Concurrent attenuated reactivity of alpha-amylase and cortisol is related to disruptive behavior in male adolescents

Abstract

Attenuated reactivity of salivary alpha-amylase has been proposed as a specific sympathetic marker of disruptive behavior in juveniles and may have additional value to studying other autonomic parameters and hypothalamic–pituitary–adrenal axis activity. Investigating the interrelationships between neurobiological parameters in relation to juvenile disruptive behavior may enhance insight into the complex mechanisms at play.

We investigated salivary alpha-amylase, cortisol, heart rate (HR), and heart rate variability (HRV) in response to a standardized public speaking task, and examined interactions between these parameters in relation to disruptive behavior. Participants were 48 delinquent male adolescents (mean age 18.4 years, SD 0.9), with and without a disruptive behavior disorder (resp. DP+, DP −) and 16 matched normal controls (NC). A structured psychiatric interview as well as the Youth Self Report and Child Behavior Checklist were administered to assess disruptive behavior.

Alpha-amylase and cortisol reactivity, but not HR or HRV, showed significant inverse associations with dimensional measures of disruptive behavior. Moreover, both cortisol and alpha-amylase reactivity were significantly lower in the DP + group as compared to the NC group. The mentioned relationships remained present when nicotine use was entered as a covariate. Combining alpha-amylase and cortisol in one model explained a larger part of the variance of disruptive behavior than either single parameter. There were no interactions between alpha-amylase and cortisol or HRV in relation to disruptive behavior.

Attenuated alpha-amylase responsivity to stress is a correlate of disruptive behavior in late-adolescent males. Although nicotine use explains a considerable part of the variance of disruptive behavior, both alpha-amylase and cortisol are related to disruptive behavior, over and above the effect of nicotine use. Combining alpha-amylase and cortisol improved insight into neurobiological mechanisms involved with disruptive behavior; concurrent low reactivity of both parameters was related to higher levels of disruptive behavior.

Introduction

Increasing evidence suggests that disruptive behavior in juveniles is associated with decreased activity of stress-related neurobiological systems, such as the hypothalamic–pituitary–adrenal (HPA) axis and the autonomic nervous system (ANS) (Beauchaine, 2001, Raine, 2002, van Goozen et al., 2007). Regarding the ANS, general autonomic measures (i.e. heart rate) as well as parasympathetic measures (i.e. heart rate variability, HRV) and sympathetic measures (i.e. skin conductance, (nor)epinephrine) have been studied. The study of the sympathetic component, however, has been hampered by difficulties in obtaining biological measures such as (nor)epinephrine. Recently, an easily obtainable marker of sympathetic nervous system activity, namely salivary alpha-amylase has been reported on, and an inverse relationship was found with disruptive behavior (Granger et al., 2007, Nater and Rohleder, 2009). Therefore, in the current study we examined the additional value of alpha-amylase to cortisol, heart rate and HRV as potential correlate of juvenile disruptive behavior. Moreover, we investigated the combined activity as well as interactions between the various parameters in relation to disruptive behavior.

Measures of the ANS like heart rate are regulated by both parasympathetic and sympathetic nervous systems (resp. PNS, SNS), and may therefore be less specific than ‘pure’ PNS or SNS measures (Berntson et al., 1991). While parasympathetic measures have been studied rather extensively in relation to disruptive behavior (Beauchaine, 2001), research on measures of SNS reactivity has been hampered by various methodological difficulties. Although skin conductance is determined as a marker of SNS, it appears to be most valuable for measuring phasic responses to stimuli presented for milliseconds to seconds, rather than psychosocial stress experiments lasting for minutes or hours (Lahey et al., 1993, Popma et al., 2006, van Goozen et al., 2000). Catecholamines as SNS measures are relatively difficult to obtain, which may explain the small number of studies so far. Correlations between plasma catecholamines, particularly norepinephrine, and alpha-amylase have been reported (Chatterton et al., 1996, Rohleder et al., 2004), although findings are not overall consistent (Nater et al., 2006). Because norepinephrine stimulates the output of alpha-amylase by the salivary glands in response to adrenergic sympathetic activation (Bosch et al., 2003), salivary alpha-amylase is likely to act as a specific measure of SNS reactivity (Rohleder et al., 2004, Van Stegeren et al., 2006). Although the specificity of alpha-amylase as a sympathetic measure depends on methodological issues like sampling procedures (Bosch et al., 2011, Nater and Rohleder, 2009, Rohleder and Nater, 2009), it may serve as a correlate of juvenile disruptive behavior.

Regarding the relation between alpha-amylase and disruptive behavior, an overview of the current literature by Granger et al. (2007) indicated inverse relationships in healthy children and adolescents. In a recent study in a general population sample of early-adolescent boys and girls, attenuated alpha-amylase reactivity to the Trier Social Stress Test (TSST) was found in relation to parent-reported disruptive behavior (Susman et al., 2010). To date, studies on the relationship between alpha-amylase reactivity and disruptive behavior in clinic-referred and delinquent samples are lacking.

