) to test the hypothesis that initially low sBDNF levels of drug–free depressed patients will increase parallel with their clinical response to antidepressant treatment. The severity of depression and response to treatment were assessed with Hamilton Rating Scale for Depression (HAM–D). sBDNF was assayed with the sandwich ELISA method. Baseline sBDNF levels of patients (mean, 20.8 ng/ml; [S.D., 6.7]) were significantly lower than those of controls (mean, 26.8 ng/ml; [S.D., 9.3]; p = 0.015), and were negatively correlated with HAM–D scores (r = –0.49, p = 0.007). After 8 weeks of treatment, sBDNF levels of patients had increased significantly (mean, 33.3 ng/ml; [S.D., 9.9]; p < 0.001) and no longer differed from those of controls. These results support the hypothesis that BDNF might play a critical role in the pathophysiology of major depressive disorder and successful antidepressant treatment increases the attenuated BDNF levels in depressed patients.     

Alterations of serum levels of brain-derived neurotrophic factor (BDNF) in depressed patients with or without antidepressants.

BACKGROUND: Because researchers have reported that antidepressants increase the expression of brain-derived neurotrophic factor (BDNF) in the rat hippocampus, we investigated whether serum BDNF levels may be used as a putative biological marker for major depressive disorders (MDD). METHODS: We measured serum BDNF in the following three groups: antidepressant-naive patients with MDD (n = 16), antidepressant-treated patients with MDD (n = 17), and normal control subjects (n = 50). Patients were evaluated using the Hamilton Rating Scale for Depression (HAM-D). Serum BDNF was assayed with the sandwich ELISA method. RESULTS: We found that serum BDNF was significantly lower in the antidepressant-naive group (mean, 17.6 ng/mL; SD, 9.6) than in the treated (mean, 30.6 ng/mL; SD, 12.3; p =.001) or in the control group (mean, 27.7 ng/mL; SD, 11.4; p =.002). There was a significant negative correlation (r = -.350, z = -2.003, p =.045) between serum BDNF and HAM-D scores in all patients. In a preliminary examination, reduced BDNF values of three drug-naive patients recovered to basal levels after antidepressant treatment. CONCLUSIONS: Our study suggests that low BDNF levels may play a pivotal role in the pathophysiology of MDD and that antidepressants may increase BDNF in depressed patients.     

Serum brain-derived neurotrophic factor (BDNF) levels in patients with panic disorder: as a biological predictor of response to group cognitive behavioral therapy

Little is known about biological predictors of treatment response in panic disorder. Our previous studies show that the brain-derived neurotrophic factor (BDNF) may play a role in the pathophysiology of major depressive disorders and eating disorders. Assuming that BDNF may be implicated in the putative common etiologies of depression and anxiety, the authors examined serum BDNF levels of the patients with panic disorder, and its correlation with therapeutic response to group cognitive behavioral therapy (CBT). Group CBT (10 consecutive 1 h weekly sessions) was administered to the patients with panic disorder after consulting the panic outpatient special service. Before treatment, serum concentrations of BDNF and total cholesterol were measured. After treatment, we defined response to therapy as a 40% reduction from baseline on Panic Disorder Severity Scale (PDSS) score as described by [Barlow, D.H., Gorman, J.M., Shear, M.K., Woods, S.W., 2000. Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: A randomized controlled trial. JAMA. 283, 2529-2536]. There were 26 good responders and 16 poor responders. 31 age- and sex-matched healthy normal control subjects were also recruited in this study. The serum BDNF levels of the patients with poor response (25.9 ng/ml [S.D. 8.7]) were significantly lower than those of the patients with good response (33.7 ng/ml [S.D. 7.5]). However, there were no significant differences in both groups of the patients, compared to the normal controls (29.1 ng/ml [S.D. 7.1]). No significant differences of other variables including total cholesterol levels before treatment were detected between good responders and poor responders. These results suggested that BDNF might contribute to therapeutic response of panic disorder. A potential link between an increased risk of secondary depression and BDNF remains to be investigated in the future.     

