Breast feeding and smoking hygiene: major influences on cotinine in urine of smokers' infants.

The determinants of urine cotinine levels were studied in a group of 101 infants aged 3 months, including 79 infants whose mothers were current smokers. At a pre-arranged home visit the infants' mothers completed an interviewer-administered questionnaire, and samples of maternal urine and breast milk and infants' urine were collected. Cotinine and nicotine levels were determined by gas-liquid chromatography. Infant urine cotinine levels ranged from 0 to 140 micrograms/l (0-1120 ng cotinine/mg creatinine). A linear dose response relation between mother's smoking rate and infant urine cotinine level was observed among breast-fed infants (r = 0.79, p less than 0.001). The relation was weaker among infants fed by both breast and bottle (r = 0.56, p = 0.01) and was not apparent among bottle-fed infants (r = 0.15, p = 0.16). In addition to mode of feeding and mother's smoking rate, mother's smoking "hygiene" (assessed by the reported frequency of smoking while feeding and with infant in same room) was independently associated with infant urine cotinine level. Father's smoking pattern and exposure to smoke outside the household did not relate significantly to infant cotinine levels. We conclude that when mothers smoke, breast feeding is the principal determinant of cotinine in infants' urine. It is likely that most of this cotinine comes from cotinine in mothers' breast milk, but further research is needed to establish how much nicotine is ingested by breast-fed infants of mothers who smoke, and to investigate possible health effects.