Gender differences in a fenfluramine-activated FDG PET study of borderline personality disorder

Psychiatry Research: Neuroimaging 138 (2005) 183 – 195 www.elsevier.com/locate/psychresns

Gender differences in a fenfluramine-activated FDG PET study of borderline personality disorder Paul H. Soloff a,T, Carolyn Cidis Meltzerb, Carl Beckerb, Phil J. Greerb, Doreen Constantinea b

a Department of Psychiatry, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA Department of Radiology and PET Facility, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA

Received 24 June 2004; received in revised form 7 February 2005; accepted 25 February 2005

Abstract Neuroimaging studies of impulsive–aggressive subjects with borderline personality disorder (BPD) demonstrate hypometabolism in areas of prefrontal and frontal cortex, and a blunted cortical metabolic response to challenge with serotonergic agonists. Neuroendocrine responses to serotonergic challenge are known to vary greatly by gender, and may be related to sex differences in expression of impulsive aggression. We conducted single-blind, placebo-controlled fenfluramineactivated positron emission tomography (PET) studies in impulsive male and female subjects with BPD to look for gender differences in cortical response. The sample comprised 22 BPD (15 female, 7 male) and 24 control subjects (10 female, 14 male) who received placebo on Day 1 and d,l-fenfluramine on Day 2 before PET neuroimaging. In response to placebo, female, but not male, controls had areas of increased uptake of fluorodeoxyglucose-F18 in prefrontal cortex compared with BPD subjects, with greatest uptake in the medial orbital frontal cortex, bilaterally. Male, but not female, BPD subjects, showed areas of increased glucose utilization compared with controls in large areas of parietal and occipital cortex, bilaterally. In response to fenfluramine (relative to placebo), significant decreases in glucose uptake were found in male, but not female, BPD subjects, centered in the left temporal lobe. Female, but not male, control subjects showed significantly decreased uptake in areas of right frontal and temporal cortex. Covarying for impulsive-aggression rendered nonsignificant the gender differences in responses to placebo or fenfluramine. Gender differences in serotonergic function may mediate differences in behavioral expression of impulsive aggression in subjects with BPD. D 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Impulsivity; Aggression; Serotonin; Prefrontal cortex; Positron emission tomography; d,l-fenfluramine

1. Introduction T Corresponding author. Western Psychiatric Institute and Clinic, 3811 O’Hara St. Pittsburgh, PA 15213, USA. Tel.: +1 412 687 2666; fax: +1 412 621 2308. E-mail address: [email protected] (P.H. Soloff).

Impulsivity is a personality trait related to temperament, and a core characteristic of borderline personality disorder (BPD), where it is significantly

0925-4927/$ - see front matter D 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2005.02.008

184

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

associated with both suicidal and aggressive behavior. (Sanislow et al., 2002). Positron emission tomography (PET) studies of subjects ascertained for extremes of impulsive violence have demonstrated areas of decreased glucose utilization or hypoperfusion in frontal and temporal lobes compared with healthy controls (Raine et al., 1997; Soderstrom et al., 2000, 2002). Among criteria-defined subjects with BPD, aberrant patterns of glucose utilization have been described in areas of frontal and prefrontal cortex, including diminished uptake in ventral medial and orbital frontal cortex, although there are inconsistencies between studies (Goyer et al., 1994; De La Fuente et al., 1997; Soloff et al., 2003a,b; Juengling et al., 2003). The ventral medial cortex and the orbital frontal cortex are implicated in the regulation of impulsivity and aggressive behavior. Injury to ventral medial and orbital frontal cortex results in emotionally dysregulated, impulsive, aggressive, and socially inappropriate behaviors (Malloy et al., 1993; Damasio et al., 1994). Goyer et al. (1994) first demonstrated an inverse relationship between a lifetime history of aggression and regional cerebral metabolic rate of glucose utilization (rCMRglc) in orbital frontal, anterior medial frontal, left anterior frontal and right temporal cortex in military inpatients with personality disorders (PD). Within this group, patients with BPD had discrete areas of decreased or increased mean normalized rCMRglc in the frontal cortex compared with normal controls. De La Fuente et al. (1997) described nine discrete areas of bilateral relative hypometabolism in frontal and prefrontal cortex in BPD inpatients compared with healthy age-matched controls. Although patients were free of Axis I major depressive disorder (MDD), a mean score of 27 on the 24item Hamilton Rating Scale for Depression suggested a high degree of depressed mood co-morbidity in this sample. Juengling et al. (2003) found significant hypermetabolism in prefrontal and frontal cortex in female BPD inpatients compared with controls (with hypometabolism in hippocampus and cuneus). Patients in this study had no MDD, but they reported extensive diagnostic co-morbidity with anxiety disorders, and a high degree of depressive mood (e.g., mean Beck Depression Inventory = 20.7). Soloff et al. (2003a,b) controlled for both Axis I MDD and depressed mood in a study of female subjects with

BPD and found reductions in relative glucose utilization in BPD subjects narrowly limited to an area of medial orbital frontal cortex [Brodmann areas (BA) 9, 10, and 11], bilaterally. Differences between patients and controls were rendered insignificant when measures of impulsivity or impulsive aggression were covaried. Impulsive–aggressive and self-injurious behaviors in BPD (and other impulsive personality disorders) have long been associated with diminished central serotonergic function on the basis of indirect measures such as levels of 5-hydroxyindoleacetic acid in cerebrospinal fluid or the prolactin response to pharmacologic challenge with serotonergic agonists ˚ sberg et al., 1987; Oquendo and Mann, 2000, for (A reviews). Challenge studies using d,l-fenfluramine (FEN) or meta-chlorophenylpiperazine (m-CPP) in BPD or other impulsive PD subjects demonstrate a bbluntedQ prolactin response to the serotonergic agonist, and an inverse relationship between prolactin response and severity of impulsivity or impulsive aggression. Conducting PET neuroimaging studies during pharmacological activation with FEN (FENPET) adds brain localizing information not available in earlier physiological studies (Kapur et al., 1994; Mann et al., 1996). Using a FEN-activation paradigm in impulsive PD patients, Siever et al. (1999) reported decreased glucose uptake in response to FEN in orbital frontal, adjacent ventral medial and cingulate cortex in PD subjects compared with normal controls. These results were supported by a FEN-PET study in criteriadefined subjects with BPD (Soloff et al., 2000). New et al. (2002) expanded on the work of Siever et al. with impulsive–aggressive PD subjects but used m-CPP to challenge the serotonergic system during PET neuroimaging. In response to m-CPP, PD subjects showed significant decreases in relative metabolic rate in left medial orbital frontal cortex and anterior cingulate gyrus (BA 25) compared with healthy controls, and a significant increase in the right lateral orbital frontal cortex and posterior cingulate. There was an inverse relationship between metabolic response to m-CPP and clinical measures of impulsive aggression in lateral orbital frontal cortex. Taken together, these imaging studies suggest that impulsive aggression in BPD (and other impulsive-aggressive PDs) is mediated, in part, by diminished serotonergic function in neural circuits involving frontal and prefrontal cortex.

