Long-term validity of biological markers of psychopathy and criminal recidivism: follow-up 6–8 years after forensic psychiatric investigation

Psychiatry Research 121 (2004) 281–291

Long-term validity of biological markers of psychopathy and criminal recidivism: follow-up 6–8 years after forensic psychiatric investigation E. Gunilla Stalenheim* Departments of Neuroscience, Psychiatry, University Hospital, SE-751 85 Uppsala, Sweden Received 2 October 2002; received in revised form 11 June 2003; accepted 16 July 2003

Abstract This study is a follow-up investigation of a forensic psychiatric sub-population 6–8 years after forensic psychiatric evaluation. The aim was to examine the long-term validity of biological markers of psychopathy and antisocial behavior over time. Data on criminal records were obtained at follow-up from the National Council for Crime Prevention. Basic data included findings of psychiatric and psychological assessments, as well as values for serum triiodothyronine (T3) and free thyroxin (FT4), and platelet monoamine oxidase (MAO) activity, all obtained during the forensic psychiatric examination. Criminal recidivists at follow-up had higher serum T3 levels than non-recidivists, and much higher values than normal controls, while their levels of free T4 were lower. The T3 levels in criminal recidivists correlated to psychopathy- and aggression-related personality traits as measured by the Karolinska Scale of Personality. In violent recidivists, a remarkably high correlation was noted between T3 levels and Irritability and Detachment, traits that have previously been linked to the dopaminergic system. Stepwise multiple regression analyses confirmed the relationships of T3 levels and platelet MAO activity with personality traits in criminal recidivists. The predictive validity of biological markers of psychopathy, T3 and platelet MAO, measured during forensic psychiatric investigation, is stable over time. The results indicate chronic alterations of the hypothalamic–pituitary–thyroid axis in this group of subjects. 䊚 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Triiodothyronine; Thyroxin; Platelet MAO activity; Karolinska Scale of Personality; Criminality; Follow-up

1. Introduction The hypothalamic–pituitary–thyroid (HPT) axis has been the focus of numerous studies of somatic disorders and of depression. Nevertheless, little is known about how thyroid hormones influ*Tel.: q46-18-611-00-00; fax: q46-18-51-58-10. E-mail address: [email protected] (E. Gunilla Stalenheim).

ence personality and its behavioral manifestations. Levander et al. (1987), in a study of criminally active boys who were institutionalized and who had been exposed to severe psychosocial strain from early infancy, found that these boys had significantly higher levels of triiodothyronine (T3) than did normal schoolboys. In a long-term followup of another group of delinquents, individual T3 levels were shown to be risk indicators for per-

0165-1781/04/$ - see front matter 䊚 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2003.07.002

282

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

sistent criminality (Alm et al., 1996b). Previous studies of the present sub-population indicated that there is an intimate relationship between the thyroid hormones T3 and free thyroxin (FT4) and abuse and antisocial behavior (Stalenheim et al., 1996). Moreover, serum levels of thyroid hormones measured in adult life have been found to be retrospectively related to a history of conduct disorder (Ramklint et al., 2000). There has long been evidence of a close relationship between stress and thyroid function (Mason, 1968). Of special interest for the present investigation is the fact that an unusual thyroid profile characterized by high T3 levels has been observed in refugees. This finding has now been confirmed in studies of Vietnam combat veterans, Israel combat veterans and World War II veterans, and also in refugees from East Germany (Bauer et al., 1994; Mason et al., 1994, 1996). The elevated levels of T3 in persons with post-traumatic stress disorder (PTSD) have been related to the personality trait of Novelty Seeking and hyper-arousal symptoms. The observed alterations of thyroid function associated with PTSD symptoms seem to be chronic, as they can be detectable even after 50 years (Wang et al., 1995, 1997; Wang and Mason, 1999). However, the robust elevations observed in veterans with combat-related PTSD probably would not be discovered in routine clinical thyroid function tests. The elevations of T3 are still within the normal range as defined in the field of clinical endocrinology and do not indicate glandular pathology but point to a potentially clinically significant hormone–symptom relationship (Wang et al., 1997). Several studies have also revealed associations between alcoholism and thyroid hormones (Baumgartner et al., 1994; Heinz et al., 1996). In a previous report, we have shown that the relationship between T3 and criminal behavior and alcoholism persists after adjustment for liver function (Stalenheim et al., 1996). The reason for the deviant hormone levels in the populations referred to is still unknown. Decreased monoamine oxidase (MAO) B activity has been related to characteristics such as Impulsiveness, Sensation-Seeking and Aggressiveness (Buchsbaum et al., 1976; Puchall et al., 1980;

