129
Psychiafry Research, 39:129-139 Elsevier
to Intravenous Clonidine Growth Hormone Responses Behavioral Irritability in Challenge Correlate With Psychiatric Patients and Healthy Volunteers Emil F. Coccaro, Timothy Lawrence, Howard M. Klar, and Larry J. Siever Received 1991.
Robert
March 26, 1991; revised version received August
Trestman,
Steven
12, 1991; accepted
Gabriel,
September
1,
Abstract. To explore the relationship between central noradrenergic receptor responsivity and indices of impulsive aggression, growth hormone responses to infusions with the cr,-adrenergic receptor agonist clonidine (GH[CLON]) and responses on the Buss-Durkee Hostility Inventory (BDHI) were examined in healthy male volunteers and male patients with major affective or personality disorder. GH[CLON] values were found to correlate significantly with the BDHI “Irritability” subscale in all subjects, but especially in healthy volunteer and personality disorder patients. GH[CLON] values did not correlate with the BDHI “Assault” subscale. These results suggest a role for central a,-adrenergic receptor responsivity in the personality trait characterized by behavioral irritability, but not overt assaultiveness, in humans. Key Words. Affective disorders, Inventory, cY,-adrenergic receptor,
personality disorders, norepinephrine.
Buss-Durkee
Hostility
Abnormalities in various indices of central neurotransmitter system function suggest an important role for biological systems in the regulation of impulsive aggressive behaviors in humans (Eichelman, 1987). For example, indices of central serotonergic system function correlate inversely with indices of suicidal and impulsive aggressive behavior (Coccaro, 1989). Since serotonin (5hydroxytryptamine; 5HT) serves an inhibitory function in many neuronal systems (Vogt, 1982), these data are consistent with the notion that 5HT inhibits aggressive behavior in response to an adverse stimulus (Coccaro, 1989). In addition to 5HT, the central noradrenergic system may also influence impulsive aggressive behavior in mammalian species. Unlike 5HT, however, norepinephrine may promote, rather than inhibit, impulsive aggressive behavior in these species.
F. Coccaro, M.D., is Associate Professor and Director, Clinical Neuroscience Research Unit, Department of Psychiatry, Medical College of Pennsylvania at Eastern Pennsylvania Psychiatric Institute, Philadelphia, PA. Timothy Lawrence, M.D., is Staff Psychiatrist, Four Winds Hospital, NY. Robert Trestman, Ph.D., M.D., is Assistant Professor of Psychiatry; Steven Gabriel, Ph.D., is Assistant Research Emil
Professor; and Larry J. Siever, M.D., is Associate Professor of Psychiatry, Department of Psychiatry, Mt. Sinai School of Medicine, New York, NY. Howard M. Klar, M.D., is Associate Clinical Professor, Department of Psychiatry, St. Luke’s_Roosevelt Medical Center, New York, NY. (Reprint requests to Dr. E.F. Coccaro, Clinical Neuroscience Research Unit, Dept. of Psychiatry, MCP at EPPI, 3200 Henry Ave., Philadelphia, PA 19129, USA.) 0161781/91/$03.50
0 1991 Elsevier Scientific
Publishers
Ireland
Ltd.
