Hypothalamic-pituitary-adrenal axis, childhood adversity and adolescent nonsuicidal self-injury

Psychoneuroendocrinology 74 (2016) 203–211

Contents lists available at ScienceDirect

Psychoneuroendocrinology journal homepage: www.elsevier.com/locate/psyneuen

Hypothalamic-pituitary-adrenal axis, childhood adversity and adolescent nonsuicidal self-injury Corinna Reichl a,b,1 , Anne Heyer a,1 , Romuald Brunner b,c , Peter Parzer c , Julia Madeleine Völker b,c , Franz Resch c , Michael Kaess a,c,∗ a Section for Translational Psychobiology in Child and Adolescent Psychiatry, Department of Child and Adolescent Psychiatry, Center for Psychosocial Medicine, University of Heidelberg, Heidelberg, Germany b Section for Disorders of Personality Development, Department of Child and Adolescent Psychiatry, Center for Psychosocial Medicine, University of Heidelberg, Heidelberg, Germany c Clinic of Child and Adolescent Psychiatry, Center for Psychosocial Medicine, University of Heidelberg, Heidelberg, Germany

a r t i c l e

i n f o

Article history: Received 13 May 2016 Received in revised form 14 September 2016 Accepted 15 September 2016 Keywords: Hypothalamus-pituitary-adrenal-axis Cortisol awakening response Hair cortisol Non-suicidal self-injury Childhood adversity Adolescence

a b s t r a c t Background: Whereas childhood adversity (CA) and the hypothalamus-pituitary-adrenal (HPA) axis have been suggested to play a major role in the etiology of non-suicidal self-injury (NSSI), no study has thus far investigated both its associations and interactions with adolescent NSSI. Method: We investigated CA (antipathy, neglect, physical, psychological, and sexual abuse) and indices of HPA axis activity (salivary and hair cortisol) in a clinical sample of 26 adolescents engaging in NSSI and 26 age- and gender-matched healthy controls (HC). We used standardized interviews for the assessment of CA (CECA), NSSI (SITBI-G), and axis I diagnoses (MINI-KID). Salivary cortisol sampling was surveyed using a monitoring system and instructed via telephone calls. Results: Adolescents engaging in NSSI exhibited significantly higher cortisol awakening responses compared to HC. No differences were found with respect to the diurnal slope or hair cortisol. In the presence of CA, healthy adolescents showed flatted diurnal cortisol slopes while those engaging in NSSI exhibited significantly steeper ones. Conclusions: Our findings indicate that adolescents engaging in NSSI may exhibit a stronger cortisol awakening response, potentially in expectation of strain. However, elevated cortisol levels may not be maintained throughout the day, especially among adolescents with a history of CA. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Non-suicidal self-injury (NSSI) has been defined as the deliberate, direct destruction of body tissue without conscious suicidal intent, such as self-cutting or carving skin, self-burning, self-hitting, biting self or burning skin (Lloyd-Richardson et al., 2007). NSSI has been observed in the context of various mental disorders (Nock et al., 2006) with a worldwide prevalence rate of 17 percent among adolescents (Swannell et al., 2014). Due to its clinical relevance, NSSI has recently been included as a distinct disorder in section 3 of the new DSM-5 (American Psychiatric Association, 2013), making research on its antecedents and correlates even more important.

∗ Corresponding author at: Clinic of Child and Adolescent Psychiatry, University Hospital Heidelberg Blumenstrasse 8, 69115 Heidelberg, Germany. E-mail address: [email protected] (M. Kaess). 1 These authors contributed equally to this work and should therefore be considered first authors. http://dx.doi.org/10.1016/j.psyneuen.2016.09.011 0306-4530/© 2016 Elsevier Ltd. All rights reserved.

Previous studies especially highlighted the role of childhood adversity (CA) in the context of adolescent NSSI (Kaess et al., 2013). An invalidating, neglecting or abusive family environment may evoke strong negative emotions in children, which, at the same time, are not tolerated (Linehan, 1993). NSSI might consequently arise as a maladaptive strategy in order to regulate unresolved negative affective states. Kaess et al. (2013) reported strong relations between CA, especially maternal neglect and antipathy, and the engagement in NSSI within a representative clinical sample of adolescents and young adults. While a relationship of NSSI and CA has been established (Gratz et al., 2002), little is known about associations of CA with neurobiological mechanisms in the context of adolescent NSSI. However, environment-biology-relations that specifically arise in the context of NSSI may give key answers to the question why some individuals who experience CA start to self-harm and others don’t. The hypothalamic-pituitary-adrenal (HPA) axis is a major stress response system of the human body. Stressors activate the

204

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211

release of corticotropin releasing hormones (CRH) and argininevasopressin (AVP) from the hypothalamus, which foster the release of the hormone adrenocorticotropin (ACTH) from the pituitary gland, and consequently the release of gluco- and mineralocorticoids from the adrenal cortex. Finally, the glucocorticoid cortisol binds to responding receptors in different brain regions, which in turn inhibits the stress-induced secretion of CRH and AVP (Lupien et al., 2009). Previous research revealed CA to be associated with alterations in HPA axis functioning, and consequently with a higher vulnerability for psychopathologies in adulthood (McCrory et al., 2010). The continuous activation of the HPA axis and the associated releases of glucocorticoids may be particularly critical within the sensitive phase of adolescent development, in which it may alter brain structures that are crucial for emotion regulation (Lupien et al., 2009). However, only few studies have thus far focused on adolescent samples (Bosch et al., 2012; MacMillan et al., 2009). Moreover, the underlying mechanisms are still poorly understood with results of previous studies on relations between CA and HPA axis functioning among adolescents being best described as inconsistent (Trickett et al., 2010). With regard to HPA axis reactivity, Bosch et al. (2012) recently found experiences of adversities before puberty to be associated with increased cortisol after stressful situations, whereas adversities experienced at early adolescence resulted in decreased cortisol. Concerning non-stimulated HPA axis activity, results range from reports on a decreased cortisol awakening response (CAR; e.g., Meinlschmidt and Heim, 2005) to reports on increased morning cortisol levels (e.g., Halligan et al., 2004) after CA. Regarding basal cortisol, elevated hair cortisol levels have been reported in the literature following various forms of childhood adversities among a population-based sample of children (Simmons et al., 2016) and related to sexual abuse among adult patients with trauma-related disorders (Schalinski et al., 2015). Data on the neurobiology of NSSI are rare, and little is known about specific neurobiological factors which may either predispose to or co-occur with NSSI (Westlund Schreiner et al., 2015). Clinical experience and empirical evidence show that NSSI often occurs in reaction to stressful situations (Pagano et al., 2004), and therefore might arise from a particular vulnerability to stress or might even be understood as regulatory coping strategy regarding exaggerated emotional stress reactions. Therefore, the HPA axis is likely to play a major role in the context of NSSI. Kaess et al. (2012) found alterations in the neurobiological stress response among adolescents engaging in NSSI, reflected in an attenuated cortisol response to a social stress test compared to healthy controls. Beauchaine et al. (2015) revealed associations between NSSI and decreased post-dexamethasone cortisol. These preliminary results seem to fit previous research on the impact of CA on HPA axis functioning (MacMillan et al., 2009), but data on non-stimulated HPA axis functioning in NSSI are lacking. However, there are several reasons to believe that variations of cortisol secretion during the day may also be altered among adolescents engaging in NSSI. First, previous research found NSSI to be related to experiences of psychological distress (Richmond et al., 2015). These experiences may predispose adolescents engaging in NSSI to expectations of strain, which have been associated with an increased cortisol secretion after awakening (Hall et al., 2010). Moreover, previous research revealed stress experiences to be associated with flatter diurnal slopes (Ly et al., 2015). Second, NSSI has been discussed as a strategy to temporarily relief from negative affectivity (Westlund Schreiner et al., 2015) with individuals engaging in NSSI showing generally higher levels of neuroticism (Mullins-Sweatt et al., 2013). Neuroticism, specifically the facet of anger, has in turn been associated with increased cortisol secretion after awakening (Adam, 2006). Third, Westlund Schreiner et al. (2015) stressed the need for research on alterations in basic arousal systems among individuals engaging in NSSI, argu-