von Knorring et al., 1984; Zuckerman, 1984; Schalling et al., 1987). However, it is also related to disinhibited symptoms of conditions such as psychopathy, suicidal behavior and hyperactivity, and to alcoholism, especially type II alcoholism as defined by von Knorring et al. (Schalling et al., 1987; von Knorring et al., 1991; Devor et al., 1993; Oreland and Hallman, 1995). The relationship between platelet MAO activity and psychopathy, criminality and criminal recidivism is now well documented, although the exact mechanisms are still unknown (Lidberg et al., 1985; Alm et al., 1996a; Garpenstrand et al., 2002). In the present study population, previous research has confirmed this relationship of platelet MAO activity with psychopathy and suicidal behavior (Stalenheim et al., 1997; Stalenheim, 2001). Previous studies of forensic psychiatric patients in Sweden have indicated the presence of different biological markers of vulnerability for criminality (Alm et al., 1996a; Belfrage et al., 1992; Stalenheim et al., 1996, 1997, 1998). The stability of these findings needs to be established by means of prospective studies. Follow-up studies of forensic psychiatric populations have mainly aimed at prediction of violent recidivism. Psychopathy has been found to be one heavy risk factor for recidivism in violent criminality in Sweden, as in other countries (Grann et al., 1999; Hare et al., 2000). Several biological correlates of psychopathy and socially deviant behavior have been reported, but the relevance of these findings and the biological background of antisocial behavior are still unknown. 1.1. Aims of the study The aims of the present study, using a clear-cut prospective design, were to investigate – whether previous data defining thyroid hormones and platelet MAO activity as biological markers of psychopathy and antisocial behavior are valid over time and – whether these relationships are mediated through personality characteristics that may indicate underlying biological mechanisms.

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

2. Methods 2.1. Research subjects This follow-up study comprised 60 male subjects who underwent a major forensic psychiatric examination at the Department of Forensic Psychiatry in Uppsala during the period 1992–1994. For detailed information on the collection of the cohort, see Stalenheim and von Knorring (1996). Their mean age was 34.0"11.0 (range 18–56) years. Persons with obvious psychotic or severe somatic disorders, those with mental retardation and those who had difficulty in speaking the Swedish language were excluded from the study. The subjects had committed serious criminal acts, most of them violent, and the courts had found them guilty. The overall psychiatric morbidity among the 60 subjects was high. Although psychotic persons were excluded, all of the subjects had been diagnosed as having at least one DSM-III-R Axis I psychiatric andyor Axis II personality disorder (American Psychiatric Association, 1987). The most frequent Axis I disorder (ns36) was substance abuse or dependence, and the second most frequent one was anxiety disorder (ns13). Antisocial personality disorder (ns19) was diagnosed in 19 subjects and borderline personality disorder in 17. However, 10 of these subjects had both diagnoses. For detailed information, see Stalenheim and von Knorring (1996). Almost all subjects were regular smokers during the forensic psychiatric examinations, and thus smoking will not influence comparison between criminal subgroups of assessments made at this time point. Routine laboratory screening in all research subjects included the liver function test gamma glutamyltransferase (gamma-GT). Blood samples were taken during the forensic psychiatric examination for assays of hormones and platelet MAO activity. They were taken fasting between 07.00 and 08.00 h, after at least 1 week at the department. A group of 66 healthy male medical students (mean age 26.5"4.4 years; range 21– 43) served as controls for the thyroid hormone levels. They had all given informed consent. All had studied at the medical faculty for at least 2

283

years. Detailed information on past or current physical or psychiatric disorders was collected as well as medication. Subjects with abnormal liver or thyroid function tests were excluded. Blood samples were obtained and analyzed in the same way as those of the research subjects. All subjects gave their informed consent to the study, which had been approved by the Ethics Committee of the Medical Faculty of Uppsala University. 2.2. Assessments made at the forensic psychiatric investigation 2.2.1. DSM-III-R Axis I and II diagnoses The subjects were diagnosed on the basis of a semi-structured interview for Axis I and II DSMIII-R diagnoses (SCID I and II) (Spitzer et al., 1987, 1992). All information obtained at the forensic psychiatric investigations, including ward observations, was used to complete the diagnoses. A team including a psychiatrist, a psychologist and a social worker conducted the major forensic investigations. Extensive filed information was available for all patients, including previous and current medical and psychiatric records, criminal and social records, previous forensic and psychological assessments, and police reports concerning the present offence(s). 2.2.2. The Karolinska Scale of Personality Karolinska Scale of Personality (KSP) ratings were obtained from 58 of the 60 subjects. The KSP is a self-report instrument for measurements of stable temperament traits. The scales were developed by Schalling et al. (Schalling, 1978; Schalling and Edman, 1993), partly through modifications of existing scales covering different aspects of personality. The purpose was to find stable traits correlating with relevant biological markers that, in combination, would constitute vulnerability factors for different kinds of psychopathology. The KSP is a questionnaire with 135 items concerning a subject’s habitual behavior and feelings. The response is given in a 4-point scale from 1s‘Does not apply at all’ to 4s‘Applies completely’. The 135 items are grouped into 15 scales: Somatic anxiety, Psychic anxiety, Muscular