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

The phenomenology of impulsive aggression in BPD varies greatly by gender. For example, BPD males score higher than females on test measures of trait impulsivity, anger-hostility, and lifetime aggressive behavior (Soloff et al., 2003a,b). BPD males are more likely to direct aggression externally, while females tend to demonstrate more self-injurious behaviors. Healthy male control subjects also have significantly increased lifetime aggression compared with healthy females (Soloff et al., 2003a,b) and have different patterns of functional cortical activation in response to imagined aggression (Pietrini et al., 2000). Gender differences in behavioral expression of impulsive-aggression or cortical processing of aggressive imagery may be mediated by differences in underlying neurobiology, especially involving the central serotonin system. In healthy subjects, gender differences in serotonergic neurobiology have been reported in some, but not all, studies of the serotonin transporter (assessed by platelet 3H-imipramine or 3Hparoxetine binding), binding potential of the 5HT1A, and 5HT2A receptors, and neuroendocrine response to challenge with serotonin agonists, all potentially relevant to behavioral expression of impulsive aggression (reviewed below). PET neuroimaging studies of impulsive–aggressive BPD subjects have not directly addressed this issue. Gender differences may also contribute to differences in findings between studies. The current study builds upon our pilot experience with FEN-PET neuroimaging in criteria-defined subjects with BPD (Soloff et al., 2000) with an enlarged patient sample, allowing separate analyses by gender. We postulated that gender-related differences in serotonin function in impulsive subjects with BPD would be reflected in the cortical metabolic responses to FEN on PET neuroimaging. On the basis of earlier studies, we hypothesized that cortical responses at baseline and following FEN would be related to measures of impulsivity and impulsive-aggression.

2. Methods This study was approved by the Institutional Review Board of the University of Pittsburgh. Written informed consent was obtained from each participant. Physically healthy subjects were recruited from the investigators’ longitudinal studies on BPD, from

185

outpatient clinics of the Western Psychiatric Institute and Clinic, and by community advertisement. Subjects for the FEN-PET neuroimaging protocol were recruited from a larger sample of subjects who participated in fenfluramine challenge studies without imaging (Soloff et al., 2003a). All subjects were examined for Axis I disorders by the Structured Clinical Interview for DSM-III-R (SCID; Spitzer et al., 1988), and for Axis II disorders by the International Personality Disorders Examination (IPDE; Loranger et al., 1987). (DSM-III-R is used for continuity with longitudinal studies.) BPD was diagnosed by the Diagnostic Interview for Borderline Patients (DIB, scaled score z 7), with a 2-year time frame (Gunderson et al., 1981), and by the IPDE, for lifetime diagnosis. Patients were excluded for any lifetime diagnoses of bipolar disorder (type I, type II or not otherwise specified), organic mood disorders, or any psychotic disorders. Subjects with symptoms of current dependence or withdrawal from alcohol or other psychoactive substances were also excluded. Depressive symptoms were assessed by the 24-item version of the Hamilton Rating Scale for Depression within 1 week of the PET study (HamD-24; Guy, 1976). Impulsivity and aggression were assessed by the self-rated Barratt Impulsiveness Scale (BIS; Barratt, 1965) and the Brown–Goodwin Lifetime History of Aggression (LHA), a semi-structured interview (Brown et al., 1982). Healthy control subjects were medication free and had no Axis I or II diagnoses. All subjects had normal physical and laboratory examinations, including complete blood counts and tests of liver and thyroid function. All were free of psychoactive medication and oral contraceptives for 3 months or longer. Subjects were admitted on the evening before the PET study to the General Clinical Research Center (GCRC) of the University of Pittsburgh Medical Center, where they had a urine drug screen (for marijuana, cocaine, opiates, barbiturates, amphetamines, and benzodiazepines) and a pregnancy test. Subjects were excluded if either test was positive. Female patients and controls were studied in the first 10 days of the menstrual cycle. Food and fluids other than water were not permitted after 2300 h on nights preceding testing. FEN challenge studies were single-blind and placebocontrolled, and were conducted in a specialized

186

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

pharmacologic intervention room of the PET facility over 2 consecutive days. On Day 1, a placebo (PLA) was administered before the PET scan. On Day 2, patients received d,l-fenfluramine (FEN), in a dose of 0.8 mg/kg to a maximum of 60 mg. FEN was used in this study under an FDA-approved protocol (IND#54,952). The method for conducting the FEN challenge study has been standardized and is reported elsewhere (McBride et al., 1990). Baseline blood samples were collected at T = 15 min and T = 0 before ingestion of drug, and hourly thereafter for 5 h. Plasma was assayed for prolactin, fenfluramine and norfenfluramine metabolites. Four mCi of 18F-fluorodeoxyglucose (FDG) were injected intravenously 3 h after ingestion of PLA (Day 1) or FEN (Day 2) (Mann et al., 1996). A magnetic resonance imaging (MRI) study, performed before Day 1, was used for anatomical co-registration. 2.1. Imaging methods The ECAT 951r/31 scanner (CTI PET Systems, Knoxville, TN) was used for 15 subjects, 7 BPD patients and 8 controls. The newer ECAT HR + scanner was used for 31 subjects, 15 BPD patients and 16 controls. Advantages of the HR + scanner include higher spatial resolution (4.1-mm full-width at half-maximum in-plane) and a larger axial field of view (15 cm) for whole-brain sampling in 63 slices. Data from the two scanners were initially analyzed separately. Since similar results were found with both scanners independently, the total data set of 46 subjects was examined to optimize statistical power. For this combined analysis, similar reconstruction parameters were used for the PET images from each scanner (0.4 Hanning filter  2, 0.5 Ramp filter, no decay correction). Since both data sets had been acquired in two-dimensional (2D) mode (septa extended), no correction for scatter was applied. With these reconstruction parameters, the spatial resolution of the 951r/31 camera is measured as 10  10  6 mm, and the calculated resolution of the HR + scanner would be approximately 9  9  5 mm. Due to the similar resolution of the reconstructed data, both data sets were smoothed with a 12  12  12 mm smoothing kernel, which further served to minimize small differences between the spatial resolution of the two scanners.