Increased serum S100B protein in schizophrenia: a study in medication-free patients

S100B protein, a calcium binding protein produced and released by glial cells, has been used as a sensitive marker of brain damage. Previous studies have found alterations in peripheral S100B levels in schizophrenic patients on medication. We compared serum S100B levels of 20 medication-free DSM-IV schizophrenic patients and 20 age-gender matched healthy controls. Schizophrenic patients presented higher serum S100B levels (mean 0.120 ng/ml+/-S.D. 0.140) compared to controls (mean 0.066 ng/ml+/-S.D. 0.067; P=0.014) and there was a negative correlation with illness duration (r=-0.496, P=0.031). The results of this study indicate that serum S100B levels may be a state marker of a limited neurodegenerative process, particularly in the early course of schizophrenia or, at least, in a subgroup of schizophrenic patients.     

Serum BDNF levels in suicide attempters related to psychosocial stressors: a comparative study with depression

Although many studies have examined the neurobiological aspects of suicide, the molecular mechanisms and pathophysiologic mechanisms associated with suicide remain unclear. In this study, it is aimed to investigate whether there is a difference in serum brain-derived neurotrophic factor (BDNF) levels among suicide attempters without a major psychiatric disorder, compared to major depressive disorder patients and healthy subjects. It was undertaken with the hypothesis that suicide per se lowers serum BDNF levels, since it is a source of stress. The study was carried out in Celal Bayar University Hospital, Manisa, Turkey. Ten suicide attempters, 24 patients with major depressive disorder and 26 subjects without any psychiatric diagnosis and any psychiatric treatment were included in the study. All subjects were asked to give their written consent. Blood samples were collected at the baseline. Serum BDNF was kept at -70 degrees C before testing, and assayed with an ELISA kit (Promega; Madison, Wisc., USA) after dilution with the block and sample solution provided with the kit. The data were subjected to the Kruskal-Wallis test for nonparametric analysis of variance. Mean serum BDNF levels were significantly lower in the suicide group (21.2 +/- 12.4 ng/ml) and the major depressive disorder group (21.2 +/- 11.3 ng/ml) than the control group (31.4 +/- 8.8 ng/ml; p = 0.004). These results suggest that BDNF may play an important role in the neurobiology of suicidal behavior. BDNF levels may be a biological marker for suicidal behavior. To investigate the role of BDNF in suicide, further studies with a wider sample size and a variety of psychiatric diagnoses accompanying suicide attempt are needed.     

Reduced serum BDNF levels in schizophrenic patients on clozapine or typical antipsychotics

Neurotrophic factors regulate neuronal development and synaptic plasticity, possibly playing a role in the pathophysiology of schizophrenia and other psychiatric disorders. Decreased brain-derived neurotrophic factor (BDNF) levels have been found in brains and in the serum of schizophrenic patients, but results are inconsistent. Also, clozapine may upregulate brain BDNF expression. In the present study, we assessed serum BDNF immunoreactivity in 44 schizophrenic patients (20 on clozapine and 24 on typical antipsychotics) and in 25 healthy volunteers. Serum BDNF levels were measured using an enzyme immunoassay. Healthy controls showed significantly higher levels of BDNF compared to the whole group of schizophrenic patients (p<0.001) as well as to the subgroups on typical antipsychotics and clozapine (p<0.001). Serum BDNF values for controls were 168.8+/-26.3pg/ml, for the clozapine group were 125.4+/-44.5pg/ml and for the group on typicals were 101.3+/-51.6pg/ml. BDNF values from patients on clozapine were non-significantly higher than values from patients on typical antipsychotics (p=0.09). Serum BDNF was strongly and positively correlated with clozapine dose (r=0.643; p=0.002) but not with other demographic characteristics. These results reinforce previous findings of reduced serum BDNF levels in schizophrenic patients and suggest a differential effect of clozapine compared to typical antipsychotics on such levels.     