284

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

tension, Social desirability, Impulsiveness, Monotony avoidance, Detachment, Psychasthenia, Socialization, Indirect aggression, Verbal aggression, Irritability, Suspicion, Guilt and Inhibition of aggression. Some scales that cover specific temperament profiles have been condensed to specified higher-order factors. The impulsive sensation seeking factor (psychopathy factor) is composed of three psychopathy-related scales (scores for ImpulsivenessqMonotony avoidanceySocialization). The KSP aggressivity-related scales are based on theories of aggressivity by Buss (1961). The aggression factor is built on this theory (scores of Indirect aggressionqVerbal aggressionqIrritability). Personality traits as assessed by means of the KSP have been found to be stable after 9 years (Gustavsson et al., 1997). Moreover, within the present population, psychopathy and aggressiveness have been shown to be stable over time independently of life events (Stalenheim, 1997). The raw scores of the KSP scales were transformed into T-scores (xymean of healthy volunteersystandard deviation (S.D.) of the healthy volunteers=10q50). The comparison group comprised 228 males and 240 females drawn from the general population (Schalling and Edman, 1993). Thus, the comparison group would have a mean T-score of 50 and an S.D. of 10. 2.2.3. Psychopathy Checklist-Revised The psychopathy checklist-revised (PCL-R) is an item checklist that measures interpersonal, affective and behavioral characteristics by scores on a 3-point scale (Hare, 1991). The scores are summed to form a total score, with a maximum of 40 points. With this method, psychopathy is described as a constellation of rather specific deviant traits and behaviors. The assessment includes interpersonal and affective traits such as selfishness, callousness and remorseless use of others (Factor 1) and behavioral manifestations such as a chronically unstable and antisocial lifestyle and social deviance (Factor 2). Ratings were performed carefully, by interview and from all information obtained during the forensic psychiatric investigation, by a PCL-R-trained forensic psychiatrist, formally authorized to use the PCL-R. A reliability test was performed using file-based ret-

rospective testing of psychopathy. The following intraclass correlation coefficients were found: total scores0.88, Factor 2s0.89 and Factor 1s0.69 (Grann et al., 1998). Thus, the dimensional agreement was very good, at the same level as that reported by Hare et al. (1990). 2.2.4. Thyroid hormones All blood samples for hormonal tests were analyzed similarly and simultaneously, using the same standards. After centrifugation, the serum was frozen at y70 8C. FT4 in serum was measured with a time-resolved competitive fluoroimmunoassay (Delfia Free Thyroxine kit, Wallac OY, Turku, Finland), using mouse monoclonal antibodies and based upon the back-titration principle. The withinand between-assay values for coefficient of variation (CV) within the reference range (9–21 pmoly l) were 2.2 and 5%, respectively. Total T3 in serum was measured with a competitive solid phase double action antibody radioimmunoassay (Pharmacia T3 RIA, Kabi Pharmacia Diagnostics, Uppsala, Sweden). The within- and between-assay CVs within the reference range (1.2–2.8 nmolyl) were 2.6 and 3.2%, respectively. Thyroid-stimulating hormone (TSH) was measured with a timeresolved sandwich fluoroimmunoassay (Delphia TSH Ultra, Wallac OY, Turku, Finland). Three different mouse monoclonal antibodies against hTSH were used. One was directed against hTSH and immobilized onto the walls of microtiter wells. The other two were in solution, labeled with europium chelate and specific for the b-subunit of the h-TSH. The TSH values were expressed in Uy l using reference standard, the Second International Reference Preparation (80y558) of hTSH for immunoassay. The minimal detection limit was 0.003 mUyl. The within- and between-assay CVs within the reference range (0.3–4.0 mUyl) were 1.7 and 2.3%, respectively. 2.2.5. Thrombocyte monoamine oxidase activity Platelet MAO activity was measured with 2phenylethylamine (b-PEA) and tryptamine as substrates, as described by Hallman et al. (1987). Platelet-rich plasma was prepared, the platelet count was measured with a Coulter Counter (Dunstable, England) and the plasma was kept

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

frozen at y70 8C until analyzed. For estimation of the enzyme activity, 14C-tryptamine and 14C-bPEA were used as substrates. Analyses with both substrates resulted in linear relationships between amount of enzyme and activity. A correlation coefficient of 0.92 (P-0.0001) between the activities against the two substrates was obtained, indicating a high degree of reliability of the method. Only the results with b-PEA were used in the calculations. 2.3. Follow-up 2.3.1. Criminal recidivism Register files for all 60 subjects were obtained from the National Council for Crime Prevention in Sweden. The Council is an agency under the Ministry of Justice and is responsible for the official Swedish juridical statistics. Detailed information about new prosecutions, types of crimes and sentences was recorded. According to a report by the National Board of Health and Welfare (2002) about mentally disordered offenders in Sweden, 27.9% of patients sentenced to forensic psychiatric care relapsed into criminality during their forensic psychiatric treatment, many of them committing violent crimes. Such crimes also occur in prison. We have therefore chosen to calculate the time from verdict and forward for all the patients. Thus, the study embraced all crimes for which the subjects were convicted after the forensic psychiatric evaluation. 2.4. Statistical methods Differences between means were tested by the use of one-wayythree-group analysis of variance (ANOVA) and one-wayyfour-group ANOVA. Correlations were sought by means of the linear correlation coefficient, and exact P-values are given. Stepwise multiple regression analyses were used to search the relationships between T3 levels and the different KSP scales and platelet MAO activity, with controls for age and signs of liver disease. 3. Results All subjects were convicted in connection with the forensic psychiatric investigation. Thirty per-

