|Home | About | Journals | Submit | Contact Us | Français|
The best known neurobehavioral effects of testosterone are on sexual function and aggression. However, testosterone and other androgens may be involved in the pathophysiology of mood disorders and suicidal behavior. This is the first study to examine whether there is a relation between testosterone levels and clinical parameters in bipolar suicide attempters.
Patients with a DSM-IV diagnosis of a bipolar disorder (16 males and 51 females), in a depressive or mixed episode with at least one past suicide attempt were enrolled. Demographic and clinical parameters, including lifetime suicidal behavior, were assessed and recorded. Plasma testosterone was assayed using a double antibody radioimmunoassay procedure.
The number of major depressive episodes, the maximum lethality of suicide attempts, and the testosterone levels were higher in men compared to women. Current suicidal ideation scores were higher in women compared to men. Controlling for sex, we found that testosterone levels positively correlated with the number of manic episodes and the number of suicide attempts.
Our findings are consistent with previous observations of the association between testosterone levels and parameters of mood and behavior. This study suggests that testosterone levels may be related to the course of bipolar disorder and suicidal behavior. Further studies of the role of testosterone in the neurobiology of mood disorders and suicidal behavior are merited.
In addition to its gonadal functions, testosterone has many central nervous system effects (Rubinow and Schmidt, 1996; Schmidt and Rubinow, 1997; Zarrouf et al., 2009; Ebinger et al., 2009). Testosterone can influence neuronal function through binding to intracellular receptors, through modulation of ligand-gated ion channels, and through binding to neurotransmitter receptors. It is also involved in modeling the developing brain and influences the continuous process of neuronal adaptation to new environmental demands.
The best known neurobehavioural effects of testosterone are on sexual function and aggression (Rubinow and Schmidt, 1996; Ebinger et al., 2009). However, testosterone and other androgens might be involved in the pathophysiology of mood disorders and suicidal behavior.
Depressed men have lower plasma or serum testosterone levels (Barrett-Connor et al., 1999; Schweiger et al., 1999; Almeida et al., 2004; Ebinger et al., 2009) although this association is not observed consistently (Levitt and Joffe, 1988; O’Connor et al., 2004; Ebinger et al., 2009). Hypogonadal men frequently show depressive symptoms, and testosterone replacement may improve these symptoms (Wang et al., 1996; Pope et al., 2003). Testosterone and other androgens might have antidepressant properties (Altschule and Tillotson, 1948; Itil et al., 1984; Pope et al., 2003; Amiaz and Seidman, 2008; Zarrouf et al., 2009). For example, it has been shown that testosterone gel may produce antidepressant effects in depressed men with low testosterone levels (Pope et al., 2003).
The data on the relationship of testosterone to suicidal behavior have been less consistent. Some studies reported low plasma testosterone levels after suicide attempts (Tripodianakis et al., 2007; Markianos et al., 2009). However, a study of male veterans with posttraumatic stress disorder showed that in this patient population, serum levels of testosterone were not associated with a history of suicide attempt (Butterfield et al., 2005). A recent study reported that there was no difference in testosterone levels between male suicide attempters and healthy controls (Perez-Rodriguez et al., 2011).
We have performed an exploratory study to examine whether there is a relationship between testosterone levels and suicidal behavior and related clinical parameters in bipolar disorder.
Participants were recruited through a combination of emergency department referrals, referrals from other outpatient services, and self-referral in response to advertisements. All participants provided written informed consent as approved by the New York State Psychiatric Institute Institutional Review Board. To be included, patients had to have a DSM-IV diagnosis of a bipolar disorder based on the Structured Clinical Interview for DSM-IV; be in a depressive or mixed episode; have at least one past suicide attempt; and be 18 to 75 years of age. Exclusion criteria were lack of capacity to provide informed consent; pregnancy or lactation; active medical problems, including substance abuse problems requiring detoxification. Forty five subjects had bipolar I disorder, and 22 subjects suffered from bipolar II disorder. Ten female participants were taking oral contraceptives.
Axis I and II pathology was assessed at baseline with the Structured Clinical Interview for DSM-IV Axis I and II Disorders. Current severity of depression was assessed by the Hamilton Depression Rating Scale (HDRS) (Hamilton, 1960). Lifetime aggression and impulsivity were assessed with the Aggression History Scale (Brown-Goodwin, revised; Brown and Goodwin, 1986) and the Barratt Impulsivity Scale, respectively (Barratt, 1965). Hostility (lifetime) was rated with the Buss-Durkee Hostility Inventory (Buss and Durkee, 1957). The Scale for Suicide Ideation (SSI) was used to measure the severity of suicidal ideation (Beck et al., 1979), and hopelessness during the previous week was measured with the Beck Hopelessness Scale (BHS) (Beck et al., 1974). A lifetime history of all suicide attempts, including number of attempts and the method of the attempt, was recorded on the Columbia Suicide History Form (Oquendo et al., 2003). A suicide attempt was defined as a self-destructive act that was committed with some intent to end one’s life. The Medical Lethality Rating Scale was used to measure the degree of medical damage caused by each suicide attempt (Beck et al., 1975). The scale was scored from 0 to 8 (0=no medical damage, 8=death), with anchor points for different suicide attempt methods. The degree of suicide intent for the worst attempt was rated with the Suicide Intent Scale (Beck et al., 1974). Interviewers were Masters or PhD-level psychologists. Inter-rater reliability was good to excellent (ICC 0.71 – 0.97).