Serum ferritin in children with attention-deficit hyperactivity disorder

To further define a role for iron deficiency in attention-deficit hyperactivity disorder, serum ferritin was included in a battery of laboratory tests obtained in children attending a clinic for attention deficit disorders. In a total of 68 consecutive patients (ages 5-16 years; 54 male, 14 female), the range of serum ferritin levels was a low of 7.7 ng/mL and a high of 150 ng/mL. The mean (+/-S.D.) serum ferritin level in patients with attention-deficit hyperactivity disorder (39.9 +/- 40.6 ng/mL) was not different from that of control subjects. Seventy-four percent of patients had serum ferritin levels below 50 ng/mL; 44% were below 30 ng/mL; and 18% below 20 ng/mL. None had evidence of iron deficiency anemia. A comparison of the clinical characteristics of the 12 patients with the lowest serum ferritin levels (<20 ng/mL) and 12 with the highest levels (>60 ng/mL) disclosed no significant difference in severity or frequency of attention-deficit hyperactivity disorder and comorbid symptoms or response to medications. In this patient cohort, a causative role for low serum ferritin in attention-deficit hyperactivity disorder has not been confirmed.     

No pituitary gland volume change in medication-free depressed patients

Increased serum cortisol levels and a hyperactive hypothalamo-pituitary-adrenal (HPA) axis have been proposed to play an important role in the pathophysiology of Major Depressive Disorder (MDD). However, there are inconsistent results regarding pituitary gland volume (PGV), which is one of the key elements of the HPA axis evaluated by MRI in depressed patients. In this study, we analyzed the PGV of medication-free moderately depressed MDD patients (N=34) and age and sex matched healthy controls (N=39). PGV did not differ between MDD patients and healthy controls [mean volume+/-S.D.; 0.76+/-0.17 cm3 and 0.75+/-0.14 cm3; ANCOVA, F1,69=1.25 p>0.05; respectively]. Our results confirm that volumetric PGV changes are not crucial for depression pathophysiology among unmedicated, moderately depressed adults.     

Amisulpride-induced hyperprolactinemia is reversible following discontinuation

BACKGROUND: Although amisulpride is considered to be a prolactin-raising atypical antipsychotic drug, a limited number of studies have documented the extent of its prolactin-elevating properties. In the present study the effect of amisulpride on plasma levels of prolactin and the reversibility of this untoward side effect were investigated. METHODS: 17 patients with various diagnoses received amisulpride (50-800 mg/day) or a combination of amisulpride plus other medication as needed. Plasma prolactin was determined 26.7+/-9.4 days (range: 13-50 days) after initiation of treatment and in 3 cases after a much longer period, and 14-51 days following its withdrawal. RESULTS: All patients on amisulpride had hyperprolactinemia (mean+/-S.D. prolactin levels: 62.5+/-33.0 ng/ml) with females exhibiting considerably higher prolactin levels than males. Following amisulpride discontinuation prolactin levels were significantly (p<000) reduced (mean+/-S.D. prolactin levels: 12.3+/-6.7 ng/ml). No significant correlation was detected between prolactin levels and either amisulpride dosage or duration of administration. CONCLUSION: Amisulpride has a pronounced prolactin-elevating effect which appears to be independent of dosage and duration of administration. Hyperprolactinemia rapidly reverses following amisulpride discontinuation.     

Effect of age and gender on citalopram and desmethylcitalopram steady-state plasma concentrations in adults and elderly depressed patients