Four mCi of FDG were injected intravenously, followed by a 40-min uptake during which the subject rested quietly, staring at a cross-hair on a blank wall. Subjects were then positioned in the scanner such that the lowest scanning plane was parallel to the canthomeatal line. Head movement was minimized through the use of a thermoplastic mask and headholder system. During scanning, subjects rested with eyes closed and ears unplugged in a dimly lit, quiet room. A 10-min transmission scan was acquired using rotating rods of 68Ge / 68Ga. Emission scanning was then performed for 45 min, collected as nine 5-min frames. PET data for each subject were registered to the subject’s volumetric spoiled gradient recalled (SPGR) MRI study using Automated Image Registration (AIR; Woods et al., 1993). The MR sequence was as follows: TE = 5, TR = 25, flip angle = 408, NEX = 1; field of view = 24 cm, image matrix = 256  192 pixels, acquired in the coronal plane at 1.5 Tesla (GE Medical Systems, Milwaukee, WI). The MR images were spatially normalized to the brain template developed by the Brain Imaging Center, Montreal Neurological Institute [distributed with Statistical Parametric Mapping-99 (SPM99)], using a linear transformation with AIR. The PET data were then converted into this standard space using the transforms created previously. Mean global activity levels within- and between-subject were removed by analysis of covariance (ANCOVA) on a pixel-by-pixel basis with mean global counts as covariates. Relative regional uptake is measured after normalization using whole brain uptake (Friston et al., 1995). Normalization does not affect degrees of freedom in the final analysis. SPM99 was used to analyze the PET data, with significance set at Z z 4.00, P b 0.05, corrected for multiple comparisons at the cluster level. Areas of significantly increased or decreased uptake were visually inspected for continuity using sequential MR images. Statistical contrasts conducted by SPM included the following: (1) Response to PLA by diagnosis (DX), i.e., BPD vs. Control, separately by gender, and in a combined gender sample; (2) response to FEN within DX, separately by gender, and in a combined gender sample; (3) response to FEN between DX, relative to placebo response, separately by gender, and in a combined gender sample; (4) repetition of those analyses with covariation for measures of impulsivity (BIS) and impul-

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

187

a. Female

b. Male

Fig. 1. Statistical Parametric Mapping (SPM99) projection images comparing FDG uptake in patients and control subjects in response to placebo. Areas of decreased FDG uptake in female BPD patients compared with control subjects are shown in (a). With a sample size of 25 female subjects, 2 groups, and 1 covariate, there are 22 degrees of freedom. Areas of increased FDG uptake in male BPD patients compared with control subjects are demonstrated in (b). With a sample size of 21 male subjects, 2 groups, and 1 covariate, there are 18 degrees of freedom.

sive aggression (LHA) separately; and (5) analysis of the interaction of Gender with DX on response to PLA and FEN. HamD-24 scores were covaried in every analysis, using the SPM bnuisanceQ covariate function (Fig. 1).

3. Results A total of 22 BPD subjects (15 females, 7 males) completed the placebo (Day 1) and FEN PET studies (Day 2), and were compared with 24 healthy control subjects (10 females, 14 males). The mean age of the BPD group (28.9 years) was not significantly different from that of the control group (27.0 years), though BPD males were older than male controls (Table 1). Axis I co-morbidity included three subjects with current MDE

(2 females, 1 male), and nine with dysthymic disorder (7 females, 2 males). The three subjects with current MDE were included in the study as their HamD-24 scores clustered with other BPD subjects, and excluding them produced no significant changes in relative glucose utilization compared with control subjects. Five subjects (2 males, 3 females) met criteria for a second Axis II disorder: Narcissistic PD (1 male), Histrionic PD (1 female), Passive–aggressive PD (1 female), Self-defeating PD (1 female), and Antisocial PD (1 male). Eight subjects (6 female, 2 male) met criteria for alcohol abuse, and one subject (male) for cannabis abuse in the month before the initial SCID intake interview. (All were drug and alcohol free for at least 1 week before admission to the GCRC, had no withdrawal symptoms, and had clean drug screens the evening before the PET study). A history of medically

Table 1 Clinical characteristics Female N Age HamD-24 BIS LHA

Male

BPD

Control

15 26.9 11.4 75.5 27.7

10 29.6 0.70 61.0 15.6

BPD vs. control

t(P), df = 23

BPD

Control

0.75 5.70 4.50 6.2

7 33.3 13.3 79.9 31.9

14 25.1 2.0 62.4 20.7

(NS) (b0.001) (b0.001) (b0.001)

t(P), df = 19

t(P), df = 44

2.43 5.3 4.08 4.56

46 0.81 7.08 6.04 6.34

(0.025) (0.001) (0.001) (b0.001)

(NS) (b0.001) (b0.001) (b0.001)

188

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

significant suicide attempts, i.e., a score of 2 or greater on the Medical Lethality Scale (Beck et al., 1975), was reported in 19 BPD subjects, self-mutilation without suicide attempts in three. As expected, BPD subjects were significantly more depressive (HamD-24), impulsive (BIS) and aggressive (LHA) than controls, by group and within gender (Table 1). For female BPD and control subjects, there was no significant difference in time from onset of menstrual flow to FEN challenge: BPD 8.5 (5.1) days versus Control 8.3 (8.2) days (t = 0.07, df = 17, NS). BPD subjects and controls received similar mean (SD) weight-relative doses of d,l-fenfluramine (BPD: 0.77 (0.18) mg/kg., controls: 0.76 (0.10) mg/kg., t = 0.12, df = 40, NS). There was no significant difference between groups in absorption of FEN, assessed as the maximal level of fenfluramine + norfenfluramine metabolites: BPD, 73.5 (19.5) ng/ml; control, 67.6 (21.2) ng/ml, t = 0.806, df = 30, NS). Prolactin respon-

ses to FEN were robust: e.g., Male control baseline prolactin = 6.23 (2.71) ng/ml, peak = 12.9 (8.7) ng/ml; Female control baseline prolactin = 5.8 (1.3) ng/ml, peak = 15.6 (4.0) ng/ml). Detailed neuroendocrine results on the larger study sample of 64 BPD and 57 controls are presented elsewhere (Soloff et al., 2003a). Mean (SD) radiation doses of administered [18F]-FDG did not differ significantly between diagnostic groups: BPD, 4.02 (0.14) mCi; Control, 4.00 (0.12) mCi; t = 1.15, df = 42, NS). 3.1. Imaging results 3.1.1. Response to placebo by diagnosis (DX: BPD vs. Control), separately by gender Fifteen female BPD subjects were compared with 10 female controls. BPD subjects had significantly decreased uptake of FDG compared with control females in the prefrontal cortex (PFC), with the greatest decrease in the medial orbital frontal cortex

Table 2 Anatomical location of clusters Contrast

Cluster size K, ( P corr)

Z

Peak voxel x, y, z (mm)

Anatomical definition of clusters

3012 (0.001) 2496 (0.004) 11,384 (b0.001)

4.69 3.91 4.02

12, 64, 16T 42, 16, 16 16, 70, 30

1487 (0.06)