130 Indices of increased noradrenergic (NE) function in brain correlate directly with the number of shock-induced aggressive episodes in rodents (Stolk et al., 1974). In addition, agents that enhance central NE function (i.e., tricyclic/ monoamine oxidase inhibitor antidepressants) have been shown to increase shock-induced fighting in rodents (Eichelman and Barchas, 1975). Similarly, clinical treatment with these agents has been associated with agitation and behavioral “irritability’‘-particularly in patients who tend to be impulsive aggressive. For example, a subset of patients with DSM-III borderline personality disorder have been reported to have increased agitation or behavior dyscontrol during treatment with tricyclici monoamine oxidase inhibitor antidepressant agents (Soloff et al., 1986; Cowdry and Gardner, 1988). Conversely, agents that diminish noradrenergic function, such as propranolol, appear to have antiaggressive properties (Silver and Yudofsky, 1986). Finally, concentrations of lumbar cerebrospinal fluid (CSF) 3-methoxy-4-hydroxyphenylglycol (MHPG) have been found to correlate directly with a history of impulsive aggressive (Brown et al., 1979) or extraverted (Roy et al., 1989) behavior in some, though not all (Linnoila et al., 1983), studies. Overall, these data are consistent with animal data which demonstrate an important role for central noradrenergic system function in regulating the level of arousal and the responsiveness of the organism to environmental stimuli (Aston-Jones and Bloom, 1981; Foote et al., 1983; Rasmussen et al., 1986a, 19866). Accordingly, increased noradrenergic system function would be associated with increased arousal and readiness to respond to environmental cues, be they adversive or not. To explore more fully the noradrenergic correlates of impulsive aggressive behaviors, we examined the growth hormone (GH) response to intravenous clonidine challenge in healthy volunteers, and in patients with personality disorder, remitted major affective disorder, or both. Clonidine is an cy,-noradrenergic receptor agonist (Kobinger and Pichler, 1976) capable of stimulating central NE receptors and assessing their physiological sensitivity to endogenous norepinephrine (NE). In animals (Rudolph et al., 1980) and humans (La1 et al., 1975) an acute challenge of clonidine yields a robust and reliable GH response. In animals, this response is blocked by the a,-antagonist yohimbine but not by the a,-antagonist prazosin (Eriksson et al., 1982; Kruhlich et al., 1982; McWilliam and Meldrum, 1983). Further evidence that the GH response to clonidine challenge reflects the sensitivity of the postsynaptic (r2-NE receptor comes from studies in which presynaptic NE function is experimentally diminished or enhanced. Depletion or destruction of presynaptic NE neurons is associated with an augmentation of the GH response to clonidine (Eriksson et al., 1982; Siever et al., 1987). Further, enhancement of noradrenergic availability, by NE uptake inhibitors, is associated with a diminution of the GH response to clonidine (Glass et al., 1982). Hence, the GH response to clonidine challenge appears to reflect the activation of the postsynaptic crZ-NE receptor and thus can be used as an index of the sensitivity of a,-adrenergic receptors in the limbic-hypothalamic-pituitary axis in humans (Siever et al., 198 1). For these reasons, we hypothesized that GH responses to clonidine challenge would correlate directly with indices of impulsive aggressive behavior in human subjects.
131
Methods Sample. Data were collected during a study of the monoaminergic correlates of behavior in healthy male volunteers and in physically healthy male patients with personality disorder, major affective disorder, or both conditions conducted at the Bronx VA Medical Center of the Mt. Sinai School of Medicine (Coccaro et al., 1989a, 19896, 1990; Siever et al., in press). Axis I and II diagnoses were made according to D&V-III criteria (American Psychiatric Association, 1980). Interview instruments for the diagnostic assessment of the patients included the Schedule for Schizophrenia and Affective Disorders-Lifetime Version (SADS-L; Spitzer and Endicott, 1978) and the Structured Interview for Diagnosing Personality Disorders (SIDP; Stangl et al., 1985) as described in detail elsewhere (Coccaro et al., 1989b). Final diagnosis was made by consensus, on the basis of all available data, in a meeting chaired by a senior clinician (H.M.K.) without knowledge of the biological data base. Healthy male volunteers were determined to be free of a lifetime history of any Axis I or Axis II disorder by a research psychiatrist (E. F.C.) who used a DSM-III checklist during a semistructured interview. The study group was composed of 10 male patients with DSM-III major affectivedisorder in remission (MAD; mean age = 