Intended for healthcare professionals

Research Article

Opiate addiction in adult offspring through possible imprinting after obstetric treatment.

British Medical Journal 1990; 301 doi: (Published 10 November 1990) Cite this as: British Medical Journal 1990;301:1067
  1. B Jacobson,
  2. K Nyberg,
  3. L Grönbladh,
  4. G Eklund,
  5. M Bygdeman,
  6. U Rydberg
  1. Karolinska Institute, Stockholm, Sweden.


    OBJECTIVE--To test the hypothesis that opiate addiction in adults might stem partly from an imprinting process during birth when certain drugs are given to the mother. DESIGN--Retrospective study by logistic regression of opiate addicts with siblings as controls. SETTING--Stockholm, Sweden. SUBJECTS--200 Opiate addicts born in Stockholm during 1945-66, comprising 41 identified during interviews of probands for an earlier study; 75 patients whose death from opiate addiction had been confirmed during 1978-88; and 84 accepted for the methadone programme. 262 Siblings (controls) born in Stockholm during the same period, 24 of whom were excluded for drug addiction or being brought up outside the family. Birth records were unavailable for eight, leaving 230 siblings and 139 corresponding probands. MAIN OUTCOME MEASURES--Administration of opiates, barbiturates, and nitrous oxide (for greater than 1 h) to mothers of all subjects during labour within 10 hours before birth as a risk factor for adult opiate addiction. RESULTS--In subjects who had subsequently become addicts a significant proportion of mothers had received opiates or barbiturates, or both, compared with unmatched siblings (25% v 16%, chi 2 = 5.83, df = 1, p = 0.02), and these mothers had received nitrous oxide for longer and more often. After controlling for hospital of birth, order of birth, duration of labour, presentation other than vertex, surgical intervention, asphyxia, meconium stained amniotic fluid, and birth weight the relative risk for offspring subsequently becoming an adult opiate addict increased with the number of administrations of any of the three drugs. When the addicts were matched with their own siblings the estimated relative risk was 4.7 (95% confidence interval 1.8 to 12.4, p for trend = 0.002) for three administrations compared with when no drug was given. CONCLUSIONS--The results are compatible with the imprinting hypothesis. Therefore, for obstetric pain relief methods are preferable that do not permit substantial passage of drugs through the placenta.