PSYCHOSOMATIC MEDICINE, Volume 35, Number 2: Pages 121-128,
March-April 1973.

Plasma Cortisol Levels and
Behavioral States in Early Infancy



Plasma cortisol was determined on 40 full-term infants on the third day of life. The relationships between cortisol and sex, birthweight, Apgar scores, circumcision, and behavioral state were examined. Plasma cortisol was found to correlate most closely with behavioral state. In three infants plasma cortisol was assessed at weekly or fortnighly intervals from one to twelve weeks of age. Cortisol levels were found to vary more closely with behavioral state at the time the blood sample was taken than with changes in chronic irritability.


Blood samples were obtained from 40 full-term infants on the third day of life in conjunction with the blood obtained for routine screening for phenylketonuria. The behavioral states of the infants were observed during the half hour preceding the blood sample. Low cortisol levels were found to be associated with the sleep state and high levels with a period of fussiness or crying with little overlap between the two groups. No systematic relationship was found between cortisol and other parameters examined—sex, birthweight, birthweight for gestational age, recency of circumcision, or Apgar scores. One exception was that the infant whose Apgar score indicated fetal distress had an extremely high level of cortisol.

In three normal infants an attempt was made to examine the relationship between plasma cortisol and the varying levels of chronic irritability which commonly occur during the first three months of life. The infants were observed in their homes for an hour during which developmental tests were administered, the mother was interviewed, and a blood sample was obtained. Plasma cortisol levels were found to vary with ratings of behavioral state at the time the blood sample was taken. Cortisol levels were independent of chronic fussiness ratings in two cases but weer highly correlated (+0.78) in the third case. This high correlation was discounted because of the apparent overriding influence of immediate state.

The need for further investigation of the developmental course of the interrelationship between psychological factors and cortisol response is discussed.

From the Department of Psychiatry, University of Colorado Medical Center, 4200 East Ninth Avenue, Denver, Colorado 80220.
       Supported in part by Fluid Research Funds, University of Colorado Medical Center.
       The authors wish to thank Dr. Antonia Vernadakis of the Department of Psychiatry for her invaluable help in all phases of the study. Dr. Donald Stilson for statistical consultation and Ann Schrive and Judy Shearer for their technical assistance. We also gratefully acknowledge the cooperation of Dr. Denis Rodgerson and his assistants of the Pediatric Microchemistry Laboratory in obtaining the blood samples, and of Dr. Lula Lubchenko and her staff in the nursery.
       Received for publication January 13, 1972; final revision received April 24, 1972.
       Address for reprint requests: Katherine Tennes, Dept. of Psychiatry, University of Colorado Medical Center, 4200 East Ninth Ave., Denver, Colorado 80220.

(File revised 4 September 2006)

Return to CIRP library