Abnormal cortisol dynamics after traumatic brain injury. Lack of utility in predicting agitation or therapeutic response to tricyclic antidepressants.

A period of significant agitation affects up to 30% of patients after traumatic brain injury. The severity and persistence of this agitation may be such as to require pharmacologic methods as part of the treatment plan. To define which subgroup of patients develop severe agitation warranting intervention and to utilize the information to predict therapeutic responsiveness to tricyclic antidepressants (TCA), we examined cortisol dynamics in 35 traumatically brain-injured (TBI) patients 2-10 months post-TBI. Fasting hypercortisolemia (cortisol greater than 20 micrograms/dl) and/or an absent diurnal variation (1600:0800 cortisol greater than 0.5) was noted in more than 70% of TBI subjects. These abnormalities in cortisol dynamics were not predictive of severe agitation (chi 2 = 0, df = 1, P = 0.99 for hypercortisolemia; chi 2 = 0.163, df = 1, P = 0.7 for absent diurnal variation) and did not differ significantly between TCA responders and nonresponders. The cortisol response to dexamethasone suppression was abnormal (postdexamethasone cortisol value at 0800 and 1600 greater than 5 micrograms/dl) in 34 of 35 subjects and was also not predictive of the presence of agitation. The 0800 cortisol was lower in TCA nonresponders in comparison with TCA responders (8.3 +/- 5 v 17.2 +/- 9). In summary, severe TBI warranting inpatient rehabilitation results in hypothalamic-pituitary-adrenal dysfunction. The extent of these abnormalities renders the assessment of cortisol secretion of limited value in making clinical judgments concerning the development of post-traumatic agitation or the management of that agitation by tricyclic therapy.