The effect of aging on steady-state plasma concentrations of citalopram (CIT) and desmethylcitalopram (DCIT) was investigated in 128 depressive patients treated with 10-80 mg/day CIT. They were separated into three groups, with age up to 64 years (mean age+/-S.D.: 47+/-12 years; n=48), between 65 and 79 years (72+/-1 years; n=57), and from 80 years or older (84+/-1 years; n=23). Body mass index (BMI), renal and hepatic functions were similar in the three groups. A large interindividual variability of plasma levels of CIT (16-fold) and DCIT (12-fold) was measured for a given dose. The mean plasma levels of CIT corrected for a 20 mg daily dose were 55% higher in the very elderly (>=80 years) patients (65+/-30 ng/ml; p<0.001) and 38% higher in the elderly (65-79 years) patients (58+/-24 ng/ml; p<0.001) when compared to the adult patients (42+/-17 ng/ml). DCIT mean plasma level was 38% higher (p<0.05) in the group of very elderly patients (22+/-10 ng/ml) when compared to the adult patients (16+/-9 ng/ml). As a consequence, the mean plasma concentration of CIT+DCIT was 48% higher in the very elderly patients (86+/-36 ng/ml; p<0.001) and 33% higher in the elderly patients (77+/-28 ng/ml; p<0.001) when compared to the adult patients (58+/-21 ng/ml). Age correlated significantly with CIT (r=0.43, p<0.001), DCIT (r=0.28, p<0.01), and CIT+DCIT plasma levels (r=0.44, p<0.001), and thus accounts for 18% of the variability of CIT plasma levels, with no influence of gender. The recommended dose reduction of CIT in elderly patients seems therefore justified.     

Plasma brain-derived neurotrophic factor as a peripheral marker for the action mechanism of antidepressants

Numerous studies have demonstrated that depression is associated with a decreased expression of brain-derived neurotrophic factor (BDNF). BDNF shows antidepressant-like effects in animal models. Therefore, we tested the hypothesis that BDNF might be a peripheral marker for the mechanism of action of antidepressant agents in humans. Thirty-two patients meeting the DSM-IV criteria for major depressive disorder and 50 normal control subjects were recruited for this study. Plasma BDNF levels and Hamilton Depression Rating Scales were measured at baseline and 6 weeks after antidepressant administration. At baseline, the mean plasma BDNF level was lower in the depressive patients (698.1 +/- 537.7 pg/ml) than in the control subjects (830.7 +/- 624.8 pg/ml), although the difference was not statistically significant (p = 0.33). The plasma BDNF levels in depressive patients significantly increased from 698.1 +/- 537.7 to 1,028.9 +/- 744.5 after 6 weeks of antidepressant treatment (p = 0.01). Moreover, plasma BDNF levels were significantly increased after 6 weeks of treatment in the responder group, while there was no statistically significant change in the unresponsive group. These results suggest that the therapeutic response after antidepressant administration might be attributable to the increase in BDNF levels. BDNF may play a critical role in the action mechanism of antidepressant drugs. Further studies with a larger number of subjects are needed to verify these findings.     

Effects of short- and long-term risperidone treatment on prolactin levels in children with autism.

BACKGROUND: The effects of short- and long-term risperidone treatment on serum prolactin were assessed in children and adolescents with autism. METHODS: Patients with autism (N = 101, 5-17 years of age) were randomized to an 8-week trial of risperidone or placebo and 63 then took part in a 4-month open-label follow-up phase. Serum samples were obtained at Baseline and Week-8 (N = 78), and at 6-month (N = 43) and 22-month (N = 30) follow-up. Serum prolactin was determined by immunoradiometric assay; dopamine type-2 receptor (DRD2) polymorphisms were genotyped. RESULTS: Baseline prolactin levels were similar in the risperidone (N = 42) and placebo (N = 36) groups (9.3 +/- 7.5 and 9.3 +/- 7.6 ng/ml, respectively). After 8 weeks of risperidone, prolactin increased to 39.0 +/- 19.2 ng/ml, compared with 10.1 +/- 8.8 ng/ml for placebo (p < .0001). Prolactin levels were also significantly increased at 6 months (32.4 +/- 17.8 ng/ml; N = 43, p < .0001) and at 22 months (N = 30, 25.3 +/- 15.6 ng/ml, p < .0001). Prolactin levels were not associated with adverse effects and DRD2 alleles (Taq1A, -141C Ins/Del, C957T) did not significantly influence baseline levels or risperidone-induced increases in prolactin. CONCLUSIONS: Risperidone treatment was associated with two- to four-fold mean increases in serum prolactin in children with autism. Although risperidone-induced increases tended to diminish with time, further research on the consequences of long-term prolactin elevations in children and adolescents is needed.