285

cent of the subjects (ns18) were referred to forensic psychiatric hospitals for inpatient treatment because of severe psychiatric disorders. The remaining subjects (ns42) were given a prison sentence or placed under probation within the criminal justice system. There were no differences between these groups regarding age, PCL-R scores, KSP scores or frequency of abuseydependence disorders. Moreover, analyses of data from the National Council for Crime Prevention showed no differences between these groups regarding frequency of crime recidivism, either violent or nonviolent. Before the follow-up, six of the subjects had died, all of them after release from prison or psychiatric ward. Four subjects had not been discharged from prison since being sentenced for the index crime leading to forensic psychiatric evaluation. The length of time ‘in freedom’ varied from 0 (ns4) to 7 years. In total, criminal recidivism, including both violent and non-violent crimes, was recorded in approximately 50% of the subjects (ns31), and violent crimes were committed by approximately 25% of the subjects (ns16). Not surprisingly, there was a very strong relationship between psychopathy as assessed by means of the PCL-R, especially PCL-R Factor 2 scores, and criminal recidivism (Table 1). The mean total PCL-R score in the study population was 18.7"11.9. The highest PCL-R scores were noted in the group with violent recidivism, while non-recidivists had the lowest scores. However, four subjects registered for violent criminality during the follow-up period (4y16) had fairly low PCL-R scores (i.e. between 12 and 17). There was no statistical relationship between PCL-R scores of psychopathy and either serum concentrations of thyroid hormones or platelet MAO activity. The study population had a highly increased T3 level and a decreased level of free T4 compared with normal controls. The difference in T3 level was even more pronounced when the subjects were divided into those with and those without criminal recidivism. No relationship was found between TSH level and criminal behavior. Although the serum T3 level was not related to psychopathy as assessed by PCL-R criteria, it was correlated with psychopathy- and aggression-related personality traits as measured by the KSP. In

286

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

Table 1 Criminal recidivism in males followed up after forensic psychiatric investigation (ns60)

Criminal recidivism total Violent Non-violent No criminal recidivism Controls (only T3) ANOVA

n

PCL-R total score

PCL-R Factor 1 score

PCL-R Factor 2 score

T3

FT4

31 16 15 29 66

24.1"10.6 26.1"8.7 22.0"12.3 13.0"10.9 – Fs8.49 Ps0.0006

10.5"4.1 11.0"3.3 9.5"4.9 7.1"5.2 – Fs4.11 Ps0.0215

11.0"5.8 12.3"5.0 9.6"6.5 5.0"5.0 – Fs10.24 Ps0.0002

2.22"3.9 2.23"0.44 2.20"0.34 1.99"0.32 1.85"0.22 Fs11.32 P-0.0001

13.05"2.05 13.17"2.31 12.91"1.77 13.25"2.72 14.65"2.04 Fs4.91 Ps0.003

Comparison of degree of psychopathy as assessed by the PCL-R scores, and comparison of serum T3 and free T4 concentrations in criminal groups. Mean total PCL-R scores and Factor 1 and Factor 2 scores"S.D. and serum hormone concentrations (mean"S.D.) presented with calculated one-way ANOVA, F- and P-values given.

fact, in the group of criminal recidivists, a correlation emerged between serum T3 level and scores of Impulsiveness (rs0.44, Ps0.02), Socialization (rsy0.42, Ps0.02), Irritability (rs0.45, Ps 0.02) and Detachment (rs0.54, Ps0.003). The associations between Detachment and T3 and between Irritability and T3 were even stronger in the group of violent criminal recidivists (Fig. 1a and b). However, Detachment and Irritability are per se correlated in the total study population (rs 0.52), with the strongest correlation in the group of violent recidivists (rs0.67), and the next strongest in the non-violent recidivists (rs0.61). No significant correlation was found between personality variables and T3 and T4 in the nonrecidivist group. Since no control group was available, the exact values of platelet MAO activity are not presented. There is no significant difference between groups with or without criminal recidivism in mean level of platelet MAO activity. However, after division of the study population into groups with low or high MAO activity, a difference appeared. Low platelet MAO activity was associated with high scores on the KSP Psychopathy and Aggression factors in violent recidivists, but not in non-violent recidivists and not in subjects without criminal relapse (Table 2). Finally, a multiple stepwise regression model was used to further demonstrate the close relationship in the group of criminal recidivists (ns27) between personality traits and the biological markers of antisocial behavior. Actually, 64% of the

variance of the T3 concentrations could be explained by Detachment, platelet MAO activity and liver injury (gamma-GT) (Table 3). 4. Discussion The very interesting direct association between the serum concentrations of the thyroid hormone triiodothyronine and antisocial behavior was found to persist in this prospective study: subjects with the highest concentrations of T3 relapsed into criminality. In the study population there appeared to be a direct association between T3 hormone levels and behavioral manifestations of psychopathy. The involvement of the HPT axis in psychopathic individuals seems to be persistent, starting at young ages and persisting into adulthood (Levander et al., 1987; Alm et al., 1996b; Ramklint et al., 2000). The similarity with the alterations of thyroid function in patients with PTSD symptoms is striking, especially since T3 is associated with hyperarousal symptoms of PTSD, which includes symptoms such as irritability and unpredictable aggressive behavior (Wang and Mason, 1999). Our study population, although heterogeneous, consisted mainly of psychosocially maladjusted subjects (Stalenheim and von Knorring, 1996). In spite of the vast literature on the HPT system over the last 60 years, little is known about how thyroid hormones influence personality and its behavioral expressions. It would be of great value if individual medical histories of all sorts of trauma, e.g. prenatal and postnatal trauma, child abuse and sub-

