OBJECTIVES: To determine whether total daily 6-sulfatoxymelatonin excretion and diurnal variation of melatonin secretion was maintained during the early phase of PICU admission through examination of the melatonin urinary metabolite, 6-sulfatoxymelatonin.

DESIGN: Exploratory prospective, observational study.

SETTING: Twelve-bed medical-surgical PICU of a Children's Hospital.

PATIENTS: Fifty children 3 months to 18 years old enrolled within 24 hours of PICU admission with access for urinary sampling.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN RESULTS: Urine samples were collected at 4-hour intervals for 24 hours and stored at -80C. 6-sulfatoxymelatonin was determined in duplicate by direct enzyme-linked immunosorbent assay. Patients were heterogeneous for diagnosis, had a mean age of 8.1 years (SD = 6.1 yr), and median (interquartile range) Pediatric Risk of Mortality III of 10 (4-13). Mean (SD) total daily 6-sulfatoxymelatonin production was 30.0 µg (25.6 µg) for the first 24 hours, which did not differ significantly from the means on days 2 (p = 0.56) or 3 (p = 0.29), and was similar to literature controls. Mean 6-sulfatoxymelatonin production for the population fit a periodic function well, with a reliable amplitude of 326 ng/hr and peak excretion from 04:00 to 08:00 (F = 4.4, p = 0.01), even when 6-sulfatoxymelatonin was corrected for body weight (F = 3.4, p = 0.03) and when sedation was included in the model (F = 3.95, p = 0.004). There was no significant correlation between lighting and 6-sulfatoxymelatonin excretion at any time period (R values: 0.11-0.25, p = 0.10-0.94). Mean 6-sulfatoxymelatonin excretion did not fit the model for a periodic function well for the subpopulations studied (sepsis [n = 18, F = 1.1, p = 0.32], respiratory failure requiring deep sedation [n = 10, F = 0.4, p = 0.66], and neurologic injury [n = 7, F = 0.6, p = 0.55]).

CONCLUSIONS: Total daily and diurnal variation of 6-sulfatoxymelatonin excretion is heterogeneously maintained early in pediatric critical illness. However, this may not hold true for specific diagnostic categories.