3.44

8, 60,

Bilateral frontal lobe, medial orbital gyrus Rt. temporal lobe, medial gyrus Rt. parietal–occipital lobe; rt. parietal lobe, inferior gyrus; lt. occipital lobe, angular gyrus; bilateral superior cerebellum Bilateral frontal lobe, medial orbital gyrus

B. Response to FEN: within-group comparisons Female control: placebo N FEN 3045 (b0.001) Placebo N FEN 1285 (0.01) Male BPD: placebo N FEN 3549 (b0.001)

4.27 3.23 4.86

14, 28, 58 52, 26, 10 32, 12, 8T

Male BPD: placebo N FEN Combined: control placebo N FEN Placebo N FEN

997 (0.02) 5312 (b0.001) 1282 (0.01)

4.78 4.06 4.05

60, 44, 16T 24, 48, 36 14, 68, 4

C. Response to FEN: between-group comparisons Female: control vs. BPD 1456 (0.006) Male: control vs. BPD 2875 (b0.001)

3.52 4.90

18, 30, 54 34, 14,

Male: control vs. BPD Male: control vs. BPD

4.64 4.33

58, 44, 18TT 18, 32, 38

A. Placebo: BPD vs. control Female: control N BPD Control N BPD Male: BPD N control

Combined: control N BPD

1079 (0.01) 886 (0.03)

20

8T

Rt. frontal lobe, superior gyrus Rt. temporal lobe, superior gyrus Lt. temporal lobe, medial gyrus; lt. fronto-parietal lobe and post-central gyrus Lt. temporal lobe, superior gyrus Rt. frontal lobe, superior, medial, precentral gyri Rt. occipital lobe, lingual gyrus

Rt. frontal lobe, superior gyrus Lt. temporal lobe, medial gyrus; parietal lobe; post-central gyrus Lt. temporal lobe, superior gyrus, parietal lobe Rt. posterior cingulate and white matter

Results from SPM99 showing cluster size with P values (corrected for multiple comparisons), Z scores, coordinates of peak voxels (Montreal Neurological Institute Atlas), and anatomical definition of clusters. T Peak voxel Z z 4.00, P b 0.05, corrected for multiple comparisons. TT Peak voxel Z z 4.00, P = 0.07, corrected for multiple comparisons.

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

189

(BA 11), bilaterally (Table 2). A second large area of decreased uptake in BPD subjects was noted in the medial gyrus of the right temporal lobe (BA 21), though falling short of our significance threshold (Z z 4). There were no areas in which BPD subjects demonstrated greater uptake than controls. Seven male BPD subjects were compared with 14 healthy male controls. There were no areas of decreased FDG uptake in male BPD subjects compared with controls; however, BPD males demonstrated a very large area of increased uptake in the parietal–occipital cortex (BA 19) bilaterally. This large area of increased metabolic activity extended into the inferior gyrus of the parietal lobe on the right, the gyrus angularis of the occipital lobe on the left (BA 19 and 39), and the superior aspect of the cerebellum bilaterally. In a combined gender sample, 22 BPD subjects were compared with 24 healthy controls. BPD subjects demonstrated a nonsignificant trend for decreased uptake compared with control subjects in the right medial orbital frontal cortex (BA 11), attributable to the decreased uptake in female subjects alone, though much diminished in size and significance. There were no areas in which BPD subjects demonstrated greater uptake compared with controls.

increased or decreased uptake relative to placebo response. In contrast, male BPD subjects demonstrated significantly decreased uptake in response to FEN in large areas of the left temporal lobe, centered on the left medial temporal gyrus, with extensions into the left frontal and parietal lobes, and the postcentral gyrus. A second, smaller, noncontiguous area of decreased uptake was noted in the left superior temporal gyrus, with extension into the left parietal lobe and angular gyrus. There were no areas of significantly increased uptake in male BPD subjects following FEN (Fig. 2).

3.1.2. Response to FEN: within-group comparisons

3.1.3. Response to FEN: between-group comparisons Response to FEN was compared between DX (each relative to their own placebo response) separately by gender, and in a combined-gender sample. In the comparison of female BPD subjects to controls, an area of decreased uptake was noted in the right superior frontal gyrus (BA 8 and 9). This difference fell short of our significance threshold (Z z 4), and it reflects a greater decrease in uptake in female control subjects in response to FEN compared with the response of BPD females. In a comparison of the response to FEN of male control and BPD subjects (each relative to placebo), the following three noncontiguous areas of decreased uptake were noted: (a) an area centered in the left medial temporal lobe, with extensions into the left parietal lobe and postcentral gyrus; (b) an area of left superior temporal gyrus, with extension into the parietal lobe; and (c) an area encompassing part of the right posterior cingulate gyrus (BA 31) and associated white matter. These differences largely

3.1.2.1. Female subjects. In a comparison of uptake after FEN to uptake after placebo (PLA), areas of increased activation (e.g. FEN N PLA ) and decreased activation (e.g. FEN b PLA ) were computed within each diagnostic group (DX). The response to FEN within DX was computed separately by gender, and in a combined gender sample. Female control subjects demonstrated significant decreases in uptake relative to placebo in a large area of right frontal cortex, centered in the superior frontal gyrus (BA 8 and 9), and, at a trend level, in right temporal lobe, centered in the superior temporal gyrus (Table 2). There were no significant areas of increased uptake following FEN in female controls. Among female BPD subjects, there were no areas with significant increases or decreases in uptake following FEN relative to placebo response. 3.1.2.2. Male subjects. In response to FEN, male control subjects showed no areas of significantly

3.1.2.3. Combined gender sample. In response to FEN, control subjects demonstrated a robust decrease in a large area of the right frontal lobe, with centers of greatest significance in the superior frontal gyrus (BA 9), the medial frontal gyrus, and the right precentral gyrus (BA 6). An area of right occipital cortex, the lingual gyrus, also showed decreased metabolic activity with FEN in control subjects. There were no areas with significantly increased activity in control subjects following FEN. BPD subjects in the combined gender sample demonstrated no significant areas of increased or decreased uptake in response to FEN relative to placebo (Fig. 3).

190

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

a. Female Controls

b. Male BPD

Fig. 2. Within-group comparisons of response to FEN (relative to placebo) illustrate decreased FDG uptake in female control (a) and male BPD subjects (b).

reflect the robust decreases in metabolic response to FEN among BPD males relative to male control subjects. In the combined gender sample, there were no significant differences between control and BPD subjects in response to FEN relative to placebo.