54.5 years, SD q 7.3), 8 male patients with DSM-III personality disorder (PD; mean age = 35.9 years, SD = 6.9 years), and 10 healthy male volunteer subjects (mean age q 53.9 years, SD = 14.7). Remitted MAD patients were determined to have had at least one episode of major depression with significant impairment or a successful response to antidepressant medication by clinical history as determined by the SADS-L. These patients had not been hospitalized for at least 6 months and no longer required continued treatment with psychotropic medication. Scores on the 23-item Hamilton Rating Scale for Depression (Guy, 1976) for the remitted MAD patients at time of study were < 8 (range = O-8; median = 2). The MAD patients were drawn from a larger group of MAD patients reported elsewhere (Siever et al., in press). While several of the PD patients had a history of at least one episode of major depression, they differed from the MAD patients in that the personality disorder (1) always preceded the onset of the first (or only) episode of major depression and (2) was always the primary cause of functional morbidity. Three PD patients received a diagnosis of mixed personality disorder; three other patients received more than one Axis II diagnosis (avoidantschizotypal; borderline-histrionic-paranoid; and avoidant-borderline-compulsive-paranoidschizotypal); the remaining two patients received a diagnosis of passive-aggressive personality disorder. Intravenous Clonidine Challenge. All subjects were medicially healthy as determined by a comprehensive medical evaluation including medical history, physical examination, and laboratory examination. Patients had not received psychotropic agents for at least 14 days (range q 14-90 days; median q 30 days); healthy volunteers denied any history of treatment with psychotropic agents. In addition, all subjects were free of all drugs (including nonpsychotropic agents) for at least 14 days before study. In addition, each subject was carefully instructed to maintain a low monoamine diet for at least 3 days before the procedure. Challenge protocols began at 8 a.m. after an overnight fast and abstention from smoking (if applicable) since midnight. At 8 a.m., an intravenous catheter was placed in a forearm vein and was kept open with normal saline. In addition, subjects were not permitted to eat, smoke, or sleep during the protocol. Baseline samples for plasma GH were obtained at 9:45 and 9:55 a.m. Clonidine (2 pg/ kg) was administered by a slow infusion over 5 minutes beginning at 10 a.m. Post-clonidine samples for plasma GH were then obtained every 15 minutes until 11 a.m. Peak A GH (i.e., peak GH after clonidine minus the averaged baseline GH), hereafter referred to as GH[CLON]), was used as the main outcome index of GH response to clonidine challenge. GH was assayed by radioimmunoassay (Davis et al., 1983); intra-assay and interassay coefficients of variation were 3% and 8%, respectively. Assessment of Impulsive Aggressive Behaviors in Subjects. The Buss-Durkee
Hostility
132 Inventory (BDHI; Buss and Durkee, 1957) constituted the primary baseline assessment of impulsive aggression and was administered on a different day from the clonidine challenge. The BDHI, a 75-item true-false questionnaire related to aggression (e.g., irritability, verbal aggression, indirect aggression, and physical assault) and hostility (e.g., resentment and suspiciousness), was rated by the subjects. The BDHI scales Irritability and Assault have been found to correlate with indices of central 5HT function in patients with DSM-III personality disorder: BDHI Irritability, with concentrations of lumbar CSF 5-hydroxyindoleacetic acid (Brownand Goodwin, 1984) and with the response of prolactin to the central 5HT releasing/uptake inhibiting agent fenfluramine (Coccaro et al., 19896) the postsynaptic 5HT agonist metachlorophenylpiperazine (m-CPP; Coccaro et al., 1989a), and the postsynaptic SHT-la agonist buspirone (Coccaro et al., 1990); BDHI Assault, with the prolactin response to fenfluramine (Coccaro et al., 19896) and to m-CPP (Coccaro et al., 1989~; Moss et al., 1990). In addition, scores on both these BDHI scales have been found to correlate highly with history of actual impulsive aggressive behavior (Irritability: r q 0.58, p < 0.01; Assault: r= 0.74,~ HV group (analysis variance: F = 7.67; df = 2, 25; p = 0.003; vs. PD: p < 0.003; vs. MAD: p < 0.0031.
of
Fig. 2. Correlation between log peak growth hormone response to i.v. clonidine challenge (GH[CLON]) and Irritability and Assault scores 16 0
r - 0.71 n 918 p
Psychiafry Research, 39:129-139 Elsevier
to Intravenous Clonidine Growth Hormone Responses Behavioral Irritability in Challenge Correlate With Psychiatric Patients and Healthy Volunteers Emil F. Coccaro, Timothy Lawrence, Howard M. Klar, and Larry J. Siever Received 1991.