Plasma testosterone was assayed using a double antibody 125I radioimmunoassay procedure (MP BIOMEDICALS Costa Mesa, California). All assays were run in duplicate with an eight point standard curve encompassing 0.1–10ng/ml. Three quality control low, medium and high levels were run with each batch assay. Inter and intra assay coefficient of variation were < 8% and <10% respectively.
Demographic and clinical features of male and female patients were compared using t-test (for continuous variables) and chi-square test (for categorical variables). Differences in testosterone levels between smokers and non-smokers were tested controlling for sex and using linear regression analysis. Pearson’s correlations were used to test the relation between testosterone levels and clinical parameters. Partial correlation analysis was employed to control for sex. The SPSS 19 statistical program was used to perform statistical analyses.
Demographic and clinical characteristics of study participants are presented in Table 1. Sixteen men and 51 women were recruited into the study. As expected, testosterone levels were significantly higher in males compared with females. The number of major depressive episodes and the maximum lethality of suicide attempts were significantly higher in men compared to women, yet current suicide ideation scores were significantly higher in women compared to men. There were no other sex-related differences in the sample.
Controlling for sex, we found that testosterone levels positively correlated with the number of manic episodes and the number of suicide attempts (Table 2). We have not observed a correlation between testosterone levels and aggression scale scores.
The higher number of major depressive episodes in male participants is likely to be a sampling issue. In fact, most studies of patients with bipolar disorder found no consistent sex differences in number of depressive episodes across sexes (Baldassano et al., 2005; Diflorio and Jones, 2010; Grant et al., 2005; Hendrick et al., 2000; Kessing, 2004; Kawa et al., 2005; Robb et al., 1998). To the contrary, several reports suggest that bipolar women are more likely than men to show a predominance of depressive polarity (Angst, 1978; Roy-Byrne et al., 1985; Nivoli et al., 2011). However, it is possible that this observation regarding polarity does not apply to bipolar suicide attempters who represent a very specific patient population. It is worth noting that there is a trend towards older age among male participants in the study. This difference, however, is not statistically or clinically significant and does not explain the significantly higher number of major depressive episodes in males compared to females in our study (Table 1).
In our study, male subjects made more medically damaging suicide attempts than females, which is consistent with a recent report that suicide attempts are more often violent amongst bipolar men than among women with bipolar illness (Nivoli et al., 2011). This finding is also consistent with observations that males take their own lives at nearly four times the rate of females and comprise approximately 80% of all suicides, even though female suicide attempt rates are estimated to be three to four times higher than men’s (Moscicki et al., 1994; Cannetto and Lester, 1995; Callanan and Davis, in press). One of the primary reasons given for the substantial sex gap in suicide rates is the difference in suicide methods used by males and females. In general, men are more likely to use methods that ensure high lethality than are women, which is consistent with our observation that the maximum lethality of suicide attempts was significantly higher in men compared to women in our sample.
As expected, testosterone levels were significantly higher in males than in females. The mean testosterone level in males in our sample was nine times higher than females, a ratio consistent with the literature (Mooradian et al., 1987; Molina, 2006).
We have observed that testosterone levels correlate with the number of manic episodes. This has never been reported on before. This finding may be related to the observation that anabolic steroid users, who consume high doses of testosterone and related androgens, sometimes develop manic or hypomanic symptoms during androgen use and depressive symptoms during androgen withdrawal (Pope and Katz, 1988; Pope and Katz, 1994). A study of the effects of supraphysiologic doses of testosterone on mood in healthy men aged 20 to 50 years showed that 16% of study participants became hypomanic (Pope et al., 2000). Two other studies of the effects of testosterone on healthy men showed that some study subjects developed symptoms of mania (Su et al., 1993; Yates et al., 1999). A case of mania in a patient taking testosterone as replacement therapy following bilateral orchidectomy, a selective serotonin reuptake inhibitor and St John’s Wort has been described (Barbenel et al., 2004). A case of testosterone patch-induced mania in a patient with a history of bipolar II disorder has also been reported (Weiss et al., 1999). These observations suggest that there is a relation between testosterone blood levels and mania.
We found that the number of suicide attempts in bipolar suicide attempters correlate with testosterone levels. Some (Tripodianakis et al., 2007; Markianos et al., 2009) but not all (Butterfield et al., 2005; Perez-Rodriguez et al., 2011) studies find relationships between testosterone and suicidal behavior.