In separate analyses, our measures of trait impulsivity (BIS) and impulsive aggression (LHA) were entered as covariates in each contrast. As before, the significance threshold for these analyses was maintained at Z z 4.0, P b 0.05 for cluster size, controlled for multiple comparisons. Covarying for LHA ren-

a. Female

b. Male

Fig. 3. Comparisons of responses to FEN between diagnostic groups, each relative to placebo and separated by gender. Areas of significantly decreased uptake are noted in the comparison of female BPD subjects to controls (a), reflecting a greater decrease in uptake among female control subjects in response to FEN compared with the response of BPD females. Areas of decreased uptake in the comparison of male BPD subjects to controls (b) reflect the greater decrease in uptake among BPD males in response to FEN compared with the response of control subjects.

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

dered insignificant all differences between BPD and control groups in response to Placebo on all contrasts (male, female and combined gender). Covarying for BIS scores did not alter previously demonstrated significant differences. Differences in the response to FEN within diagnostic groups (i.e. response to placebo vs. response to FEN in BPD and control subjects, separately by gender) were rendered insignificant by both BIS and LHA as covariates. There were no meaningful significant interactions between gender and diagnosis on response to FEN.

4. Discussion We found marked gender differences in relative glucose utilization in response to placebo and FEN in both BPD and control subjects. At baseline (placebo response), female, but not male, BPD subjects have areas of decreased uptake compared with control subjects, while male, but not female, BPD subjects, show areas of increased glucose utilization compared with controls. Relative prefrontal hypometabolism, previously reported in combined gender samples of BPD subjects (Goyer et al., 1994; De La Fuente et al., 1997), is found only among female subjects with BPD in this study. Differences between BPD and control subjects in FDG uptake in response to placebo are rendered nonsignificant by covarying for impulsive aggression (LHA), but not trait impulsivity (BIS), suggesting that baseline metabolic differences may be more closely related to the neurobiology of behavioral aggressivity than to impulsivity, which is a more general construct. While our hypothesis implies an aberration in serotonergic function, it is important to note that metabolic response to placebo may be unrelated to serotonin function and reflect other correlates of neuronal glucose metabolism such as glutamate neurotransmission, which is the major correlate of neuronal glucose consumption (Shulman, 2001). In response to FEN (relative to placebo), significant decreases in glucose uptake are found in male but not female BPD subjects, and in female but not male controls. The patterns of uptake at baseline and in response to FEN appear so gender-specific that significant differences between BPD and control groups (each relative to placebo) are lost when the

191

genders are combined. Covarying for either LHA or BIS renders insignificant differences within groups in the response to FEN, suggesting that this serotonergic response is related to the mediation of impulsivity and impulsive aggression. Imaging studies of impulsive violence (e.g., murderers, violent criminal offenders, and aggressive psychiatric patients) have generally been conducted on male subjects in specialized settings, and have reported hypoperfusion or decreased metabolism in frontal and temporal cortex related to impulsive aggression or severity of psychopathy (Raine et al., 1997; Soderstrom et al., 2000, 2002). Male BPD subjects in the present study did not demonstrate the relative prefrontal hypometabolism that has been associated with impulsive aggression in BPD females (Soloff et al., 2003b). However, they had a robust gender-specific decrease in relative glucose uptake in frontal and temporal lobes following FEN compared with male control subjects (but not seen in BPD females). Gender differences in the responsiveness of the central serotonergic system may mediate differing expressions of impulsive aggression in males and females with BPD. (Male aggression may also involve other biologic factors, such as testosterone.) Important differences were also noted in laterality of metabolic responses to FEN in both groups. These findings should be viewed as descriptive results as SPM does not conduct statistical tests on laterality as an independent variable. Control females showed decreased activation in right frontal and temporal areas. BPD males demonstrated areas of decreased metabolic function in left frontal and temporal lobes. Laterality of findings may be relevant to regulation of impulsive aggression. Decreased relative glucose utilization in the left frontal cortex following activation with FEN or m-CPP has previously been reported in subjects with impulsive–aggressive personality disorders (New et al., 2002) and BPD (Soloff et al., 2000), suggesting decreased serotonergic function in this region. Regulation of emotion, especially impulsive anger and hostility, may be predominately left frontal and temporal functions (New et al., 2002). In contrast, decreased activation to FEN in right hemisphere has been reported as a normative response in healthy female subjects (Mann et al., 1996). Gender differences have been reported in many functions of the central serotonergic system, including

192

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

the neuroendocrine response to serotonergic challenge with FEN (McBride et al., 1990). The inverse relationship between impulsive aggression and indices of central serotonergic function has been widely replicated among male patients with BPD (or other impulsive PDs) (Soloff et al., 2003b), and among nonpatient or healthy male subjects in community settings (Muldoon et al., 1996; Cleare and Bond, 1997; Manuck et al., 1998) but less consistently among females. Some (Hollander et al., 1994; Stein et al., 1993; Martial et al., 1997; New et al., 1997; Soloff et al., 2003b), but not all (Herpertz et al., 1997; Rinne et al., 2000), studies involving female BPD subjects have failed to demonstrate either bbluntingQ of prolactin responses or significant inverse relationships between prolactin responses and measures of behavioral impulsivity. Gender-related differences have also been reported in some studies of the human platelet serotonin transporter, assessing the maximum binding capacity (B max) and dissociation constants (K d) of 3H-imipramine or 3H-paroxetine binding to the serotonin transporter complex. In some studies (Klompenhouwer et al., 1990; Halbreich et al., 1991; Marazziti et al., 1998), though not all (Langer et al., 1980), young females have a significantly lower K d (higher affinity) than young males. A significant interaction between gender and age may result in a reversal of affinity over time, i.e., aged females have a higher K d (lower affinity) than aged males (Marazziti et al., 1998). Gender may affect seasonal variability in 3Himipramine binding, with B max lowest in spring in women, but not men (Soria et al., 1996). Post-mortem studies of nonpsychiatric subjects also show gender-related differences in studies of the serotonin transporter site, though results are inconsistent. For example, Arato´ et al. (1991) reported higher B max values for 3H-imipramine binding in women compared with men in right orbital cortex, an important site for regulation of impulsivity and aggression. Arora and Meltzer (1989) found a significantly higher B max in males in frontal cortex using 5HT-sensitive 3H-imipramine binding. Gender-related differences are reported in some studies of the binding potential of the 5HT1A and 5HT2A receptors in healthy subjects (Biver et al., 1996; Parsey et al., 2002), and of aging effects on the 5HT1A receptor (Meltzer et al., 2001). Females have

