Relationship of hypothalamus-pituitary-adrenal (HPA) axis function and suicidal behavior in patients with depression

BACKGROUND There have been few neuroendocrinology studies of suicidal behaviors among patients with depression and the results of these studies have been inconsistent. AIM To explore the association between the function of the hypothalamus-pituitary-adrenal (HPA) axis and suicidal behaviors in Chinese patients with depression. METHODS Several measures of HPA functioning in 14 depressed patients who had had suicidal behaviors in the two prior months ('depressed cases') were compared to those of 15 depressed inpatients who did not have prior suicidal behaviors ('depressed controls'): a dexamethasone suppression test (DST), the diurnal changes in serum cortisol levels during a single day before and after 6 weeks of treatment with paroxetine; and 24 h urinary 17-OH cortisol and free corticosterone before and after treatment. The Hamilton Depression Scale (HAMD) was used to measure the severity of depression. Daytime cortisol levels were also assessed in 15 non-depressed controls selected from individuals who had a routine health exam. RESULTS There were no statistically significant differences in the 24 h urinary measures of cortisol and corticosterone between depressed cases and depressed controls. In both groups the normal midnight drop in serum cortisol was nonsignificant prior to treatment but after treatment it became more pronounced. The DST was positive in more of the depressed cases than depressed controls (57% v. 20%, χ(2)=4.24, p=0.039). The correlation of cortisol serum levels with the HAMD total score and the item scores for hopelessness and suicidal ideation were statistically significant in the depressed case group both before and after treatment, but in the depressed control group these correlation coefficients did not reach statistical significance. The 08.00 h serum cortisol level in depressed cases was significantly greater than the level in non-depressed controls both before and after treatment, but the level in depressed controls was not significantly greater than that in non-depressed controls. CONCLUSION These findings are broadly consistent with those of prior studies about the relationship of depression and the functioning of the HPA axis. There were, however, some differences between depressed patients that did and did not report prior suicidal behavior which may indicate suicide-specific characteristics of HPA axis dysfunction. These differences merit further assessment in larger studies that distinguish patients who have made suicide attempts from those who only report prior suicidal ideation.