ing that those persons may suffer from sleep difficulties which may impair daily functioning. Finally, it has been argued that relations between CA and HPA axis activity depend on clinical symptoms (Strüber et al., 2014). Previous research discussed the possibility that CA would result in a hypoactivity of the HPA axis in the context of externalizing problems, such as antisocial behavior (Susman, 2006), whereas interactions between CA and internalizing problems, such as depressive symptoms, may result in a hyperactivity of the HPA axis (Heim et al., 2004). Cicchetti et al. (2010) further revealed that children who have been physically or sexually abused during the first 5 years of life showed an attenuated diurnal slope in cortisol only if they showed a large extend of internalizing symptoms. To our knowledge, no study has yet investigated whether engagement in NSSI moderates relations between CA and diurnal cortisol secretion. In our study, we aim at investigating several indices of nonstimulated HPA axis functioning (CAR, diurnal slope, hair cortisol levels) in adolescents engaging in NSSI compared to a healthy control group (HC). We further examine relations between CA (antipathy, neglect, psychological, physical, sexual abuse) and HPA axis indices for both groups. Finally, we test whether associations between CA and HPA axis indices differ between adolescents engaging in NSSI and the HC. The outlined considerations led us to formulate the following hypotheses: First, that adolescents engaging in NSSI show alterations in HPA axis functioning compared to age- and gender-matched HC. Second, that CA is associated with elevated baseline HPA axis activity (hair cortisol) and altered diurnal HPA axis activity (salivary cortisol). Third, that the relationship between CA and HPA axis functioning is moderated by NSSI.

2. Method 2.1. Participants and recruitment Adolescents engaging in NSSI were consecutively recruited from our specialized outpatient clinic for adolescent risk-taking and self-harm behavior (AtR!Sk; “Ambulanz für Risikoverhalten & Selbstschädigung”) as well as our inpatient units at the Clinic of Child and Adolescent Psychiatry, University Hospital Heidelberg, Germany. Participants from the HC group were recruited via public advertisement. Adolescents with acute psychotic symptoms, acute suicidality, poor knowledge of the German language, pregnancy, being on glucocorticoid medication or having endocrine disorders were not included (Stalder et al., 2016). Our study was approved by the institutional ethics committee of the Medical Faculty, University of Heidelberg and was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Informed and written consent was obtained from participants and their parents/caregivers. Our sample consists of 26 adolescents aged between 14 and 18 years who had engaged in NSSI at least 5 times during the last 6 months (24 females; Mage = 16.28 years; SDage = 1.28) and 26 age-, gender- and school type-matched HC who had neither received any psychiatric diagnosis, nor undergone any psychiatric treatment or engaged in NSSI (HC; 24 females; Mage = 16.22 years; SDage = 1.11). In the NSSI group, frequency of NSSI was on average 112 times during the last year (SD = 101.6) and 10.5 times during the last month (SD = 20.0). The most common forms of NSSI were deliberate cutting (100%), followed by manipulating wounds (69.2%), scratching (57.7%), hitting one’s body (50%) and biting (46.2%). Twenty-two adolescents (84.6%) in the NSSI group reported that they had already attempted suicide at least once during lifetime. On average, adolescents in the NSSI group met the criteria for 2.81 axis I diagnoses (SD = 1.52), and fulfilled 4.38 (SD = 2.17) diagnos-

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211

205

Table 1 Number of BPD diagnostic criteria (DSM-IV) and clinical diagnoses (ICD-10) for the NSSI group (n = 26).

BPD diagnostic criteria Fear of abandonment Unstable relationships Identity disturbances Impulsivity Self-harm/Suicidality Affective instability Inner emptiness Inappropriate anger Paranoia/Dissociation Clinical diagnoses (ICD-10) F0 Organic, including symptomatic, mental disorders F1 Mental and behavioral disorders due to psychoactive substance use F2 Schizophrenia, schizotypal and delusional disorders) F3 Mood [affective] disorders F4 Neurotic, stress-related and somatoform disorders F5 Behavioural syndromes associated with physiological disturbances and physical factors F8 Disorders of psychological development F9 Behavioral and emotional disorders with onset usually occurring in childhood and adolescence

N

%

4 11 10 8 26 17 19 11 9

15.38 42.31 38.46 30.77 100 65.38 73.08 42.31 34.62

0 13 2 22 20 7

0 50 7.69 84.62 76.92 26.92

0 9

0 34.62

Note: NSSI = non-suicidal self-injury; BPD = Borderline Personality Disorder. Multiple diagnoses per participant are possible.

tic criteria of BPD. Detailed information on the psychopathological profile of the NSSI group is given in Table 1. In the HC group, no participant had ever engaged in NSSI or had ever attempted suicide. 2.2. Psychological measures Our interviews assessed basic socio-demographic variables along with variables that may influence HPA-axis functioning, such as height, weight, days since the last menstrual cycle, regular use of medication, smoking behavior, and substance abuse. Use of medication and smoking behavior during the last three months were measured by dichotomous variables and drug abuse by a 3-point scale ranging from 1 (never) to 3 (at least 20 days per month). All interviews were performed by a trained clinician in the field of child and adolescent psychiatry. Non-suicidal self-injury (NSSI) was assessed using the German version (SITBI-G; Fischer et al., 2014) of the Self-Injurious Thoughts and Behavior Interview (Nock et al., 2007). The semi-structured interview assesses the presence, frequency, and characteristics of a variety of thoughts and behavior of NSSI as well as suicidal thoughts, gestures, plans, and attempts. Previous research reported good reliability estimates for the German version (SITBI-G) and good convergent validity in relation to an established questionnaire of self-harming behavior (Fischer et al., 2014). Childhood adversities were assessed with the German version (Kaess et al., 2011) of the Childhood Experiences of Care and Abuse Interview (CECA; Bifulco et al., 1994), which has recently been judged as “gold standard” for the assessment of CA (Thabrew et al., 2012). The CECA is a standardized interview which measures subjective experiences of adversities during childhood or adolescence within the family environment. The subscales antipathy, neglect, psychological, physical, and sexual abuse are rated on a 4-point scale ranging from 0 (severe adversity/abuse) to 3 (no adversity/abuse). The subscales have been shown to be highly interrelated in our study, so we calculated a general score of CA (Cronbach’ ␣ = 0.81). Psychiatric diagnoses of the NSSI group were assessed by means of the Mini-International Neuropsychiatric Interview for Children and Adolescents (M.I.N.I.-KID; Sheehan et al., 2010), a structured interview for the assessment of axis I psychiatric disorders according to DSM-IV and ICD-10. Moreover, adolescent borderline-