Robert
March 26, 1991; revised version received August
Trestman,
Steven
12, 1991; accepted
Gabriel,
September
1,
Abstract. To explore the relationship between central noradrenergic receptor responsivity and indices of impulsive aggression, growth hormone responses to infusions with the cr,-adrenergic receptor agonist clonidine (GH[CLON]) and responses on the Buss-Durkee Hostility Inventory (BDHI) were examined in healthy male volunteers and male patients with major affective or personality disorder. GH[CLON] values were found to correlate significantly with the BDHI “Irritability” subscale in all subjects, but especially in healthy volunteer and personality disorder patients. GH[CLON] values did not correlate with the BDHI “Assault” subscale. These results suggest a role for central a,-adrenergic receptor responsivity in the personality trait characterized by behavioral irritability, but not overt assaultiveness, in humans. Key Words. Affective disorders, Inventory, cY,-adrenergic receptor,
personality disorders, norepinephrine.
Buss-Durkee
Hostility
Abnormalities in various indices of central neurotransmitter system function suggest an important role for biological systems in the regulation of impulsive aggressive behaviors in humans (Eichelman, 1987). For example, indices of central serotonergic system function correlate inversely with indices of suicidal and impulsive aggressive behavior (Coccaro, 1989). Since serotonin (5hydroxytryptamine; 5HT) serves an inhibitory function in many neuronal systems (Vogt, 1982), these data are consistent with the notion that 5HT inhibits aggressive behavior in response to an adverse stimulus (Coccaro, 1989). In addition to 5HT, the central noradrenergic system may also influence impulsive aggressive behavior in mammalian species. Unlike 5HT, however, norepinephrine may promote, rather than inhibit, impulsive aggressive behavior in these species.
F. Coccaro, M.D., is Associate Professor and Director, Clinical Neuroscience Research Unit, Department of Psychiatry, Medical College of Pennsylvania at Eastern Pennsylvania Psychiatric Institute, Philadelphia, PA. Timothy Lawrence, M.D., is Staff Psychiatrist, Four Winds Hospital, NY. Robert Trestman, Ph.D., M.D., is Assistant Professor of Psychiatry; Steven Gabriel, Ph.D., is Assistant Research Emil
Professor; and Larry J. Siever, M.D., is Associate Professor of Psychiatry, Department of Psychiatry, Mt. Sinai School of Medicine, New York, NY. Howard M. Klar, M.D., is Associate Clinical Professor, Department of Psychiatry, St. Luke’s_Roosevelt Medical Center, New York, NY. (Reprint requests to Dr. E.F. Coccaro, Clinical Neuroscience Research Unit, Dept. of Psychiatry, MCP at EPPI, 3200 Henry Ave., Philadelphia, PA 19129, USA.) 0161781/91/$03.50
0 1991 Elsevier Scientific
Publishers
Ireland
Ltd.