increased presynaptic 5HT1A binding compared with males. Some studies (Biver et al., 1996), but not all (Baeken et al., 1998), also report decreased postsynaptic 5HT2A binding in healthy females compared with males, suggesting postsynaptic down-regulation. A pattern of increased transporter availability and presynaptic 5HT1A binding, and postsynaptic 5HT2A down-regulation may be related to impulse regulation in healthy females (Arato´ et al., 1991). An inverse relationship between age and binding potential of the 5HT1A receptor in men but not women may be related to differential susceptibility to psychiatric disorders with age (Meltzer et al., 2001). Gender differences are also found in personality traits associated with gene polymorphisms for the serotonin transporter promoter and tryptophan hydroxylase (Manuck et al., 1999; Du et al., 2000; Baca-Garcia et al., 2002). For example, the bshortQ variant of the serotonin transporter promoter gene is associated with conduct disorder, aggressivity and attention deficit hyperactivity disorder in males (Cadoret et al., 2003). The bUQ allele of the tryptophan hydroxylase gene in males, but not females, is associated with blunting of the prolactin response to FEN, higher aggression and anger-related personality traits (Manuck et al., 1999). Gender differences in the neurobiology of the serotonin system clearly play an important role in the behavioral expression of impulsive aggression. 4.1. Limitations Contrary to expectation, we found no areas of increased metabolic activity (relative to placebo) following FEN challenge in either the male or female control samples, despite evidence of robust prolactin responses. Mann et al. (1996) and Kapur et al. (1994) also noted areas of relative regional decreases in metabolic activity in response to FEN in healthy controls; however, the major findings of their studies were of increased metabolic activity in wide areas of frontal, temporal and temporal–parietal cortex. Technical considerations may contribute to different findings in these FEN-PET studies. Image acquisition in our study occurs during a brief 45min bwindowQ within a 5-h FEN challenge. FDG injection is timed for 3 h post-ingestion of FEN to capture the peak pharmacologic effects of FEN as

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

defined by the prolactin response in normal subjects. However, prolactin data reveal great variation between subjects in time-to-peak, and may vary with psychiatric diagnosis. Martial et al. (1997) found that time-to-peak in female BPD subjects (e.g. 2.2 F 0.3 h) was significantly earlier than in controls (3.25 F 0.2 h). Differences in timing of FDG injection and scanning relative to the peak effect of FEN can alter imaging results as FEN-mediated increases and decreases in cortical metabolic rate may occur at different times. In female subjects, prolactin responses to FEN vary greatly with phase of menstrual cycle. If uncontrolled, variations in menstrual phase could increase variability in response to FEN, potentially affecting results in small sample studies. Histories of suicidal behavior and lethality of attempts may contribute to variation in cortical metabolic response to FEN in our BPD sample. Suicidal behavior is related to decreased central serotonergic function independent of impulsivity or impulsive aggression and may share related but not identical neurobiologic mediation (Virkkunen et al., 1987, 1989; Stanley et al., 2000). Oquendo et al. (2003) recently reported greater regional metabolic decreases following FEN in high-lethality attempters with major depressive disorder compared with lowlethality attempters. Relative regional glucose utilization was inversely related to lethality and positively correlated with impulsivity in preselected regions of interest. We controlled for severity of depressed mood by covarying HamD-24 scores rather than excluding three subjects meeting categorical criteria for a current major depressive episode. HamD-24 scores of these three subjects were continuous with other scores and did not represent statistical outliers. Our results are virtually unchanged from a preliminary study of 13 BPD females without major depressive disorder, using HamD-24 as covariate (Soloff et al., 2003a). Discrete cerebral blood flow and metabolic abnormalities have been described in subjects with MDE in areas of prefrontal cortex including dorsolateral, medial, orbital frontal PFC, and anterior cingulate (subgenual PFC) (Drevets, 2000; Bremner et al., 2002). These aberrant perfusion and metabolic changes may be state-dependent, correlating with depressive symptom severity (Mayberg et al., 1999; Drevets, 2000; Brody et al., 2001). We did not

193

control for lifetime history of major depressive disorder, which would be difficult in a study of BPD. Persisting reductions in volume in orbital frontal PFC and hippocampus have been reported in remitted subjects with major depressive disorder, which could potentially confound results (Bremner et al., 2002).

Acknowledgments This research was supported by NIMH grant MH48463 (P.H.S.) and NIHCRC grant MO1RR00056.

References Arato´, M., Frecska, E., Tekes, K., MacCrimmon, D.J., 1991. Serotonergic interhemispheric asymmetry gender differences in the orbital cortex. Acta Psychiatrica Scandinavica 84, 110 – 111. Arora, R.C., Meltzer, H.Y., 1989. 3H-imipramine binding in the frontal cortex of suicides. Psychiatry Research 30, 125 – 135. 2sberg, M., Schalling, D., Tr7skman-Benz, L., Wagner, A., 1987. Psychobiology of suicide, impulsivity, and related phenomena. In: Meltzer, H.Y. (Ed.), Psychopharmacology: The Third Generation of Progress. Raven Press, New York. Baca-Garcia, E., Vaquero, C., Diaz-Sastre, C., Saiz-Ruiz, J., Fernandez-Piqueras, J., de Leon, J., 2002. A gender-specific association between the serotonin transporter gene and suicide attempts. Neuropsychopharmacology 26, 692 – 695. Baeken, C., D’haenen, H., Flamen, P., Mertens, J., Terriere, D., Chavatte, K., Boumon, R., Bossuyt, A., 1998. 123I-5-I-R91150, a new single photon emission tomography ligand for 5-HT2A receptors: influence of age and gender in healthy subjects. European Journal of Nuclear Medicine 25 (12), 1617 – 1622. Barratt, E.L., 1965. Factor analysis of some psychometric measures of impulsiveness and anxiety. Psychological Reports 16, 547 – 554. Beck, A.T., Beck, R., Kovacs, M., 1975. Classification of suicidal behavior: I. Quantifying intent and medical lethality. American Journal of Psychiatry 132, 285 – 287. Biver, F., Lotstra, F., Monclus, M., Wikler, D., Damhut, P., Mendlewicz, J., Goldman, S., 1996. Sex difference in 5HT2 receptor in the living human brain. Neuroscience Letters 204 (1–2), 25 – 28. Bremner, J.D., Vythilingam, M., Vermetten, E., Nazeer, A., Adil, J., Khan, S., Staib, L.H., Charney, D.S., 2002. Reduced volume of orbitofrontal cortex in major depression. Biological Psychiatry 51, 273 – 279. Brody, A.L., Saxena, S., Mandelkern, M.A., Fairbanks, L.A., Ho, M.L., Baxter Jr., L.R., 2001. Brain metabolic changes associated with symptom factor improvement in major depressive disorder. Biological Psychiatry 50, 171 – 178.