personality disorder (BPD) traits were assessed with the respective module of the German version of the Structured Clinical Interview for DSM-IV Personality Disorders (SKID II; Fydrich et al., 1997). During the three days of saliva collection at home, adolescents additionally kept a cortisol diary, reporting the time and type of medication use, subjective stress experiences at the time of saliva collection on a 10-point scale ranging from 1 (very low) to 10 (very high), and sleep quality on a 10-point scale ranging from 1 (very bad) to 10 (very good).

2.3. Cortisol measures Saliva samples were collected for the analysis of diurnal cortisol patterns. Previous research revealed saliva cortisol to constitute a good proxy for the assessment of HPA axis activity (Kirschbaum and Hellhammer, 1994). Participants were instructed to collect saliva samples at home at four time points (t1: immediately after awakening, t2: 30 min after awakening, t3: 60 min after awakening, t4: before going to bed) during three consecutive days. They were instructed not to eat food, have drinks, smoke or engage in sports for at least one hour prior sampling. The collection of saliva samples was recorded using the electronic monitoring system (MEMS® , Medication Event Monitoring System). Participants received salivettes marked with colors and numbers within a box that registered each opening of the box with a time signature. The assessment of saliva samples was instructed via telephone by a student assistant, who called participants at individually agreed fixed time points, the first call representing a wake-up call. Participants were called the evening before and reminded of the instructions. Previous research revealed that temporal inaccuracy in saliva sampling affects the validity of CAR data (Griefahn and Robens, 2011). Specifically, delay between awakening and sampling times results in an overestimation of cortisol secretion levels at awakening and an underestimation of the CAR (see Stalder et al., 2016 for a rationale). This is especially crucial, given that only about 6 percent of previous studies used methods allowing for the objective control of awakening and sampling times (Stalder et al., 2016). In order to maximize the accuracy of our data, we (a) regularly briefed participants about sampling procedures and post-awakening behavior, (b) made use of methods allowing for an objective control of awak-

206

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211

ening and sampling times, and (c) excluded inaccurately sampled data. Saliva samples were frozen and stored at −20 ◦ C until assayed. Samples were not included in our analyses in case that the time signature of the MEMS was missing (e.g. due to improper closing) or if the collection after awakening deviated more than 5 min from the proposed time frame (30 or 60 min), resulting in the exclusion of 61 out of 624 saliva samples. We further collected hair samples for the analysis of cumulative cortisol levels of the past 3 months. The analysis of cortisol in hair has been shown to provide a reliable and valid method for the retrospective assessment of cumulative cortisol up to several months (Stalder and Kirschbaum, 2012). Two strands of hair were cut scalp-near from a posterior vertex position. Data were excluded if hair samples were shorter than 3 cm, which was the case for 3 out of 52 samples. Biochemical assays for the analyses of saliva and hair samples were performed at the Laboratory of the Department of Biological Psychology at the Technical University of Dresden (Prof. Clemens Kirschbaum).

2.4. Statistical analyses Analyses were performed using the statistical software STATA 14.0. With respect to our interview data, there were no missing values. In case of missing cortisol indices, participants were excluded from the respective analyses. With respect to indices assessed via the cortisol diary (sleep quality, perceived stress), we imputed missing values due to a high number of missing values (13% for items about sleep quality; 57% for items about stress). The number of missing values did not vary between groups and was not related to any study variable. We first tested for significant differences between the NSSI and the HC group with respect to potential covariates of cortisol (Stalder et al., 2016), CA, sleep quality and perceived stress at time of saliva collection. With respect to diurnal cortisol patterns, we computed three different indices for the cortisol awakening response (CAR): delta between the cortisol increase and ground (g), the area under the curve with respect to ground (AUCg), the area under the curve with respect to increase (AUCi); and one index for the diurnal slope (DSl). Indices were computed if valid cortisol data existed for the respective measurement points from at least two out of three days. For the assessment of g, which describes the increase in cortisol levels across the first 30 min after awakening (Stalder et al., 2016), we computed a change score by subtracting cortisol levels at t1 from levels at t2. The AUCg refers to the post-awakening cortisol secretion during a specific time, in our study during the first 60 min after awakening, taking also account of the hormonal secretion at awakening as the starting point of the g. In contrast, the AUCi measures changes during the first 60 min after awakening and is thus related to the sensitivity of the system (Pruessner et al., 2003). The AUCg and AUCi were computed according to the formulas provided by Pruessner et al. (2003). Finally, the DSl refers to the decline of cortisol secretion throughout the day (Adam and Kumari, 2009). We calculated the DSl by subtracting the individually highest cortisol level in the morning from the cortisol level at bedtime (t4) and dividing this score by the total time awake, which varied individually. Consequently, we tested for significant mean differences in g, AUCg, AUCi, DSl, and hair cortisol levels between the NSSI and the HC group by calculating analyses of variances (ANOVAs), controlling for BMI and smoking behavior (H1). We used two-sided tests of significance. We further regressed g, AUCg, AUCi, DSl, and hair cortisol levels on the general score of CA within each group (H2). We consequently tested for significant differences between regression coefficients in the NSSI and HC group (H3).