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

Fig. 1. (a) Relationship between serum T3 concentrations and Karolinska Scales of Personality ratings of Detachment at follow-up 6–8 years after forensic psychiatric examination. Subjects with violent criminal recidivism, ns14y58, presented in figure. (b) Relationship between serum T3 concentrations and Karolinska Scales of Personality ratings of Irritability at follow-up 6–8 years after forensic psychiatric examination. Subjects with violent criminal recidivism, ns14y58, presented in figure.

stance abuse and head injuries, were available. Such histories are not easily obtained retrospectively. It has been proposed that elevated T3 concentrations may be due to adaptation mecha-

287

nisms developed early in life as a response to severe strain (Alm, 1996). However, individual factors and exposure interact as causes of aggression and antisocial behavior. Animal and human studies have suggested that 5-hydroxytryptamine (5-HT) is a key modulator of aggressive behavior (Golden et al., 1991). It has been proposed that decreased MAO activity may reflect some property of the central serotonergic system, such as the density, size or functional capacity (Oreland and Shaskan, 1983; Oreland and Hallman, 1995). The present results confirm the role of platelet MAO activity as a biological marker of psychopathy, in line with previous findings: it was found here to be related to common personality factors among criminal recidivists, namely, psychopathy- and aggressionrelated personality traits. It is possible that there are common genetic factors that might lead to vulnerability to abuse, criminality and emotional difficulties and thereby contribute to an individual’s criminal development. 5-HT exerts regulatory influence on dopamine (DA) turnover, and an increased CSF concentration of the DA metabolite homovanillic acid (HVA) in combination with a low CSF concentration of the serotonin metabolite 5-HIAA indicates ˚ a regulatory imbalance (Agren et al., 1986; Risby et al., 1987). The importance of such imbalance was illustrated in a recent study by Soderstrom et al. (2001), who found that the CSF concentration of HVA and its ratio to that of 5-HIAA correlated with all aspects of psychopathy. This was especially true regarding the behavioral features captured by Factor 2 of the PCL-R, such as Impulsivity, Irresponsible lifestyle and Aggression. They conclude that this seems to link the outwarddirected aggression of psychopathy to serotonergic hypo-functioning and high DA turnover, which might account for disinhibition of destructive impulses (Soderstrom et al., 2001). In this context it is interesting to note the unexpected high correlation between the serum T3 concentration and personality traits such as Detachment and Irritability. These traits have previously been linked to the dopamine-2 receptor (Farde et al., 1997; Breier et al., 1998; Laakso et al., 2000). However, multiple comparisons in small populations, such as corre-

288

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

Table 2 Association between KSP Psychopathy- and Aggression-related personality traits and platelet MAO activity (ns54) in males followed up after forensic psychiatric investigation Criminal recidivism

n Lowyhigh MAO

Violent recidivism Non-violent recidivism No criminal recidivism ANOVA

9y5 5y8 14y13

KSP psychopathy factor

KSP aggression factor

MAO low

MAO high

MAO low

MAO high

112.1"36.9 84.2"59.2 87.9"27.8 F 10.85 Ps0.0019 MAO

40.8"18.0 85.2"32.2 58.8"31.9

186.7"23.0 158.5"41.4 158.1"23.7 F 11.57 Ps0.0014

128.7"21.9 158.9"20.5 140.5"23.6

The study population is divided into groups with high or low MAO activity (cutting point at b-PEA median values12.57). Mean KSP factor scores"S.D. in subjects with criminal recidivism, violent or non-violent, and with no crime recidivism are presented. ANOVA F- and P-values are given.

lations between serum T3 levels and scores of KSP traits pose statistical problems. It is extremely hard to achieve the level of significance using the Bonferroni procedure for multiple comparisons, which would demand a level of significance of 0.003. This level is reached in the correlation between serum levels of T3 and scores of Detachment, indicating a very strong connection. The role of DA in violent criminality should be further investigated, especially in the light of the emotional disturbances, type-II alcoholism and aggressiveness recorded in violent criminal recidivists. Not surprisingly, the PCL-R scores, like the serum concentrations of T3, were highest among the criminal recidivists. Yet, within the group of criminal recidivists, individual PCL-R scores were not correlated with any biological marker, whereas individual KSP scores revealed significant relationships between T3 concentrations and psychopathyand aggression-related traits. The explanation for this may be sought in the difference between the

two instruments: The PCL-R, on the one hand, has been shown to have a very good ability to predict criminal recidivism. Although the PCL-R is a dimensional instrument, some of the item scores, such as criminality and delinquency, are not normally distributed within the normal population. In the present population there was a group with very high and a group with very low scores, with a similar pattern for PCL-R Factors 1 and 2. The KSP, on the other hand, was developed with the aim of finding biological correlates to and ultimately etiological factors underlying disorders and personality deviances. It reflects personality traits that are truly normally distributed within the normal population. The psychopathy-related traits of the KSP also seemed to be almost normally distributed in the study population. Despite these differences, the characteristics measured by the instruments are highly correlated (Stalenheim and von Knorring, 1998).