194

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195

Brown, G.L., Ebert, M.H., Goyer, P.F., Jimerson, D.C., Klein, W.J., Bunney, W.E., Goodwin, F.K., 1982. Aggression, suicide, and serotonin: relationships to CSF amine metabolites. American Journal of Psychiatry 139, 741 – 746. Cadoret, R.J., Langbehn, D., Caspers, K., Troughton, E.P., Yucuis, R., Sandhu, H.K., Philibert, R., 2003. Associations of the serotonin transporter promoter polymorphism with aggressivity, attention deficit and conduct disorder in an adoptee population. Comprehensive Psychiatry 44 (2), 88 – 101. Cleare, A.J., Bond, A.J., 1997. Does central serotonergic function correlate inversely with aggression? A study using d-fenfluramine in healthy subjects. Psychiatry Research 69, 89 – 95. Damasio, H., Grabowski, T., Frank, R., Galaburda, A.M., Damasio, A.R., 1994. The return of Phineas Gage: clues about the brain from the skull of a famous patient. Science 264, 1102 – 1105. De La Fuente, J.M., Goldman, S., Stanus, E., Vizuete, C., Morlan, I., Bobes, J., Mendlewicz, J., 1997. Brain glucose metabolism in borderline personality disorder. Journal of Psychiatric Research 31, 531 – 541. Drevets, W.C., 2000. Neuroimaging studies of mood disorders. Biological Psychiatry 48, 813 – 829. Du, L., Bakish, D., Hrdina, P.D., 2000. Gender differences in association between serotonin transporter gene polymorphism and personality traits. Psychiatric Genetics 10, 159 – 164. Friston, K.J., Holmes, A.P., Worsley, K.J., Poline, J.B., Frith, C.D., Frackowiak, R.S.J., 1995. Statistical parametric maps in functional imaging: a general approach. Human Brain Mapping 2, 189 – 210. Goyer, P.F., Andreason, P.J., Semple, W.E., Clayton, A.H., King, A.C., Compton-Toth, B.A., Schulz, S.C., Cohen, R.M., 1994. Positron-emission tomography and personality disorders. Neuropsychopharmacology 10, 21 – 28. Gunderson, J.G., Kolb, J.E., Austin, V., 1981. The Diagnostic Interview for Borderlines. American Journal of Psychiatry 138, 896 – 903. Guy, W., 1976. ECDEU Assessment Manual of Psychopharmacology—Revised (DHEW Publ. No. ADM 76-338). Superintendent of Documents. U.S. Government Printing Office, Washington, DC. Halbreich, U., Rojansky, N., Zander, K.J., Barkai, A., 1991. Influence of age, sex, and diurnal variability on imipramine receptor binding and serotonin uptake in platelets of normal subjects. Journal of Psychiatric Research 25, 7 – 18. Herpertz, S., Sass, H., Favazza, A., 1997. Impulsivity in selfmutilative behavior: psychometric and biologic findings. Journal of Psychiatric Research 31, 451 – 465. Hollander, E., Stein, D.J., DeCaria, C.M., Cohen, L., Saoud, J.B., Skodol, A.E., Kellman, D., Rosnick, L., Oldham, J.M., 1994. Serotonergic sensitivity in borderline personality disorder: preliminary findings. American Journal of Psychiatry 151, 277 – 280. Juengling, F.D., Schmahl, C., Hesslinger, B., Ebert, D., Bremner, J.D., Gostomzyk, J., Bohus, M., Lieb, K., 2003. Positron emission tomography in female patients with borderline personality disorder. Journal of Psychiatric Research 37, 109 – 115.

Kapur, S., Meyer, J., Wilson, A.A., Houle, S., Brown, G.M., 1994. Modulation of cortical neuronal activity by a serotonergic agent: a PET study in humans. Brain Research 646, 292 – 294. Klompenhouwer, J.-L., Fekkes, D., van Hulst, A.M., Moleman, P., Pepplinkhuizen, L., Mulder, P.G.H., 1990. Seasonal variations in binding of 3H-paroxetine to blood platelets in healthy volunteers: indications for a gender difference. Biological Psychiatry 28, 509 – 517. Langer, S.Z., Briley, M.S., Raisman, R., Henry, J.-F., Morselli, P.L., 1980. Specific 3H-imipramine binding in human platelets. Archives of Pharmacology 313, 189 – 194. Loranger, A.W., Susman, V.L., Oldham, J.M., Russakoff, L.M., 1987. The Personality Disorder Examination: a preliminary report. Journal of Personality Disorders 1, 1 – 3. Malloy, P., Bihrle, A., Duffy, J., 1993. The orbitomedial frontal syndrome. Archives of Clinical Neuropsychology 8, 185 – 201. Mann, J.J., Malone, K.M., Diehl, D.J., Perel, J., Nichols, T.E., Mintun, M.A., 1996. Positron emission tomographic imaging of serotonin activation effects on prefrontal cortex in healthy volunteers. Journal of Cerebral Blood Flow and Metabolism 16, 418 – 426. Manuck, S.B., Flory, J.D., McCaffery, J.M., Matthews, K.A., Mann, J.J., Muldoon, M.F., 1998. Aggression, impulsivity and central nervous system serotonergic responsivity in a nonpatient sample. Neuropsychopharmacology 19, 287 – 299. Manuck, S.B., Flory, J.D., Ferrell, R.E., Dent, K.M., Mann, J.J., Muldoon, M.F., 1999. Aggression and anger related traits associated with a polymorphism of the tryptophan hydroxylase gene. Biological Psychiatry 45, 603 – 614. Marazziti, D., Rossi, A., Palego, L., Barsanti, A., Carrai, M., Giannaccini, G., Serra, P., Lucacchini, A., Cassano, G.B., 1998. Effect of aging and sex on the [3H]-paroxetine binding to human platelets. Journal of Affective Disorders 50, 11 – 15. Martial, J., Paris, J., Leyton, M., Zweig-Frank, H., Schwartz, G., Teboul, E., Thavundayil, J., Larue, S., Kin, N.M.K., Nair, N.P., 1997. Neuroendocrine study of serotonin function in female borderline personality disorder patients: a pilot study. Biological Psychiatry 42, 737 – 739. Mayberg, H.S., Liotti, M., Brannan, S.K., McGinnis, S., Mahurin, R.K., Jerabek, P.A., Silva, J.A., Tekell, J.L., Martin, C.C., Lancaster, J.L., Fox, P.T., 1999. Reciprocal limbic–cortical function and negative mood: converging PET findings in depression and normal sadness. American Journal of Psychiatry 156, 675 – 682. McBride, P.A., Tierney, H., DeMeo, M., Chen, J.-S., Mann, J.J., 1990. Effects of age and gender on CNS serotonergic responsivity in normal adults. Biological Psychiatry 27, 1143 – 1155. Meltzer, C.C., Drevets, W.C., Price, J.C., Mathis, C.A., Lopresti, B., Greer, P.J., Villemagne, V., Holt, D., Mason, N.S., Houck, P.R., Reynolds, C.F., DeKosky, S.T., 2001. Gender-specific aging effects on the serotonin 1A receptor. Brain Research 895 (1–2), 9 – 17. Muldoon, M.F., Manuck, S.B., Jansma, C.L., Moore, A.L., Perel, J., Mann, J.J., 1996. d,l-Fenfluramine challenge test: experience in non-patient sample. Biological Psychiatry 39, 761 – 768.