3. Results 3.1. Descriptive statistics Adolescents engaging in NSSI did not differ significantly from the HC group in regard to the use of medication (␹2 = 0.59, p = 0.44), drug abuse (␹2 = 2.88, p = 0.24) or the number of days since the beginning of the last menstrual cycle (t = 1.30, df = 38, p = 0.20). However, adolescents engaging in NSSI had a significantly higher body mass index (BMI) compared to the HC group (t = 2.04, df = 50, p = 0.05) and showed significantly more smoking behavior (␹2 = 12.24, p < 0.01). Participants of the NSSI group reported significantly more CA with respect to antipathy (␹2 = 16.27, p < 0.01), neglect (␹2 = 10.60, p = 0.01), physical abuse (␹2 = 8.44, p = 0.04), and the general score of CA (t = 3.67, df = 50, p < 0.01), but no difference with respect to sexual (␹2 = 2.20, p = 0.53) or psychological abuse (␹2 = 7.14, p = 0.07). Adolescents engaging in NSSI reported a significantly worse sleep quality than adolescents of the HC group (F = 14.44, df = 50, p < 0.01). There were no group differences with respect to mean stress experiences at the time of saliva collection (t1: F =0.15, df = 50, p = 0.71; t2: F = 0.22, df = 50, p = 0.65; t3: F = 1.00, df = 50, p = 0.32; t4: F = 0.41, df = 50, p = 0.52). Neither sleep quality nor perceived stress was related to diary cortisol secretion. Means, standard deviations and intercorrelations between study variables for the NSSI and the HC group are presented in Table 2. 3.2. Cortisol, non-suicidal self-injury, and childhood adversity Concerning diurnal cortisol patterns, the NSSI group showed a significant higher mean g (F = 4.74, df = 48, p = 0.03) and a significant higher cumulative cortisol secretion in regard to the mean AUCg (F = 6.68, df = 44, p = 0.01) than the HC group while controlling for BMI and smoking behavior. No group differences were found concerning the AUCi (F = 2.54, df = 44, p = 0.12), the DSl (F = 0.28, df = 49, p = 0.60) or the hair cortisol levels during the past three months (F = 0.45, df = 48, p = 0.51). Mean values of cortisol indices of the NSSI and HC groups are presented in Table 3. Morning cortisol patterns of the NSSI and HC groups are shown in Fig. 1. The mean cortisol levels for the four time points are M(t1) = 10.22 (SD = 4.87), M(t2) = 19.39 (SD = 8.36), M(t3) = 16.84 (SD = 7.17), and M(t4) = 2.89 (SD = 2.85) for the NSSI group and M(t1) = 10.36 (SD = 3.48), M(t2) = 16.76 (SD = 6.14), M(t3) = 16.39 (SD = 7.48), and M(t4) = 1.73 (SD = 2.10) for the HC group. In the NSSI group, frequency of NSSI during the last month was not significantly correlated with the mean g (␤ = −0.04, p = 0.85), the AUCg (␤ = −0.03, p = 0.90), the AUCi (␤ = −0.09, p = 0.70), the DSl (␤ = −0.17, p = 0.40) or hair cortisol (␤ = −0.05, p = 0.80). Within the NSSI group, there were no significant differences between adolescents who met the criteria for a BPD diagnosis (n = 12) and those who did not (n = 14) with respect to the mean g (t =0.71; df = 22; p = 0.49), the AUCg (t = 1.00; df = 20; p = 0.33), the AUCi (t = 1.24; df = 20; p = 0.23), the DSl (t = 0.91; df = 23; p = 0.37) or hair cortisol levels (t = 1.97; df = 23; p = 0.06). In the NSSI group, CA was associated with the frequency of NSSI during lifetime (ˇ = 0.39, p < 0.05) but not during the last year (ˇ = 0.19, p = 0.36) or month (ˇ = 0.05, p = 0.81). The experience of CA was unrelated to the g (NSSI: ˇ = −0.13, p = 0.52; HC: ˇ = −0.04, p = 0.80), the AUCg (NSSI: ˇ = 0.19, p = 0.35; HC: ˇ = −0.28, p = 0.07), the AUCi (NSSI: ˇ = −0.07, p = 0.74; HC: ˇ = −0.03, p = 0.85) or hair cortisol levels (NSSI: ˇ = 0.01, p = 0.96; HC: ˇ = 0.15, p = 0.34) in both groups. However, the DSl was significantly related to the experience of CA within groups. Specifically, the group affiliation (NSSI vs. HC) significantly moderated associations between CA and DSl (ˇ = −0.53, p < 0.01). Whereas higher levels of CA were associated to a steeper slope of cortisol during day within the NSSI group (ˇ = −0.47, p = 0.02), CA were positively, though not significantly

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211

207

Table 2 Means, Standard Deviations, and Pearson Correlations among Study Variables for the NSSI group (n = 26) and the HC group (n = 26). CAR CAR AUCg AUCi DSl haircortisol antipathy neglect abusephys abusesex abusepsych CA M SD

0.73 0.89 −0.04 0.23 −0.16 −0.30 −0.19 0.26 −0.20 −0.16 10.15 5.28

AUCg

AUCi

DSl

haircortisol

antipathy

neglect

abusephys

abusesex

abusepsych

CA

M

SD

0.78

0.98 0.75

0.44 −0.30 0.38

0.03 −0.07 −0.05 0.27

−0.11 −0.43 −0.15 0.32 0.04

−0.07 −0.36 −0.07 0.31 0.15 0.60

−0.04 −0.36 −0.03 0.40 0.24 0.64 0.46

0.03 −0.21 0.07 0.35 0.08 0.79 0.72 0.50

0.00 −0.29 0.06 0.45 0.26 0.67 0.73 0.66 0.83

−0.05 −0.38 −0.03 0.41 0.17 0.88 0.83 0.74 0.91 0.90

6.58 14.99 5.11 −0.58 2.99 0.05 0.35 0.27 0.19 0.12 1.42

6.89 4.99 5.25 0.29 2.23 0.76 0.75 0.53 0.69 0.59 2.84

0.57 −0.70 0.30 0.16 0.03 0.21 0.34 −0.13 0.18 17.86 5.30

0.06 0.44 −0.07 −0.19 −0.14 0.12 −0.13 −0.11 7.09 3.41

−0.01 −0.37 −0.06 −0.48 −0.41 −0.13 −0.43 −0.50 0.36

0.07 0.08 0.06 −0.07 −0.08 0.01 2.84 1.33

0.77 0.67 −0.04 0.49 0.84 1.58 0.99

0.55 −0.09 0.38 0.75 1.00 0.89

0.12 0.59 0.86 0.92 0.98

−0.16 0.26 0.46 1.03

0.69 0.62 1.06

4.58 3.34

Note: NSSI = non-suicidal self-injury; HC = healthy control; CAR = cortisol awakening response; AUCg = area under the curve with respect to ground; AUCi = area under the curve with respect to increase; DSl = diurnal slope; haircortisol = cortisol levels in hair during the past three months; abusephys = physical abuse; abusesex = sexual abuse; abusepsych = psychological abuse; role rev. = role reversal; CA = general score of childhood adversitiy ranging from 0 to 15; SD = standard deviation. Salivary cortisol is quantified in nmol/l. Hair cortisol is quantified in pg/mg. Values above the diagonal refer to the HC group. Values below the diagonal refer to the NSSI group. Correlations of r > 0.40| are significant at p ≤ 0.05. Correlations of r > 0.55| are significant at p ≤ 0.01. Table 3 Cortisol indices and group differences for the NSSI group and the HC group. NSSI

g AUCg AUCi DSl haircortisol

Not controlling for any covariates

Controlling for BMI and smoking behavior

n

M

SD

n

HC M

SD

F

p

d

F

p

24 22 22 25 25

10.15 17.86 7.09 −0.50 2.84

5.28 5.30 3.41 0.36 1.33

25 23 23 25 24

6.58 14.99 5.11 −0.58 2.99

6.89 4.99 5.25 0.29 2.23

4.12 3.52 2.21 0.68 0.09

0.05 0.07 0.14 0.41 0.77

0.59 0.57 0.46 0.25 0.08

4.75 6.68 2.54 0.28 0.45

0.03 0.01 0.12 0.60 0.51

Note: NSSI = non-suicidal self-injury; HC = healthy control; g = delta between the cortisol increase and ground; AUCg = area under the curve with respect to ground; AUCi = area under the curve with respect to increase; DSl = diurnal slope representing the average slope per hour; haircortisol = cortisol levels in hair during the past three months; n = sample size; M = mean; SD = standard deviation; p = probability; d = Cohen’s d. Salivary cortisol is quantified in nmol/l. Hair cortisol is quantified in pg/mg.

related to the DSl within the HC group, indicating a flattened slope of cortisol (ˇ = 0.27, p = 0.06) (see Fig. 2). The pattern of our results did not change if analyses were only performed for female adolescents.