Table 3 Multiple stepwise regression for subjects with criminal recidivism (ns27), with T3 as the dependent variable and the separate scales from the Karolinska Scales of Personality, age, liver function test (gamma-GT) and platelet MAO activity as independent variables At final step: Independent variables

Multiple Rs0.80 Coefficient

Fs13.25 S.E.

P-0.0001 F

Intercept Detachment Liver function test Platelet MAO activity

1.76 0.01 0.24 y0.04

0.26 0.03 0.07 0.12

16.02 11.65 10.49

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

The population of this study was chosen with the aim of detecting a vulnerability to antisocial behavior by identifying possible biological markers. The predictive validity of such biological markers regarding criminal relapse has previously been considered to be low, and they have been classified as pathophysiological markers. Such markers are important, since they may serve as a source of etiological hypothesis and influence treatment (Nurnberger, 1992). Interestingly, within the present population, no correlation exists between measurements of the biological markers and scores of the strongest predictor of criminal recidivism, namely, the PCL-R. One reason may be that the serum levels of T3 and platelet MAO activity are almost normally distributed within the study population and that the subjects were all euthyroid. For the reasons mentioned, it is unlikely that the biological markers recorded in this study could serve as tests that per se are able to predict criminality in juridical settings. The unique composition of the study population makes it possible to illuminate the pathology found in other populations in a magnified manner. However, the results obtained cannot be generalized to the population at large, or to criminal women, but are limited to the type of population investigated, males in forensic psychiatry. There is a strong need for long-term follow-up studies of larger samples of patients of this category. The results of the study show that the biological markers of psychopathy T3 and platelet MAO activity remain stable when assessed in prospective data collected at follow-up. These two biological markers are already related to antisocial behavior at young ages, which might indicate genetic influence in the development of psychopathy. In forensic psychiatric practice in Sweden, patients of this type are currently treated with a combination of SSRIs and antipsychotics, both traditional and atypical. The present findings may support this pharmacotherapy, which otherwise is not evidencebased. References ˚ Agren, H., Mefford, I.N., Rudorfer, M.V., Linnoila, M., Potter, W.Z., 1986. Interacting neurotransmitter systems. A non-

289

experimental approach to the 5HIAA-HVA correlation in human CSF. Journal of Psychiatric Research 20, 175–193. Alm, P., 1996. Juvenile and adult criminality: relationships to platelet MAO activity, triiodothyronine, ADHD, conduct disorder and psychopathy. Forensic Psychiatry Thesis, The Karolinska Institute, Stockholm, p. 125. ¨ Alm, P., af Klinteberg, B., Humble, K., Leppert, J., Sorensen, S., Thorell, L.-H., Lidberg, L., Oreland, L., 1996a. Psychopathy, platelet MAO activity and criminality among former juvenile delinquents. Acta Psychiatrica Scandinavica 94, 105–111. ¨ Alm, P.O., af Klinteberg, B., Humble, K., Leppert, J., Sorensen, S., Tegelman, R., Thorell, L.H., Lidberg, L., 1996b. Criminality and psychopathy as related to thyroid activity in former juvenile delinquents. Acta Psychiatrica Scandinavica 94, 112–117. American Psychiatric Association, 1987. Diagnostic and Statistical Manual of Mental Disorders. Third ed. rev. American Psychiatric Press, Washington, DC. ¨ K.-J., Baumgartner, A., ¨ Bauer, M., Priebe, S., Kurten, I., Graf, 1994. Psychological and endocrine abnormalities in refugees from East Germany: Part 1. Prolonged stress, psychopathology, and hypothalamic–pituitary–thyroid axis activity. Psychiatry Research 51, 61–73. Baumgartner, A., Rommelspacher, H., Otto, M., Schmidt, L.G., ¨ K.J., Campos-Barros, A., Platz, W., 1994. Kurten, I., Graf, Hypothalamic–pituitary–thyroid (HPT) axis in chronic alcoholism. 1. HPT axis in chronic alcoholics during withdrawal and after 3 weeks of abstinence. Alcoholism, Clinical and Experimental Research 18, 284–294. Belfrage, H., Lidberg, L., Oreland, L., 1992. Platelet monoamine oxidase activity in mentally disordered violent offenders. Acta Psychiatrica Scandinavica 85, 218–221. Breier, A., Kestler, L., Adler, C., Elman, I., Wiesenfeld, N., Malhotra, A., Pickar, D., 1998. Dopamine D2 receptor density and personal detachment in healthy subjects. American Journal of Psychiatry 155, 1440–1442. Buchsbaum, M.S., Coursey, R.D., Murphy, D.L., 1976. The biochemical high-risk paradigm: behavioural and familial correlates of low platelet monoamine oxidase activity. Science 194, 339–341. Buss, A.G., 1961. The Psychology of Aggression. Wiley, New York. Devor, E.J., Cloninger, C.R., Hoffman, P.L., Tabakoff, B., 1993. Association of monoamine oxidase (MAO) activity with alcoholism and alcoholic subtypes. American Journal of Medical Genetics 48, 209–213. ¨ Farde, L., Gustavsson, J.P., Jonsson, E., 1997. D2 dopamine receptors and personality traits. Nature 385, 590. Garpenstrand, H., Longato-Stadler, E., af Klinteberg, B., Grigorenko, E., Damberg, M., Oreland, L., Hallman, J., 2002. Low platelet monoamine oxidase activity in Swedish imprisoned criminal offenders. European Neuropsychopharmacology 12, 135–140. Golden, R.N., Gilmore, J.H., Corrigan, M.H., Ekstrom, R.D., Knight, B.T., Garbutt, J.C., 1991. Serotonin, suicide, and