130 Indices of increased noradrenergic (NE) function in brain correlate directly with the number of shock-induced aggressive episodes in rodents (Stolk et al., 1974). In addition, agents that enhance central NE function (i.e., tricyclic/ monoamine oxidase inhibitor antidepressants) have been shown to increase shock-induced fighting in rodents (Eichelman and Barchas, 1975). Similarly, clinical treatment with these agents has been associated with agitation and behavioral “irritability’‘-particularly in patients who tend to be impulsive aggressive. For example, a subset of patients with DSM-III borderline personality disorder have been reported to have increased agitation or behavior dyscontrol during treatment with tricyclici monoamine oxidase inhibitor antidepressant agents (Soloff et al., 1986; Cowdry and Gardner, 1988). Conversely, agents that diminish noradrenergic function, such as propranolol, appear to have antiaggressive properties (Silver and Yudofsky, 1986). Finally, concentrations of lumbar cerebrospinal fluid (CSF) 3-methoxy-4-hydroxyphenylglycol (MHPG) have been found to correlate directly with a history of impulsive aggressive (Brown et al., 1979) or extraverted (Roy et al., 1989) behavior in some, though not all (Linnoila et al., 1983), studies. Overall, these data are consistent with animal data which demonstrate an important role for central noradrenergic system function in regulating the level of arousal and the responsiveness of the organism to environmental stimuli (Aston-Jones and Bloom, 1981; Foote et al., 1983; Rasmussen et al., 1986a, 19866). Accordingly, increased noradrenergic system function would be associated with increased arousal and readiness to respond to environmental cues, be they adversive or not. To explore more fully the noradrenergic correlates of impulsive aggressive behaviors, we examined the growth hormone (GH) response to intravenous clonidine challenge in healthy volunteers, and in patients with personality disorder, remitted major affective disorder, or both. Clonidine is an cy,-noradrenergic receptor agonist (Kobinger and Pichler, 1976) capable of stimulating central NE receptors and assessing their physiological sensitivity to endogenous norepinephrine (NE). In animals (Rudolph et al., 1980) and humans (La1 et al., 1975) an acute challenge of clonidine yields a robust and reliable GH response. In animals, this response is blocked by the a,-antagonist yohimbine but not by the a,-antagonist prazosin (Eriksson et al., 1982; Kruhlich et al., 1982; McWilliam and Meldrum, 1983). Further evidence that the GH response to clonidine challenge reflects the sensitivity of the postsynaptic (r2-NE receptor comes from studies in which presynaptic NE function is experimentally diminished or enhanced. Depletion or destruction of presynaptic NE neurons is associated with an augmentation of the GH response to clonidine (Eriksson et al., 1982; Siever et al., 1987). Further, enhancement of noradrenergic availability, by NE uptake inhibitors, is associated with a diminution of the GH response to clonidine (Glass et al., 1982). Hence, the GH response to clonidine challenge appears to reflect the activation of the postsynaptic crZ-NE receptor and thus can be used as an index of the sensitivity of a,-adrenergic receptors in the limbic-hypothalamic-pituitary axis in humans (Siever et al., 198 1). For these reasons, we hypothesized that GH responses to clonidine challenge would correlate directly with indices of impulsive aggressive behavior in human subjects.
131
Methods Sample. Data were collected during a study of the monoaminergic correlates of behavior in healthy male volunteers and in physically healthy male patients with personality disorder, major affective disorder, or both conditions conducted at the Bronx VA Medical Center of the Mt. Sinai School of Medicine (Coccaro et al., 1989a, 19896, 1990; Siever et al., in press). Axis I and II diagnoses were made according to D&V-III criteria (American Psychiatric Association, 1980). Interview instruments for the diagnostic assessment of the patients included the Schedule for Schizophrenia and Affective Disorders-Lifetime Version (SADS-L; Spitzer and Endicott, 1978) and the Structured Interview for Diagnosing Personality Disorders (SIDP; Stangl et al., 1985) as described in detail elsewhere (Coccaro et al., 1989b). Final diagnosis was made by consensus, on the basis of all available data, in a meeting chaired by a senior clinician (H.M.K.) without knowledge of the biological data base. Healthy male volunteers were determined to be free of a lifetime history of any Axis I or Axis II disorder by a research psychiatrist (E. F.C.) who used a DSM-III checklist during a