4. Discussion In our study, we found alterations in the HPA stress system among adolescents engaging in NSSI compared to a sample of ageand gender-matched healthy participants. Adolescents engaging in NSSI showed a higher CAR, reflected in a significant higher g, a significant higher AUCg and a trend toward a higher AUCi while controlling for smoking behavior and BMI. In contrast, data did not indicate any group differences in basal cortisol levels of the last three months as reflected by hair cortisol. Whereas experiences of CA turned out to be unrelated to the CAR and to hair cortisol, CA differentially affected the DSl within both groups. CA revealed to be associated with a steeper slope in the NSSI group and a flattened slope in the HC group. Engagement in NSSI further came along with more severe reports of CA, specifically antipathy, neglect, psychological and physical abuse, which is in line with previous research on NSSI (Kaess et al., 2013). Our sample mainly consisted of female adolescents (92%), therefore results and implications drawn from our study cannot be generalized to the population of male adolescents. Whereas previous studies reported augmented morning cortisol secretion among patients with a diagnosis of BPD (Lieb et al., 2004; Rausch et al., 2015), to our knowledge no study has thus far specifically investigated whether this effect consists among adolescent patients engaging in NSSI. BPD is characterized by a pervasive pattern of instability in various life domains, resulting among others in difficulties in impulse control, strong emotional responses and

dissociative states (American Psychiatric Association, 2013). Concomitant, BPD patients generally experience more daily hassles and life stress than healthy individuals (Jovev and Jackson, 2006), which has been urged as one possible explanation for BPD patients showing an increased CAR. NSSI is often used as a maladaptive strategy to regulate strong negative emotions and to reduce stress reactions (Klonsky et al., 2015). Accordingly, previous research found NSSI to be associated with increased psychological distress (Richmond et al., 2015). Our finding of an elevated CAR among adolescents engaging in NSSI may consequently be interpreted in the sense of those individuals experiencing more psychosocial stress than HC or of being prone to contemplate strain and conflicts of the day ahead (Chida and Steptoe, 2009). In contrast, adolescents of the NSSI group did not report significantly more stress experiences at the time of saliva collection, though these results have to be interpreted with caution due to the high percentage of missing data. Group differences in the CAR may also be attributed to adolescents engaging in NSSI having a poorer sleep quality. This interpretation would be in line with the assumptions of Westlund Schreiner et al. (2015), who argued that individuals engaging in NSSI may suffer from sleep difficulties which may impair daily functioning. However, sleep quality was not related to salivary cortisol in our study. Since the exact function and meaning of the CAR is not yet sufficiently clarified (Stalder et al., 2016), further research on the phenomenon of the CAR will likely support interpretation of our results. In contrast to our findings on the CAR, basal hair cortisol, which represented cumulative cortisol levels over the past 3 months, the diurnal slope, as well as evening cortisol did not differ significantly between the NSSI and the HC group. Overall, our findings indicate that adolescents engaging in NSSI may exhibit a stronger CAR, potentially in expectation of strain, whereas HPA axis activity may not generally be elevated through-

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211

5

10

Cortisol [nmol/l] 20 15

25

208

awakening

awakening + 30" Mean NSSI Mean HC

awakening + 60" 95% CI NSSI 95% CI HC

-1.5

-1

diurnal slope -.5

0

.5

Fig. 1. Mean diurnal saliva cortisol levels of the NSSI and the HC group. Note. Mean values include data from at least two out of three days. CI = confidence interval; NSSI = non-suicidal self-injury; HC = healthy control.

0

5 10 severity of childhood adversities NSSI participants NSSI fitted values

15

HC participants HC fitted values

Fig. 2. Regression between childhood adversity and the diurnal slope within groups. Note. NSSI = non-suicidal self-injury group; HC = healthy control group. The diurnal slope represents the mean slope in cortisol per hour. Childhood adversity is indicated by high scores on the general trauma score.

out the day. Cortisol can on the short term help individuals to recover from stressful experiences and reduce excitability (de Kloet et al., 2008); thus, one might argue that individuals with NSSI might require higher levels of cortisol in order to regulate their stress levels. However, if the HPA axis is excessively activated over a long period, this may on the long term result in tissue damages and health problems (allostatic load theory; McEwen, 1998). Meta-analytic findings even found cortisol levels to be associated with suicide attempts among young adults (O’Connor et al., 2016). On the one hand, if cortisol secretion may primarily be raised in expectation of strain in the morning but increased cortisol levels cannot be maintained throughout the day, it probably cannot fully exert its protective function. NSSI may consequently be adapted as a maladaptive strategy to cope with experiences of psychological distress. This hypothesis is in line with previous results on an atten-

uated cortisol response to acute stress in both adolescents with NSSI (Kaess et al., 2012) and adults with BPD (Nater et al., 2010). On the other hand, cortisol may on the long term foster ill-being (McEwen, 1998). If engagement in NSSI helps adolescents to cope with stressors, this may alternatively explain why elevated cortisol levels are not maintained throughout the day and may even protect from allostatic load. Finally, our results could be due to methodological problems of our study. All but one individual from the NSSI group met the criteria of at least one axis I diagnosis. We therefore cannot rule out the possibility that significant differences between the NSSI and the HC group are due to the confounding of engagement in NSSI with psychopathology. In line with previous research (Kaess et al., 2013; Muehlenkamp et al., 2010), adolescents engaging in NSSI reported significantly more experiences of CA than an age- and gender-matched HC

