Introduction
In the last few years a substantial body of scientific literature has highlighted common elements regarding clinical and neurobiological mechanisms of pathological gambling and substance use disorders (Fauth-Buhler et al.
2016; Leeman and Potenza
2012; Petry et al.
2013; Goudriaan et al.
2014). Indeed, activation of the dopaminergic mesolimbic system represents the neurobiological substrate of the reinforcing properties and of the enhanced incentive salience both of drugs of abuse and addictive behaviours like pathological gambling (Brancato et al.
2014; Berridge
2007). Thus, in the new edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) pathological gambling has been identified as a behavioural addictive disorder and moved to the category “Subtance-Related and Addictive Disorders” (American Psychiatric Association. Diagnostic and statistical manual of mental disorders—5th ed.
2013). In The Netherlands, pathological gambling prevalence ranges approximately from .25 to .75%, similar to rates in several other European countries and often is associated to comorbidity with other psychiatric disorders thus provoking negative psychosocial consequences and health costs (Cowlishaw et al.
2012; Griffiths
2009; Goudriaan
2007; Lejoyeux
2012; Maniaci et al.
2015; Odlaug et al.
2013).
From clinical and neuropsychological studies it becomes clear that impulsivity is an important factor that contributes to the onset and worsening of pathological gambling (Blanco et al.
1996; Steel and Blaszczynski
1998; Van Holst et al.
2010; Verdejo-Garcia et al.
2008). Impulsivity consists in the predisposition toward rapid, unplanned reactions to internal or external stimuli with no regard to their negative consequences, and the occurrence of impulsivity leads subjects to be highly responsive to immediate positive reinforcement but rather insensitive to long-term negative consequences (Moeller et al.
2001). Impulsivity is related to impaired decision-making and higher risk behaviour in pathological gamblers (PGs; Alessi and Petry
2003; Cavedini et al.
2002; Goudriaan et al.
2005; Petry and Casarella
1999; Zermatten et al.
2005). Compared to healthy controls (HCs), PGs display increased levels of impulsivity (Marazziti et al.
2014; for a review see Goudriaan et al.
2014) and, notably score significantly higher on impulsivity and on the inability to resist craving compared to alcoholics and cocaine users (Grant et al.
2016). Moreover the presence of high impulsivity levels is considered an important factor for treatment failure in pathological gambling (Leblond et al.
2003; Goudriaan et al.
2008). The physiological and systemic responses correlated with impulsivity include a variety of signs such as heart rate and electrodermal activity (EDA) (Derefinko et al.
2014; Mathias and Stanford
2003). For instance, high impulsive and low impulsive subjects show different patterns of psychophysiological reactivity in a gambling task during an active choice and high bet size; indeed differences in heart rate accelerations to wins versus losses are positively correlated with impulsivity levels in healthy students (Studer and Clark
2011; Studer et al.
2016). Furthermore higher impulsivity scores are associated with reduced electrodermal activity (EDA) differences between wins and losses and reduced EDA in response to stress during a risky choice task (Stankovic et al.
2014). However, how impulsivity is associated with stress responses in pathological gambling is not yet clear (Kräplin et al.
2014).
Several studies have observed increased physiological arousal in recreational gamblers while engaging in gambling related activities, such as heart rate and hypothalamic–pituitary–adrenal axis (HPA) activation (Coventry and Constable
1999; Kreuger et al.
2005; Wulfert et al.
2008). Gambling has been shown to lead to moderate heart rate elevation, and to increased levels of salivary cortisol (Meyer et al.
2000). Unexpectedly, problem gamblers did not show significant differences in plasma cortisol levels during casino gambling, compared to HCs (Meyer et al.
2004). Rather, salivary cortisol levels before and after watching gambling scenarios were significantly higher in recreational gamblers compared to PGs. These findings suggest that pathological gambling is associated with hyporeactivity of the stress system to cue-related stimuli (Paris et al.
2009). In this regard lower concentrations of salivary cortisol are associated with riskier choices and monetary losses in the Iowa Gambling Task, whereas the opposite pattern—higher concentrations of salivary cortisol—is associated with less risky choices and monetary gain (van Honk et al.
2003). However, whether lower HPA axis responses to cue-related stimuli are associated with a longer duration and intensity of pathological gambling is still not clear. In addition, it is unknown whether HPA-axis reactivity in PGs is diminished only in gambling situations, or whether a more general dysregulation of the HPA-axis reactivity exists. A dysfunctional arousal of the HPA axis could be related to a more severe form of pathological gambling or longer presence of pathological gambling, since diminished HPA-axis reactivity could maintain the addictive behaviour, because risky behavior or losses do not activate the HPA-axis enough, and thus reinforce the inclination towards risky behaviour or continuation of gambling, despite losses.
For these reasons the present study investigated HPA-axis reactivity to a psychosocial stress in PGs compared to HCs using the Trier Social Stress Test (TSST). This test has been shown to lead to a robust increase in cortisol through the activation of HPA axis and sympathetic nervous system (Brkic et al.
2015; Dickerson and Kemeny
2004; Inagaki and Eisemberger
2015; Kirschbaum and Hellhammer
1994). We hypothesised that relative to HCs, PGs would display a different physiological response to the stress test. Moreover considering that hypocortisolism may be a consequence of exposure to chronic stress (Heim et al.
2000), we also investigated the potential occurrence of a negative correlation between the intensity of the physiological stress response and duration of pathological gambling (Fernald et al.