semistructured interview. The study group was composed of 10 male patients with DSM-III major affectivedisorder in remission (MAD; mean age = 54.5 years, SD q 7.3), 8 male patients with DSM-III personality disorder (PD; mean age = 35.9 years, SD = 6.9 years), and 10 healthy male volunteer subjects (mean age q 53.9 years, SD = 14.7). Remitted MAD patients were determined to have had at least one episode of major depression with significant impairment or a successful response to antidepressant medication by clinical history as determined by the SADS-L. These patients had not been hospitalized for at least 6 months and no longer required continued treatment with psychotropic medication. Scores on the 23-item Hamilton Rating Scale for Depression (Guy, 1976) for the remitted MAD patients at time of study were < 8 (range = O-8; median = 2). The MAD patients were drawn from a larger group of MAD patients reported elsewhere (Siever et al., in press). While several of the PD patients had a history of at least one episode of major depression, they differed from the MAD patients in that the personality disorder (1) always preceded the onset of the first (or only) episode of major depression and (2) was always the primary cause of functional morbidity. Three PD patients received a diagnosis of mixed personality disorder; three other patients received more than one Axis II diagnosis (avoidantschizotypal; borderline-histrionic-paranoid; and avoidant-borderline-compulsive-paranoidschizotypal); the remaining two patients received a diagnosis of passive-aggressive personality disorder. Intravenous Clonidine Challenge. All subjects were medicially healthy as determined by a comprehensive medical evaluation including medical history, physical examination, and laboratory examination. Patients had not received psychotropic agents for at least 14 days (range q 14-90 days; median q 30 days); healthy volunteers denied any history of treatment with psychotropic agents. In addition, all subjects were free of all drugs (including nonpsychotropic agents) for at least 14 days before study. In addition, each subject was carefully instructed to maintain a low monoamine diet for at least 3 days before the procedure. Challenge protocols began at 8 a.m. after an overnight fast and abstention from smoking (if applicable) since midnight. At 8 a.m., an intravenous catheter was placed in a forearm vein and was kept open with normal saline. In addition, subjects were not permitted to eat, smoke, or sleep during the protocol. Baseline samples for plasma GH were obtained at 9:45 and 9:55 a.m. Clonidine (2 pg/ kg) was administered by a slow infusion over 5 minutes beginning at 10 a.m. Post-clonidine samples for plasma GH were then obtained every 15 minutes until 11 a.m. Peak A GH (i.e., peak GH after clonidine minus the averaged baseline GH), hereafter referred to as GH[CLON]), was used as the main outcome index of GH response to clonidine challenge. GH was assayed by radioimmunoassay (Davis et al., 1983); intra-assay and interassay coefficients of variation were 3% and 8%, respectively. Assessment of Impulsive Aggressive Behaviors in Subjects. The Buss-Durkee
Hostility
132 Inventory (BDHI; Buss and Durkee, 1957) constituted the primary baseline assessment of impulsive aggression and was administered on a different day from the clonidine challenge. The BDHI, a 75-item true-false questionnaire related to aggression (e.g., irritability, verbal aggression, indirect aggression, and physical assault) and hostility (e.g., resentment and suspiciousness), was rated by the subjects. The BDHI scales Irritability and Assault have been found to correlate with indices of central 5HT function in patients with DSM-III personality disorder: BDHI Irritability, with concentrations of lumbar CSF 5-hydroxyindoleacetic acid (Brownand Goodwin, 1984) and with the response of prolactin to the central 5HT releasing/uptake inhibiting agent fenfluramine (Coccaro et al., 19896) the postsynaptic 5HT agonist metachlorophenylpiperazine (m-CPP; Coccaro et al., 1989a), and the postsynaptic SHT-la agonist buspirone (Coccaro et al., 1990); BDHI Assault, with the prolactin response to fenfluramine (Coccaro et al., 19896) and to m-CPP (Coccaro et al., 1989~; Moss et al., 1990). In addition, scores on both these BDHI scales have been found to correlate highly with history of actual impulsive aggressive behavior (Irritability: r q 0.58, p < 0.01; Assault: r= 0.74,~ HV group (analysis variance: F = 7.67; df = 2, 25; p = 0.003; vs. PD: p < 0.003; vs. MAD: p < 0.0031.
of
Fig. 2. Correlation between log peak growth hormone response to i.v. clonidine challenge (GH[CLON]) and Irritability and Assault scores 16 0
r - 0.71 n 918 p