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211

group. However, CA turned out to be unrelated to the cortisol secretion after awakening (g; AUCg; AUGi) as well as to basal cortisol levels within hair. Experiences of CA could temporarily alter HPA axis activity during childhood but it is possible that this effect does not outlast until adolescence. In this case, alterations in HPA axis activity during childhood could alter brain structures that are crucial for emotion regulation (Lupien et al., 2009), which may indirectly contribute to the engagement in NSSI during adolescence. Though, longitudinal data would be necessary to draw these causal conclusions. CA may not be correlated to cortisol indices due to methodological problems (e.g. retrospective assessment of CA; lack of consideration of current CA) or to the fact that the constructs are not related in the adolescent population. Our findings do not support the idea of effects between CA and NSSI being mediated by HPA axis activity. Alternative neurobiological factors such as brain activity or other hormonal systems as well as environmental factors (e.g. social learning of coping mechanisms) may explain relations between CA and NSSI. Previous research further reported adults engaging in NSSI to show different patterns of comorbid diagnoses as a function of type and severity of CA (Vaughn et al., 2015). In our study, subtypes of CA were highly interrelated, so we could not test for differential effects of forms of CA. Future research may benefit from investigating the effects of isolated forms of traumatic experiences on HPA axis functioning. It is possible that HPA axis functioning mediates effects from specific traumata, such as antipathy, neglect or abuse. Moreover, relations between CA and HPA axis activity may occur only among individuals with specific clinical symptoms. We found relations between CA and the DSl as a function of NSSI. Previous research in non-clinical adult samples revealed an attenuated DSl of cortisol to be associated with psychosocial stress such as poorer relationship functioning, various demands at work and home (Adam and Gunnar, 2001) or same-day feelings of tension and anger (Adam et al., 2006). Accordingly, in our study adolescents of the HC group with CA exhibited a slower rate of decline in cortisol across the day than healthy adolescence without CA, which could be interpreted as a chronic stress reaction. In contrast, CA resulted in a steeper DSl among adolescents engaging in NSSI. Importantly, this effect was not primarily due to an increased HPA axis activity after awakening − CA did not moderate the CAR. Thus, these findings may be related to a CA-dependent attenuation of HPA axis capacities that has been postulated by various researchers (Miller et al., 2007; Susman, 2006). However, previous research yielded inconsistent findings as to whether early adversity causes HPA axis hyper- or hypofunctioning later in life (for a review, see Strüber et al., 2014). These heterogeneous findings may be due to clinical symptoms moderating relations between CA and HPA axis activity (Cicchetti et al., 2010). It has been discussed that CA results in attenuated cortisol secretion primarily among individuals with externalizing problems (Cicchetti et al., 2010). It would be interesting for future research, to investigate whether the moderating effect of NSSI on relations between CA and the diurnal slope further depends on the individual functions of NSSI. Those may refer to internalizing processes, such as intrapersonal functions, or to externalizing processes such as social functions. In our study, HC with former CA exhibited elevated cortisol levels during the day, which can foster recovery from acute stressors (de Kloet et al., 2008) and may reduce stress-related health problems. In contrast, adolescents engaging in NSSI generally had a higher CAR, which may be interpreted as an indicator of expectation of strain (Hall et al., 2010). However those who had experienced CA exhibit an increased decline of cortisol during the day, which may contribute on the short term to poorer problem solving and the development of strong negative emotions that often occur in relation to NSSI. On the long term, the steeper decline in cortisol may protect ado-

209

lescents engaging in NSSI from experiencing allostatic load after CA. 5. Summary and conclusions To sum up, we found alterations in HPA axis activity among adolescents engaging in NSSI compared to an age- and gender matched HC group. NSSI was associated with higher HPA axis activity in the morning (as reflected by the CAR) but unrelated to the DSl or basal cortisol deposited in hair. CA was unrelated to all indices of HPA axis functioning except for the DSl. CA resulted in a tendentially but not significantly flattened DSl among healthy adolescents, and to a significantly steeper DSl among adolescents engaging in NSSI. Our findings indicate that adolescents engaging in NSSI may exhibit a stronger cortisol awakening response, potentially in expectation of strain. However, elevated cortisol levels may not be maintained throughout the day, especially among adolescents with a history of CA. Our results contribute to the identification of neurobiological and environmental correlates of NSSI and their potential interplay. Specifically, our findings underline the importance of alterations in the neurobiological stress response system among adolescents engaging in NSSI. Future research should try to independently replicate the findings, and investigate HPA axis activity with (e.g. acute stress test) and without stimulation (e.g. CAR or DSl) in order to confirm the hypothetical relationship of the different findings so far. Contributors Corinna Reichl and Michael Kaess had the idea for the study and developed the study design. All authors took an active part in the implementation and recruitment of the study. Anne Heyer, Corinna Rei-chl and Julia M. Völker collected the data. Corinna Reichl and Peter Parzer analyzed the data. Corinna Reichl and Michael Kaess drafted the manuscript. All authors revised the manuscript and approved its final version. Role of the funding source The Dres. Majic/Majic-Schelz-Foundation funded the cortisol analyses and reimbursement of study participants. The foundation had no involvement in the collection, analysis or interpretation of data, in the writing of the manuscript or the decision to submit the article for publication. Acknowledgement This study has been funded by the Dres. Majic/Majic-SchelzFoundation. References Adam, E.K., Gunnar, M.R., 2001. Relationship functioning and home and work demands predict individual differences in diurnal cortisol patterns in women. Psychoneuroendocrinology 26, 189–208, http://dx.doi.org/10.1016/S03064530(00)00045-7. Adam, E.K., Kumari, M., 2009. Assessing salivary cortisol in large-scale, epidemiological research. Psychoneuroendocrinology 34, 1423–1436, http:// dx.doi.org/10.1016/j.psyneuen.2009.06.011. Adam, E.K., Hawkley, L.C., Kudielka, B.M., Cacioppo, J.T., 2006. Day-to-day dynamics of experience − cortisol associations in a population-based sample of older adults. Proc. Natl. Acad. Sci. U. S. A. 103, 17058–17063, http://dx.doi. org/10.1073/pnas.0605053103. Adam, E.K., 2006. Transactions among adolescent trait and state emotion and diurnal and momentary cortisol activity in naturalistic settings. Psychoneuroendocrinology 31, 664–679, http://dx.doi.org/10.1016/j. psyneuen.2006.01.010.