P.H. Soloff et al. / Psychiatry Research: Neuroimaging 138 (2005) 183–195 New, A.S., Hazlett, E.A., Buchsbaum, M.S., Goodman, M., Reynolds, D., Mitropoulou, V., Sprung, L., Shaw Jr., R.B., Koenigsberg, H., Platholi, J., Silverman, J., Siever, L.J., 2002. Blunted prefrontal cortical [F-18] fluorodeoxyglucose positron emission tomography response to meta-chlorophenylpiperazine in impulsive aggression. Archives of General Psychiatry 59, 621 – 629. New, A.S., Trestman, R.L., Mitropoulou, V., Benishay, D., Coccaro, E., Siever, L.J., 1997. Serotonergic function and self injurious behavior in personality disorder patients. Psychiatry Research 69 (1), 17 – 26. Oquendo, M.A., Mann, J.J., 2000. The biology of impulsivity and suicidality. Psychiatric Clinics of North America 23, 11 – 25. Oquendo, M.A., Placidi, G.P.A., Malone, K.M., Campbell, C., Keilp, J., Brodsky, B., Kegeles, L.S., Cooper, T.B., Parsey, R.V., van Heertum, R.L., Mann, J.J., 2003. Positron emission tomography of regional brain metabolic responses to a serotonergic challenge and lethality of suicide attempts in major depression. Archives of General Psychiatry 60, 14 – 22. Parsey, R.V., Oquendo, M.A., Simpson, N.R., Ogden, R.T., Van Heertum, R., Arango, V., Mann, J.J., 2002. Effects of sex, age and aggressive traits in man on brain serotonin 5 HT1A receptor binding potential measured by PET using [C-11] WAY-100635. Brain Research 954 (2), 173 – 182. Pietrini, P., Guazzelli, M., Basso, G., Jaffe, K., Grafman, J., 2000. Neural correlates of imaginal aggressive behavior assessed by positron emission tomography in healthy subjects. American Journal of Psychiatry 157 (11), 1172 – 1781. Raine, A., Buchsbaum, M., LaCasse, L., 1997. Brain abnormalities in murderers indicated by positron emission tomography. Biological Psychiatry 42, 495 – 508. Rinne, T., Westenberg, H.G.M., den Boer, J.A., van den Brink, W., 2000. Serotonergic blunting to meta-chlorophenylpiperazine (m-CPP) highly correlates with sustained childhood abuse in impulsive and autoaggressive female borderline patients. Biological Psychiatry 47, 548 – 556. Sanislow, C.A., Grilo, C.M., Morey, L.C., Bender, D.S., Skodol, A.E., Gunderson, J.G., Shea, M.T., Stout, R.L., Zanarini, M.C., McGlashan, T.H., 2002. Confirmatory factor analysis of DSMIV criteria for borderline personality disorder: findings from the collaborative longitudinal personality disorders study. American Journal of Psychiatry 159, 284 – 290. Shulman, R.G., 2001. Functional imaging studies: linking mind and basic neuroscience. American Journal of Psychiatry 158, 11 – 20. Siever, L.J., Buchsbaum, M.S., New, A.S., Spiegel-Cohen, J., Wei, T., Hazlett, E., Sevin, E., Nunn, M., Mitropoulou, V., 1999.

195

d,lQFenfluramine response in impulsive personality disorder assessed with [18F]- fluorodeoxyglucose positron emission tomography. Neuropsychopharmacology 20, 413 – 423. Soderstrom, H., Tullberg, M., Wikkelso, C., Ekholm, S., Forsman, A., 2000. Reduced regional cerebral blood flow in nonpsychotic violent offenders. Psychiatry Research: Neuroimaging 98, 29 – 41. Soderstrom, H., Hultin, L., Tullberg, M., Wikkelso, C., Ekholm, S., Forsman, A., 2002. Reduced frontotemporal perfusion in psychopathic personality. Psychiatry Research: Neuroimaging 114, 81 – 94. Soloff, P.H., Kelly, T.M., Strotmeyer, S.J., Malone, K.M., Mann, J.J., 2003a. Impulsivity, gender and response to fenfluramine challenge in borderline personality disorder. Psychiatry Research 119, 11 – 24. Soloff, P.H., Meltzer, C.C., Becker, C., Greer, P.J., Kelly, T.M., Constantine, D., 2003b. Impulsivity and prefrontal hypometabolism in borderline personality disorder. Psychiatry Research: Neuroimaging 123, 153 – 163. Soloff, P.H., Meltzer, C.C., Greer, P.J., Constantine, D., Kelly, T.M., 2000. A fenfluramine-activated FDG-PET study of borderline personality disorder. Biological Psychiatry 47, 540 – 547. Soria, P.R., Vallejo, J., Oros, M., Menchon, J.M., Cortadellas, T., Navarro, M.A., Urretavizcaya, M., 1996. Seasonal variability in blood platelet 3H-imipramine binding in healthy controls: age and gender effects. Biological Psychiatry 40, 981 – 985. Spitzer, R.L., Williams, J.B.W., Gibbon, M., First, M.B., 1988. Instruction Manual for the Structured Clinical Interview for DSM-III-R (SCID). Biometrics Research, New York State Psychiatric Institute, New York. Stanley, B., Molcho, A., Stanley, M., Winchel, R., Gameroff, M.J., Parsons, B., Mann, J.J., 2000. Association of aggressive behavior with altered serotonergic function in patients who are not suicidal. American Journal of Psychiatry 157 (4), 609 – 614. Stein, D.J., Hollander, E., Cohen, L., Frenkel, M., Saoud, J.B., DeCaria, C., Aronowitz, B., Levin, A., Liebowitz, M.R., Cohen, L., 1993. Neuropsychiatric impairment in impulsive personality disorders. Psychiatry Research 48, 257 – 266. Virkkunen, M., Nuutila, A., Goodwin, F.K., Linnoila, M., 1987. Cerebrospinal fluid metabolite in male arsonists. Archives of General Psychiatry 44, 241 – 247. Virkkunen, M., DeJong, J., Bartko, J., Goodwin, F.K., Linnoila, M., 1989. Relationship of psychobiological variables to recidivism in violent offenders and impulsive fire-setters. A follow-up study. Archives of General Psychiatry 46, 600 – 603. Woods, R.P., Mazziotta, J.C., Cherry, S.R., 1993. MRI–PET registration with automated algorithm. Journal of Computer Assisted Tomography 17, 536 – 546.