Other parameters of the stress regulation system have been studied extensively in both clinic-referred or delinquent samples and the general population. Measures include the HPA-axis (represented by cortisol), the ANS (represented by heart rate) and more specific the PNS (represented by HRV). Inverse relationships between these parameters and disruptive behavior have been shown frequently (Beauchaine, 2001, Ortiz and Raine, 2004, van Goozen et al., 2007). These associations have often been explained by theories of low (autonomic) arousal. In these theories, attenuated physiological responsivity to stress is regarded as a marker of low levels of fear and low punishment sensitivity. Fearless juveniles are thought to be more likely to engage in disruptive behaviors because they do not fear the negative consequences of their actions (Raine, 1993, Raine, 2002). Genetic vulnerabilities and/or early life adversities may underlie the attenuated stress responsivity (van Goozen and Fairchild, 2008).

Similarly, it has been proposed that disruptive children are characterized by a mismatch in the interplay between different physiological systems involved in the regulation of stress. For example, regarding SNS and HPA-axis reactivity, generally, activity of both systems increases in response to stress. Bauer et al. (2002) postulated two models that specifically describe the interrelationship between both systems in relation to disruptive behavior. The additive model proposes that low reactivity in both systems concurrently (balanced low activity) is related to elevated levels of disruptive behavior. As such, this fits in with the low arousal theory. Alternatively, the interactive model proposes that low reactivity in one system together with concurrent high reactivity in the other system (asymmetrical or unbalanced activity) is associated with greater risk of disruptive behavior (Bauer et al., 2002). In this model it is thus suggested that the relationship between either of the two systems and disruptive behavior is moderated by the other system. Gordis et al. (2006) tested this hypothesis in a study in which they investigated the interaction between alpha-amylase and cortisol in relation to disruptive behavior. In a sample of maltreated early-adolescents and a control group, they found that interactions between the HPA-axis and the SNS are linked with disruptive behavior. The interaction showed that low activity in both systems was associated with more aggression (Gordis et al., 2006).

Moreover, another mismatch between generally well-coordinated physiological stress systems within the ANS, i.e. the interaction between the SNS and the PNS, has been described in disruptive juveniles. It has generally been assumed that SNS and PNS display coupled, reciprocal actions on organ systems. When SNS activity increases, the PNS activity decreases, and vice versa. However, it has also been argued that the SNS and PNS function as two separate dimensions (Berntson et al., 1991). These non-reciprocal actions may result in concurrent increases or concurrent decreases in both branches, leading to ambiguous effects on physiological arousal (Berntson et al., 1993). Indeed, several studies found concurrent low levels of SNS and PNS to be related to juvenile disruptive behavior (Beauchaine et al., 2007, Boyce et al., 2001, El-Sheikh et al., 2009). Findings warrant replication in other age samples like children or late-adolescents, as well as in specific samples like (clinic referred) disruptive behavior disordered juveniles or delinquents.

There are still inconsistencies in the literature on the relationships between neurobiological parameters and disruptive behavior (Dietrich et al., 2007, Lorber, 2004, Sondeijker et al., 2008), for which several explanations have been given. One explanation may be the different methods of measuring disruptive behavior that were used. Although for clinical purposes it is useful to study disruptive behavior categorized in disorders, dimensional measures are able to distinguish between severe or mild forms of disruptive behavior. It is thus important to relate neurobiological parameters to categorical and dimensional measures of disruptive behavior. Another explanation may be that many studies have focused on only one system at the time, not taking into account cumulative effects and interactions between involved systems (Bauer et al., 2002, Gordis et al., 2006). As explained above, it has been proposed not to focus on physiological systems independently, but to take into account interrelationships as well, to enhance understanding of the associations with juvenile disruptive behavior (Bauer et al., 2002).

Improving knowledge on the neurobiological basis of disruptive behavior may ultimately lead to improved identification of juveniles at risk for a deviant development, such as juvenile delinquents. Therefore, in the present study, we concurrently assessed reactivity of alpha-amylase, cortisol, heart rate and HRV during a public speaking task in delinquent male adolescents and matched normal controls. We related neurobiological reactivity to categorical as well as dimensional measures of disruptive behavior. We investigated whether examining combined reactivity of the parameters alpha-amylase, cortisol, heart rate and HRV improves the explanation of disruptive behavior compared to examining one of the parameters alone. Furthermore, we tested which model of Bauer (additive or interactive) best explains disruptive behavior, by investigating interactions between SNS and HPA-axis reactivity as well as between SNS and PNS reactivity in relation to disruptive behavior.