290

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291

aggression: clinical studies. Journal of Clinical Psychiatry 52 (Suppl), 61–69. ¨ ¨ Grann, M., Langstrom, N., Tengstrom, A., Stalenheim, E.G., ˚ 1998. The reliability of file-based retrospective ratings of psychopathy with the PCL-R. Journal of Personality Assessments 70, 416–426. Grann, M., Langstrom, N., Tengstrom, A., Kullgren, G., 1999. Psychopathy (PCL-R) predicts violent recidivism among criminal offenders with personality disorders in Sweden. Law and Human Behavior 23, 205–217. ¨ Gustavsson, J.P., Weinryb, R.M., Goransson, S., Pedersen, ˚ N.L., Asberg, M., 1997. Stability and predictive ability of personality traits across 9 years. Personality and Individual Differences 22, 783–791. Hallman, J., Oreland, L., Edman, G., Schalling, D., 1987. Thrombocyte amine oxidase activity and personality traits in women with severe premenstrual syndrome. Acta Psychiatrica Scandinavica 76, 225–234. Hare, R.D., Harpur, T.J., Hakstian, A.R., Forth, A.E., Hart, S.D., Newman, J.P., 1990. The Revised Psychopathy Checklist: reliability and factor structure. Psychological Assessments 2, 338–341. Hare, R.D., 1991. The Hare Psychopathy Checklist-Revised. Multi-Health Systems, Canada, Toronto, Ontario. Hare, R.D., Clark, D., Grann, M., Thornton, D., 2000. Psychopathy and the predictive validity of the PCL-R: an international perspective. Behavioral Sciences and the Law 18, 623–645. ¨ K.J., ¨ Heinz, A., Bauer, M., Kuhn, S., Kruger, F., Graf, Rommelspacher, H., Schmidt, L.G., 1996. Long-term observation of the hypothalamic–pituitary–thyroid (HPT) axis in alcohol-dependent patients. Acta Psychiatrica Scandinavica 93, 470–476. Laakso, A., Vilkman, H., Kajander, J., Bergman, J., Paranta, M., Solin, O., Hietala, J., 2000. Prediction of detached personality in healthy subjects by low dopamine transporter binding. American Journal of Psychiatry 157, 290–292. ˚ Schalling, D., Dalteg, A., 1987. Levander, S., Mattson, A., Psychoendocrine patterns within a group of male juvenile delinquents as related to early psychosocial stress, diagnostic classification, and follow-up data. In: Magnusson, D., ¨ ˚ (Eds.), Psychopathology: An Interactional PerOhman, A. spective. Academic Press, Inc, Orlando, FL, pp. 235–252. Lidberg, L., Modin, I., Oreland, L., Tuck, J.R., Gillner, A., 1985. Platelet monoamine oxidase activity and psychopathy. Psychiatry Research 16, 339–343. Mason, J.W., 1968. A review of psychoendocrine research on the pituitary-adrenal cortical system. Psychosomatic Medicine 30, 576–606. Mason, J., Southwick, S., Yehuda, R., Wang, S., Riney, S., Bremner, D., Johnson, D., Lubin, H., Blake, D., Zhou, G., Gusman, F., Charney, D., 1994. Elevation of serum free triiodothyronine, total triiodothyronine, thyroxine-binding globulin, and total thyroxine levels in combat-related posttraumatic stress disorder. Archives of General Psychiatry 51, 629–641.