2008). Because higher impulsivity levels have been correlated with increased sympathetic nervous system activity (Kreuger et al.
2005) and decreased cardiac vagal control in male adolescents (Allen et al.
2000) we wanted to investigate the interaction between impulsivity and stress reactivity in PGs, as higher impulsivity has been frequently reported in PGs. Impulsive psychological traits may correlate with the stress response of PGs during psychosocial stressing situations such as TSST, and therefore we investigated if there was a different pattern of correlation between impulsivity and stress response in PGs compared to HCs.
Discussion
This study investigated the effects of psychological stress on HPA axis activation and sympathetic nervous system response in relation to impulsivity in a sample of male PGs, compared to HCs. In particular, we measured salivary cortisol levels and IBI in PGs during and after the TSST compared to the HC response. The impact of the duration of the disorder and the role played by impulsivity in the onset of modifications in the physiological response to the stress test were also assessed.
Our study shows that the longer the duration of pathological gambling the lower the cortisol response before psychosocial stress, which is consistent with findings indicating that exposure to chronic stressful situations—similarly to gambling—can lead to hypocortisolism (Heim et al.
2000). However, there was no evidence of differences in IBI between groups. Several studies have highlighted an association between severity of psychiatric disorders and cortisol levels, such as in post-traumatic stress disorder (Yehuda et al.
1996), generalised anxiety disorder (Steudte et al.
2011) and pathological gambling (Geisel et al.
2015); consistently, our data report that a longer duration of pathological gambling is related to lower baseline salivary cortisol levels. This correlation can be interpreted as a maladaptive response of PGs’ HPA axis to the chronic distress caused by enduring gambling behaviour, thus confirming the occurrence of a dysregulation of critical central neuroendocrine circuits in pathological gambling. A theory suggests that under the influence of continuous or intermittent chronic stressful stimulus the hypocortisolism might be an adaptive self-preserving response, in order to protect the metabolic machinery, and most importantly, the brain (Hellhammer and Wade
1993). States of hypocortisolism are common in patients chronically exposed to stressful environments, in those with unpredictable schedules and in those with traumatic early life experiences (Gunnar and Vazquez
2001; Heim et al.
2000). As in several stress related disorders, also in pathological gambling the repeated or intermittent distress, caused by enduring gambling behaviour, can produce an alteration of basal HPA axis activity finally leading to the observed hypocortisolism. Notably, in PGs the baseline hypocortisol condition might promote increased vulnerability for the development of physical consequences, such as high-stress sensitivity, chronic fatigue and chronic pain (Fries et al.
2005), leading to functional and anatomical disturbances that need medical intervention. However when performing the TSST, PGs, independently from the duration of the disorder, display a remarkable activation of the HPA axis. Indeed there was a trend for higher cortisol levels at T4 and T5 in PGs compared to HCs. This suggests a vulnerability of the stress system of PGs to environmental and social challenges. Future longitudinal studies should further investigate this issue and clarify whether the correlation between hypocorticolism and duration of the disease observed in this study is a cause or a consequence of pathological gambling.
Although we replicated the robust effect of the TSST to produce an increase of salivary cortisol and IBI, we did not find significant differences between groups. This finding suggests that in PGs as in HCs, neuroendocrine and neurovegetative responses to non gambling-related stress stimuli are not compromised, at least in our experimental conditions, and for the examined parameters.
Interestingly, we show that impulsivity is positively correlated with IBI, over and above the influence of pathological gambling. Indeed, higher impulsivity subjects revealed significantly higher IBIs throughout the TSST. Thus impulsivity, independently of pathological gambling, appears to be related to a decreased heart rate during the TSST, consistent with an earlier study on the relationship between impulsivity and cardiovascular responses during a psychosocial stress task (Allen et al.
2009). Thus, our findings imply that impulsivity—regardless of the presence of pathological gambling—is associated with a diminished cardiovascular responsivity. Impulsivity, particularly during young adult life, has been implicated in an increase in the risk of unhealthy behaviours and developments due to structural brain deficits and lack of experience with novel adult behaviours (Romer
2010). This study goes some way to indicate a potential biological mechanism associated with impulsivity. Central motivational dysregulation has been proposed to underpin the link between blunted stress reactivity and outcomes such as obesity, depression, a range of substance and behavioural dependencies, and bulimia (Carroll et al.
2009,
2011; Lovallo
2011). Thus, blunted stress reactivity can be considered a peripheral marker of dysregulation of the neural systems that support motivation, emotional regulation, and goal-directed behaviour. The association between impulsivity and dysregulation of the neurovegetative stress response deserves special attention, because it represents a vulnerability factor that can affect impulsive decision making (Brand et al.
2005; Goudriaan et al.
2005), thus exacerbating all those neuropsychiatric disorders characterized by impaired impulse control.
The main limitation of the present study is the recruitment of only male participants. Further studies should be carried out in order to investigate the relation between impulsivity and physiological stress response in PGs including females. However, by including only males suffering from pathological gambling, whithout other comorbid disorders, we made sure that we had a quite homogeneous group, limiting the potentional confounding factors in our study.
In conclusion, an influence of pathological gambling on HPA-axis activity was highlighted together with a role of impulsivity in the cardiac stress response to a psychological stress test. Future studies should test whether the stress response normalizes after treatment and whether a recovered neuro-humoral reactivity to stress is associated with diminished relapse.