210

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211

American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. American Psychiatric Publishing, Arlington, USA. Beauchaine, T.P., Crowell, S.E., Hsiao, R.C., 2015. Post-dexamethasone cortisol, self-inflicted injury, and suicidal ideation among depressed adolescent girls. J. Abnorm. Child Psychol. 43, 619–632, http://dx.doi.org/10.1007/s10802-0149933-2. Bifulco, A., Brown, G.W., Harris, T.O., 1994. Childhood Experience of Care and Abuse (CECA): a retrospective interview measure. J. Child Psychol. Psychiatry 35, 1419–1435. Bosch, N.M., Riese, H., Reijneveld, S.A., Bakker, M.P., Verhulst, F.C., Ormel, J., Oldehinkel, A.J., 2012. Timing matters: long term effects of adversities from prenatal period up to adolescence on adolescents’ cortisol stress response. The TRAILS study. Psychoneuroendocrinology 37, 1439–1447, http://dx.doi.org/10. 1016/j.psyneuen.2012.01.013. Chida, Y., Steptoe, A., 2009. Cortisol awakening response and psychosocial factors: a systematic review and meta-analysis. Biol. Psychol. 80, 265–278, http://dx. doi.org/10.1016/j.biopsycho.2008.10.004. Cicchetti, D., Rogosch, F.A., Gunnar, M.R., Toth, S.L., 2010. The differential impacts of early physical and sexual abuse and internalizing problems on daytime cortisol rhythm in school-aged children. Child Dev. 81, 252–269, http://dx.doi. org/10.1111/j.1467-8624.2009.01393.x. de Kloet, E.R., Karst, H., Joëls, M., 2008. Corticosteroid hormones in the central stress response: quick-and-slow. Front. Neuroendocrinol. 29, 268–272, http:// dx.doi.org/10.1016/j.yfrne.2007.10.002. Fischer, G., Ameis, N., Parzer, P., Plener, P.L., Groschwitz, R., Vonderlin, E., Kölch, M., Brunner, R., Kaess, M., 2014. The German version of the self-injurious thoughts and behaviors interview (SITBI-G): a tool to assess non-suicidal self-injury and suicidal behavior disorder. BMC Psychiatry 14, 265, http://dx.doi.org/10.1186/ s12888-014-0265-0. Fydrich, T., Renneberg, B., Schmitz, B., Wittchen, H.-U., 1997. Strukturiertes Klinisches Interview für DSM-IV − Achse II: Persönlichkeitsstörungen. In: Structured Clinical Interview for DSM-IV − Axis II: Personality Disorders. Hogrefe, Göttingen. Gratz, K.L., Conrad, S.D., Roemer, L., 2002. Risk factors for deliberate self-harm among college students. Am. J. Orthopsychiatry 72, 128–140. Griefahn, B., Robens, S., 2011. Cortisol awakening response: are sampling delays of 15 minutes acceptable? Int. J. Psychophysiol. Off. J. Int. Organ. Psychophysiol. 82, 202–205, http://dx.doi.org/10.1016/j.ijpsycho.2011.08.005. Hall, G.B., Dollard, M.F., Tuckey, M.R., Winefield, A.H., Thompson, B.M., 2010. Job demands, work-family conflict, and emotional exhaustion in police officers: a longitudinal test of competing theories. J. Occup. Organ. Psychol. 83, 237–250, http://dx.doi.org/10.1348/096317908X401723. Halligan, S.L., Herbert, J., Goodyer, I.M., Murray, L., 2004. Exposure to postnatal depression predicts elevated cortisol in adolescent offspring. Biol. Psychiatry 55, 376–381, http://dx.doi.org/10.1016/j.biopsych.2003.09.013. Heim, C., Plotsky, P.M., Nemeroff, C.B., 2004. Importance of studying the contributions of early adverse experience to neurobiological findings in depression. Neuropsychopharmacol. Off. Publ. Am. Coll. Neuropsychopharmacol. 29, 641–648, http://dx.doi.org/10.1038/sj.npp. 1300397. Jovev, M., Jackson, H.J., 2006. The relationship of borderline personality disorder, life events and functioning in an Australian psychiatric sample. J. Personal. Disord. 20, 205–217, http://dx.doi.org/10.1521/pedi.2006.20.3.205. Kaess, M., Parzer, P., Mattern, M., Resch, F., Bifulco, A., Brunner, R., 2011. Childhood Experiences of Care and Abuse (CECA): Validierung der deutschen Version von Fragebogen und korrespondierendem Interview sowie Ergebnisse einer Untersuchung von Zusammenhängen belastender Kindheitserlebnisse mit suizidalen Verhaltensweisen [Childhood Experiences of Care and Abuse (CECA): Validation of the German version and results of investigation of correlations between adverse childhood experiences and suicidal behaviour]. Z. Für Kinder- Jugendpsychiatrie Psychother. 39, 243–252, http://dx.doi.org/10. 1024/1422-4917/a000115. Kaess, M., Hille, M., Parzer, P., Maser-Gluth, C., Resch, F., Brunner, R., 2012. Alterations in the neuroendocrinological stress response to acute psychosocial stress in adolescents engaging in nonsuicidal self-injury. Psychoneuroendocrinology 37, 157–161, http://dx.doi.org/10.1016/j. psyneuen.2011.05.009. Kaess, M., Parzer, P., Mattern, M., Plener, P.L., Bifulco, A., Resch, F., Brunner, R., 2013. Adverse childhood experiences and their impact on frequency severity, and the individual function of nonsuicidal self-injury in youth. Psychiatry Res. 206, 265–272. Kirschbaum, C., Hellhammer, D.H., 1994. Salivary cortisol in psychoneuroendocrine research: recent developments and applications. Psychoneuroendocrinology 19, 313–333, http://dx.doi.org/10.1016/0306-4530(94)90013-2. Klonsky, E.D., Glenn, C.R., Styer, D.M., Olino, T.M., Washburn, J.J., 2015. The functions of nonsuicidal self-injury: converging evidence for a two-factor structure. Child Adolesc. Psychiatry Ment. Health 9, 44, http://dx.doi.org/10. 1186/s13034-015-0073-4. Lieb, K., Rexhausen, J.E., Kahl, K.G., Schweiger, U., Philipsen, A., Hellhammer, D.H., Bohus, M., 2004. Increased diurnal salivary cortisol in women with borderline personality disorder. J. Psychiatr. Res. 38, 559–565, http://dx.doi.org/10.1016/j. jpsychires.2004.04.002. Linehan, M., 1993. Cognitive-behavioral Treatment of Borderline Personality Disorder. Guilford Press, New York. Lloyd-Richardson, E.E., Perrine, N., Dierker, L., Kelley, M.L., 2007. Characteristics and functions of non-suicidal self-injury in a community sample of