Tripodianakis et al. (2007) compared testosterone levels in males admitted to hospital wards after a suicide attempt with testosterone levels in healthy males in the same age range. Blood samples from suicide attempters were drawn during the first days of hospitalization. The authors found that attempters had significantly lower testosterone levels compared to control, and that the attempters who used violent methods had lower testosterone levels compared to the non-violent attempter subgroup. Markianos et al. (2009) examined testosterone levels in a group of male psychiatric patients who had attempted suicide by jumping, compared with a group of male subjects who were hospitalized after accidentally falling from a high height, and in healthy controls. Both accident and suicide attempt groups had lower testosterone levels compared to the control group, and there was a trend towards lower testosterone levels in suicide attempters compared to the accident group. However, a recent study found no difference with regard to plasma testosterone levels between male suicide attempters and male controls (Perez-Rodriguez et al., 2011). In that study, blood samples were taken within 24 hours after the attempt. A study which investigated associations between neuroactive steroids and suicidality in military veterans with posttraumatic stress disorder found no association between testosterone levels and a history of a suicide attempt in the last six months (Butterfield et al., 2005). The discrepancies between the results of the studies of the relation between testosterone and suicidal behavior could be related to variations in sample selection criteria, psychiatric diagnoses, differences in time intervals between suicide attempts and blood sampling, smoking status and potentially other factors. Also, the control and patient groups were often poorly matched. No prior studies have looked at the relation between testosterone levels and the number of suicide attempts.
A significant number of studies suggest that testosterone is associated with aggression. It has been shown that violent individuals have higher plasma (Ehrenkranz et al., 1974; Mattsson et al., 1980), saliva (Soler et al., 2000) and CSF testosterone (Virkkunen et al., 1994) levels compared to non-violent controls. For example, in a study of impulsive offenders with alcoholism and antisocial personality disorder, higher levels of CSF testosterone were observed compared to healthy volunteers (Virkkunen et al., 1994). The authors suggested that high CSF testosterone levels may be associated with aggressiveness or interpersonal violence. In the same paper, the authors reviewed the literature on the relation of testosterone to aggression in humans, and proposed that both a repetitive pattern of aggressive behavior starting early in life, and a repetitive pattern of aggressiveness under the influence of alcohol are associated with elevated levels of testosterone. It has also been demonstrated that adolescent males with high plasma testosterone levels are more irritable and more likely to respond aggressively to provocation and threats than subjects with lower testosterone concentrations (Olweus et al., 1980; Olweus et al., 1988). Also, individuals receiving testosterone are more likely to have an aggressive response to perceived threats than subjects receiving placebo (Pope and Katz, 1990; Su et al., 1993; Pope et al., 2000).
Some authors have postulated that there are substantial similarities between aggression against the self and aggression against others, based on the clinical and epidemiological findings that some suicide attempters may share personality traits with violent criminals (Engstrom et al., 1999). We have also observed an association between aggression and suicidal behavior (Oquendo et al., 2004; Sher et al., 2005). For example, we have demonstrated that high aggression predicts suicidal acts (Oquendo et al., 2004). We have also shown that the higher prevalence of suicide attempters among depressed patients with a history of alcoholism compared to depressed patients without a history of alcoholism was related to higher aggression scores in the group with alcoholism (Sher et al., 2005).
Some studies, however, did not find an association between testosterone and aggression (Archer et al., 1991; Zitzmann and Nieschlag, 2001) which is consistent with our observation in this study that testosterone levels do not correlate with aggression scores. It has been suggested that the connection between testosterone and aggression is probably only of importance to athletes who supplement their testosterone levels to very high levels (Zitzmann, 2006).
A strength of this study is that we have examined the relation between testosterone levels and clinical features in an unique patient population: suicide attempters with bipolar disorder. This is the first study of testosterone levels in this patient population. Another strength of this investigation is that the sample was very well characterized using multiple psychometric scales. A limitation of the study is that we did not have a bipolar nonattempter or healthy volunteer control group for comparison and the sample size is modest. In addition, due to the exploratory nature of the study, results should be interpreted with caution. Overall, the study design is not adequate to definitely conclude that testosterone is a correlate of suicidal behavior.
This study suggests that testosterone levels may be related to the course of bipolar disorders and suicidal behavior. It is possible that a) testosterone levels influence the pathophysiology of bipolar illness and suicidal behavior; b) the presence of bipolar disorder and suicidal behavior affects testosterone levels; c) certain factors (e.g., genetics or smoking) affect both testosterone levels and the course of bipolar disorder and suicidal behavior; and d) there is a complex interplay of different factors which is most likely. Future research should focus on the role of testosterone in the pathophysiology of depression not only suicidal behavior taking into account that depressive disorders are associated with suicidal behavior. Whether testosterone has antidepressant properties thereby reducing suicidal behavior in patients with depression is unknown. It may also be important to examine the role of testosterone in the pathophysiology of manic states.
Our findings are consistent with some previous observations of the association between testosterone levels and parameters of mood and behavior. This study suggests that testosterone levels may be related to the course of bipolar disorders and suicidal behavior. Further studies of the role of testosterone in the neurobiology of mood disorders and suicidal behavior are merited.
Supported by grants R01 MH59710, P50 MH62185, and R01 MH48514.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.