Abnormal cortisol levels reveal clues in children at risk for psychosis



Abnormalities in the functioning of the hypothalamic-pituitary-adrenal axis might offer insight into understanding psychosis in schizophrenia, a recent study shows.

As "the primary system involved in coordinating the physiological response to stress," the HPA axis might interact with psychosocial stress "to trigger psychosis onset among individuals with an underlying vulnerability for the disorder," based on findings of abnormal cortisol levels in at-risk children, Alexis E. Cullen and her associates at King’s College London reported.

Their findings rely on abnormal daytime cortisol levels and correlations between cortisol levels and psychosocial stress and/or neurocognitive performance in children with a family history of schizophrenia and/or antecedents of schizophrenia (Psychoneuroendocrinology 2014;46:1-13 []).

The authors screened more than 1,300 children to identify 33 willing participants exhibiting all three antecedents of schizophrenia: a speech and/or motor delay or abnormality; a social, emotional, and/or behavioral problem; and a psychoticlike experience. Then they recruited 22 children with a first- or second-degree relative with schizophrenia, including 4 in the antecedents group. The comparison group included 40 typically developing children with no first-, second-, or third-degree relatives with a schizophrenia spectrum disorder and no antecedents.

About 33 months after identifying the participants, the children, aged 11-14, collected six saliva samples each day over 2 consecutive days, four in the hour after waking up, one at noon and one at 8 p.m. They also completed two psychosocial assessments – one on negative life events and one on school-related daily hassles – and memory and executive function assessments.

The researchers used Area Under the Curve calculations to compare cortisol levels in the hour after waking up (AUCi-CAR) and throughout the day (AUCg-DAY). The family history participants had significantly lower AUCi-CAR values (mean -33.8 nmol min/L) than the typically developing participants (mean 121.6 nmol min/L; standardized mean difference (d) = -.73; P = .01), but the antecedents of schizophrenia participants (mean 81.2 nmol min/L) were not significantly different from typically developing participants (d = -.19; P = .42). AUCg-DAY values were not significantly different between typically developing participants and either family history participants or antecedents of schizophrenia participants.

Both family history and antecedents of schizophrenia participants experienced more negative life events than the control participants, and the children with antecedents of schizophrenia experienced more daily hassles and subsequent distress than the controls. AUCi-CAR values for family history children were positively correlated with current and past distress relating to negative life events; those values were negatively correlated with distress at the time of past negative life events among typically developing children.

Both family history and antecedents of schizophrenia participants performed more poorly on neurocognitive tests than the typically developing participants. AUCi-CAR values had some positive correlations with subtests among family history and among antecedents participants, and AUCg-DAY values had negative correlations with subtests among family history participants.

Ms. Cullen and her associates said their study was limited by the small number of participant groups. "However, the non–help-seeking nature of this group may be considered a potential strength of the study, as it reduces the possibility that any HPA axis abnormalities are merely due to distress associated with emerging illness," they wrote.

The study was supported by the National Institute for Health Research, the Bial Foundation, a NARSAD Young Investigator Award, the British Medical Association, the Waterloo Foundation, and the Medical Research Council. No disclosures were reported.

[email protected]

Next Article: