• cortisol
• hypothalamus-pituitary-adrenal (HPA) axis
• subthreshold mania
• suicidality
• Post-traumatic stress disorder (PTSD)
• dehydroepiandrosterone sulphate (DHEAS)

Although the association between mood disorders, and particularly bipolar disorders, comorbidity and suicidality in post-traumatic stress disorder (PTSD) patients is well established, less information is available on the impact of subsyndromal affective symptoms. The aim of the present study was, thus, to explore the frequency and relationship between subthreshold affective symptoms, suicidality and hormone levels (cortisol and DHEAS), in PTSD patients versus control subjects.
Sixty-five PTSD out-patients without bipolar disorders and 65 healthy control subjects were asked to complete the Mood Spectrum-SR-Lifetime Version (MOODS-SR), a questionnaire exploring the presence of subthreshold affective symptoms. Logistic regression models were used to analyze the relationships between suicidality, explored by 6 items of the MOODS-SR combined and dichotomized to denote the presence or absence of suicidal ideations/plans and/or attempts, and the number of manic/hypomanic or depressive symptoms. Statistically significant and positive associations were found between the presence of manic/hypomanic and depressive symptoms and the likelihood of suicidal ideation or attempts. Besides depressive, even subthreshold manic/hypomanic traits seem to be associated with higher suicidality in PTSD patients.
Biological sample included 100 subjects: 36 patients with DSM-IV diagnosis of chronic PTSD, and a comparison group of 64 healthy control subjects, without any current or lifetime Axis I psychiatric diagnosis. The results of the present study supported the findings on hypothalamus-pituitary-adrenal (HPA) axis alterations in PTSD patients, as significantly lower DHEAS and cortisol levels were reported in patients with respect to control subjects. Further significantly positive correlations between DHEAS levels and the total number of manic/hypomanic spectrum symptoms were found. Both in the whole sample and in the subgroup of trauma-exposed subjects (PTSD and controls with trauma) cortisol level was significantly lower amongst those with suicidal ideations than in those without.
Our findings suggested that decrease cortisol levels and higher total manic component might predict risk for suicidal ideation.