adolescents. Psychol. Med. 37, 1183–1192, http://dx.doi.org/10.1017/ S003329170700027X. Lupien, S.J., McEwen, B.S., Gunnar, M.R., Heim, C., 2009. Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nat. Rev. Neurosci. 10, 434–445, http://dx.doi.org/10.1038/nrn2639. Ly, J., McGrath, J.J., Gouin, J.-P., 2015. Poor sleep as a pathophysiological pathway underlying the association between stressful experiences and the diurnal cortisol profile among children and adolescents. Psychoneuroendocrinology 57, 51–60, http://dx.doi.org/10.1016/j.psyneuen.2015.03.006. MacMillan, H.L., Georgiades, K., Duku, E.K., Shea, A., Steiner, M., Niec, A., Tanaka, M., Gensey, S., Spree, S., Vella, E., Walsh, C.A., De Bellis, M.D., Van der Meulen, J., Boyle, M.H., Schmidt, L.A., 2009. Cortisol response to stress in female youths exposed to childhood maltreatment: results of the Youth Mood Project. Biol. Psychiatry 66, 62–68, http://dx.doi.org/10.1016/j.biopsych.2008.12.014. McCrory, E., De Brito, S.A., Viding, E., 2010. Research review: the neurobiology and genetics of maltreatment and adversity. J. Child Psychol. Psychiatry 51, 1079–1095, http://dx.doi.org/10.1111/j.1469-7610.2010.02271.x. McEwen, B.S., 1998. Protective and damaging effects of stress mediators. N. Engl. J Med. 338, 171–179, http://dx.doi.org/10.1056/NEJM199801153380307. Meinlschmidt, G., Heim, C., 2005. Decreased cortisol awakening response after early loss experience. Psychoneuroendocrinology 30, 568–576, http://dx.doi. org/10.1016/j.psyneuen.2005.01.006. Miller, G.E., Chen, E., Zhou, E.S., 2007. If it goes up, must it come down? Chronic stress and the hypothalamic-pituitary-adrenocortical axis in humans. Psychol. Bull. 133, 25–45, http://dx.doi.org/10.1037/0033-2909.133.1.25. Muehlenkamp, J.J., Kerr, P.L., Bradley, A.R., Larsen, M.A., 2010. Abuse subtypes and nonsuicidal self-injury: preliminary evidence of complex emotion regulation patterns. J. Nerv. Ment. Dis. 198, 258–263, http://dx.doi.org/10.1097/NMD. 0b013e3181d612ab. Mullins-Sweatt, S.N., Lengel, G.J., Grant, D.M., 2013. Non-suicidal self-injury: the contribution of general personality functioning. Personal. Ment. Health 7, 56–68, http://dx.doi.org/10.1002/pmh.1211. Nater, U.M., Bohus, M., Abbruzzese, E., Ditzen, B., Gaab, J., Kleindienst, N., Ebner-Priemer, U., Mauchnik, J., Ehlert, U., 2010. Increased psychological and attenuated cortisol and alpha-amylase responses to acute psychosocial stress in female patients with borderline personality disorder. Psychoneuroendocrinology 35, 1565–1572, http://dx.doi.org/10.1016/j. psyneuen.2010.06.002. Nock, M.K., Joiner Jr., T.E., Gordon, K.H., Lloyd-Richardson, E., Prinstein, M.J., 2006. Non-suicidal self-injury among adolescents: diagnostic correlates and relation to suicide attempts. Psychiatry Res. 144, 65–72, http://dx.doi.org/10.1016/j. psychres.2006.05.010. Nock, M.K., Holmberg, E.B., Photos, V.I., Michel, B.D., 2007. Self-injurious thoughts and behaviors Interview: development, reliability, and validity in an adolescent sample. Psychol. Assess. 19, 309–317, http://dx.doi.org/10.1037/ 1040-3590.19.3.309. O’Connor, D.B., Ferguson, E., Green, J.A., O’Carroll, R.E., O’Connor, R.C., 2016. Cortisol levels and suicidal behavior: a meta-analysis. Psychoneuroendocrinology 63, 370–379. Pagano, M.E., Skodol, A.E., Stout, R.L., Shea, M.T., Yen, S., Grilo, C.M., Sanislow, C.A., Bender, D.S., McGlashan, T.H., Zanarini, M.C., Gunderson, J.G., 2004. Stressful life events as predictors of functioning: findings from the collaborative longitudinal personality disorders study. Acta Psychiatr. Scand. 110, 421–429, http://dx.doi.org/10.1111/j.1600-0447.2004.00398.x. Pruessner, J.C., Kirschbaum, C., Meinlschmid, G., Hellhammer, D.H., 2003. Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change. Psychoneuroendocrinology 28, 916–931, http://dx.doi.org/10.1016/S03064530(02)00108-7. Rausch, J., Gäbel, A., Nagy, K., Kleindienst, N., Herpertz, S.C., Bertsch, K., 2015. Increased testosterone levels and cortisol awakening responses in patients with borderline personality disorder: gender and trait aggressiveness matter. Psychoneuroendocrinology 55, 116–127, http://dx.doi.org/10.1016/j. psyneuen.2015.02.002. Richmond, S., Hasking, P., Meaney, R., 2015. Psychological distress and non-suicidal self-injury: the mediating roles of rumination, cognitive reappraisal, and expressive suppression. Arch. Suicide Res. Off. J. Int. Acad. Suicide Res., 1–11, http://dx.doi.org/10.1080/13811118.2015.1008160 (Epub ahead of print). Schalinski, I., Elbert, T., Steudte-Schmiedgen, S., Kirschbaum, C., 2015. The cortisol paradox of trauma-related disorders: lower phasic responses but higher tonic levels of cortisol are associated with sexual abuse in childhood. PLoS One 10, e0136921, http://dx.doi.org/10.1371/journal.pone.0136921. Sheehan, D.V., Sheehan, K.H., Shytle, R.D., Janavs, J., Bannon, Y., Rogers, J.E., Milo, K.M., Stock, S.L., Wilkinson, B., 2010. Reliability and validity of the mini international neuropsychiatric interview for children and adolescents (MINI-KID). J. Clin. Psychiatry 71, 313–326, http://dx.doi.org/10.4088/JCP. 09m05305whi. Simmons, J.G., Badcock, P.B., Whittle, S.L., Byrne, M.L., Mundy, L., Patton, G.C., Olsson, C.A., Allen, N.B., 2016. The lifetime experience of traumatic events is associated with hair cortisol concentrations in community-based children. Psychoneuroendocrinology 63, 276–281, http://dx.doi.org/10.1016/j. psyneuen.2015.10.004. Stalder, T., Kirschbaum, C., 2012. Analysis of cortisol in hair- state of the art and future directions. Brain Behav. Immun. 26, 1019–1029, http://dx.doi.org/10. 1016/j.bbi.2012.02.002.

C. Reichl et al. / Psychoneuroendocrinology 74 (2016) 203–211 Stalder, T., Kirschbaum, C., Kudielka, B.M., Adam, E.K., Pruessner, J.C., Wüst, S., Dockray, S., Smyth, N., Evans, P., Hellhammer, D.H., Miller, R., Wetherell, M.A., Lupien, S.J., Clow, A., 2016. Assessment of the cortisol awakening response: expert consensus guidelines. Psychoneuroendocrinology 63, 414–432, http:// dx.doi.org/10.1016/j.psyneuen.2015.10.010. Strüber, N., Strüber, D., Roth, G., 2014. Impact of early adversity on glucocorticoid regulation and later mental disorders. Neurosci. Biobehav. Rev. 38, 17–37, http://dx.doi.org/10.1016/j.neubiorev.2013.10.015. Susman, E.J., 2006. Psychobiology of persistent antisocial behavior: stress, early vulnerabilities and the attenuation hypothesis. Neurosci. Biobehav. Rev. 30, 376–389, http://dx.doi.org/10.1016/j.neubiorev.2005.08.002. Swannell, S.V., Martin, G.E., Page, A., Hasking, P., St John, N.J., 2014. Prevalence of nonsuicidal self-injury in nonclinical samples: systematic review, meta-analysis and meta-regression. Suicide Life Threat Behav. 44, 273–303, http://dx.doi.org/10.1111/sltb.12070.

211

Thabrew, H., de Sylva, S., Romans, S.E., 2012. Evaluating childhood adversity. Adv. Psychosom. Med. 32, 35–57. Trickett, P.K., Noll, J.G., Susman, E.J., Shenk, C.E., Putnam, F.W., 2010. Attenuation of cortisol across development for victims of sexual abuse. Dev. Psychopathol. 22, 165–175, http://dx.doi.org/10.1017/S0954579409990332. Vaughn, M.G., Salas-Wright, C.P., Underwood, S., Gochez-Kerr, T., 2015. Subtypes of non-suicidal self-injury based on childhood adversity. Psychiatr. Quat. 86, 137–151, http://dx.doi.org/10.1007/s11126-014-9313-7. Westlund Schreiner, M., Klimes-Dougan, B., Begnel, E.D., Cullen, K.R., 2015. Conceptualizing the neurobiology of non-suicidal self-injury from the perspective of the Research Domain Criteria Project. Neurosci. Biobehav. Rev. 57, 381–391, http://dx.doi.org/10.1016/j.neubiorev.2015.09.011.