Mason, J., Weizman, R., Laor, N., Wang, S., Schujovitsky, A., Abramovitz-Schneider, P., Feiler, D., Charney, D., 1996. Serum triiodothyronine elevation with posttraumatic stress disorder: a cross-cultural study. Biological Psychiatry 39, 835–838. National Board of Health and Welfare, 2002. Mentally Disordered Offenders. Report. Stockholm, Sweden. Nurnberger, J.I., 1992. Should a biologic marker be sensitive and specific? Acta Psychiatrica Scandinavica 86, 1–4. Oreland, L., Shaskan, E.G., 1983. Monoamine oxidase activity as a biological marker. Trends in Pharmacological Sciences, Elsevier August, 339–341. Oreland, L., Hallman, J., 1995. The correlation between platelet MAO activity and personality—a short review of findings and a discussion on possible mechanisms. In: Yu, P.U., et al. (Eds.), Progress in Brain Research, vol 106. Current Neurochemical and Pharmacological Aspects of Biogenic Amines: Their Function, Oxidative Deamination and Inhibition. Elsevier, New York, pp. 77–84. Puchall, L.B., Coursey, R.D., Buchsbaum, M.S., Murphy, D.L., 1980. Parents of high risk subjects defined by levels of monoamine oxidase activity. Schizophrenia Bulletin 6, 338–346. Ramklint, M., Stalenheim, E., von Knorring, A.-L., vonKnorring, L., 2000. Triiodothyronine (T3) related to conduct disorder in a forensic psychiatric population. European Journal of Psychiatry 14, 33–41. ˚ Risby, E.D., Hsiao, J.K., Sunderland, T., Agren, H., Rudorfer, M.V., Potter, W.Z., 1987. The effects of antidepressants on the cerebrospinal fluid homovanillic acidy5-hydroxyindoleacetic acid ratio. Clinical Pharmacology and Therapeutics 42, 547–554. Schalling, D., 1978. Psychopathy-related personality variables and the psychophysiology of socialization. In: Hare, R.D., Schalling, D. (Eds.), Psychopathic Behaviour: Approaches to Research. Wiley, New York, pp. 85–106. ˚ Schalling, D., Asberg, M., Edman, G., Oreland, L., 1987. Markers for vulnerability to psychopathology: temperament traits associated with platelet MAO activity. Acta Psychiatrica Scandinavica 76, 172–182. Schalling, D., Edman, G., 1993. The Karolinska Scales of Personality (KSP). An Inventory for Assessing Temperament Dimensions Associated with Vulnerability for Psychosocial Deviance. The Department of Psychiatry, The Karolinska Institute, Stockholm Manual. Soderstrom, H., Blennow, K., Manhem, A., Forsman, A., 2001. CSF studies in violent offenders. I. 5-HIAA as a negative and HVA as a positive predictor of psychopathy. Journal of Neural Transmission 108, 869–878. Spitzer, R.L., Williams, J.B.W., Gibbon, M., 1987. Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II). New York State Psychiatric Institute, New York. Spitzer, R.L., Williams, J.B.W., Gibbon, M., First, M.B., 1992. The Structured Clinical Interview for DSM-III-R (SCID). I: History, rationale, and description. Archives of General Psychiatry 49, 624–629.

E. Gunilla Stalenheim / Psychiatry Research 121 (2004) 281–291 Stalenheim, E.G., 1997. Karolinska Scales of Personality in a forensic psychiatric population: stability over time and between situations. Nordic Journal of Psychiatry 51, 379–384. Stalenheim, E.G., 2001. Relationships between attempted suicide, temperamental vulnerability, and violent criminality in a Swedish forensic psychiatric population. European Psychiatry 16, 386–394. Stalenheim, E.G., von Knorring, L., 1996. Psychopathy and Axis I and Axis II disorders in a forensic psychiatric population in Sweden. Acta Psychiatrica Scandinavica 94, 217–223. Stalenheim, E.G., von Knorring, L., 1998. Personality traits and psychopathy in a forensic psychiatric population. European Journal of Psychiatry 12, 83–92. Stalenheim, E.G., von Knorring, L., Wide, L., 1996. Serum levels of thyroid hormones as biological markers in a Swedish forensic psychiatric population. Biological Psychiatry 43, 755–761. Stalenheim, E.G., von Knorring, L., Oreland, L., 1997. Platelet monoamine oxidase activity as a biological marker in a Swedish forensic psychiatric population. Psychiatry Research 69, 79–87.

291

Stalenheim, E.G., Eriksson, E., von Knorring, L., Wide, L., 1998. Testosterone as a biological marker in psychopathy and alcoholism. Psychiatry Research 77, 79–88. Wang, S., Mason, J., Southwick, S., Johnson, D., Lubin, H., Charney, D., 1995. Relationships between thyroid hormones and symptoms in combat-related posttraumatic stress disorder. Psychosomatic Medicine 57, 398–402. Wang, S., Mason, J., Charney, D., Yehuda, R., Riney, S., Southwick, S., 1997. Relationships between hormonal profile and novelty seeking in combat-related posttraumatic stress disorder. Biological Psychiatry 41, 1451. Wang, S., Mason, J., 1999. Elevations of serum T3 levels and their association with symptoms in World War II veterans with combat-related posttraumatic stress disorder: replication of findings in Vietnam combat veterans. Psychosomatic Medicine 61, 131–138. von Knorring, L., Oreland, L., Winblad, B., 1984. Personality traits related to monoamine oxidase (MAO) in platelets. Psychiatry Research 12, 11–26. von Knorring, A.-L., Hallman, J., von Knorring, L., Oreland, L., 1991. Platelet monoamine oxidase activity in type 1 and type 2 alcoholism. Alcohol and Alcoholism 26, 409–416. Zuckerman, M., 1984. Sensation seeking: a comparative approach to a human trait. The Behavioral and Brain Sciences 7, 413–471.