Male gonadal dose of ionizing radiation delivered during X-ray examinations and monthly probability of pregnancy: a population-based retrospective study
1 INSERM, National Institute of Health and Medical Research, Unité U569 "Epidémiologie, Démographie et Sciences Sociales", IFR69, Le Kremlin-Bicêtre, France; INED, National Institute for Demographic Studies, Le Kremlin Bicêtre, France; University Paris-Sud, Faculté de Médecine, Le Kremlin Bicêtre, France
2 Institut de Veille Sanitaire, Département Santé-Travail, Saint-Maurice, France
BMC Public Health 2006, 6:55 doi:10.1186/1471-2458-6-55Published: 3 March 2006
Male gonadal exposure to ionizing radiation may disrupt spermatogenesis, but its influence on the fecundity of couples has been rarely studied. We aimed to characterize the influence of male gonadal dose of ionizing radiation delivered during radiodiagnostic on the monthly probability of pregnancy.
We recruited a random sample of women who retrospectively described 1110 periods of unprotected intercourse beginning between 1985 and 1999 and leading either to a live birth or to no pregnancy; their duration was censored after 13 months. The male partner answered a telephone questionnaire on radiodiagnostic examinations. We assigned a mean gonadal dose to each type of radiodiagnostic examination. We defined male dose for each period of unprotected intercourse as the sum of the gonadal doses of the X-ray examinations experienced between 18 years of age and the date of discontinuation of contraception. Time to pregnancy was analysed using a discrete Cox model with random effect allowing to estimate hazard ratios of pregnancy.
After adjustment for female factors likely to influence fecundity, there was no evidence of an association between male dose and the probability of pregnancy (test of homogeneity, p = 0.55). When compared to couples with a male gonadal dose between 0.01 and 0.20 milligrays (n = 321 periods of unprotected intercourse), couples with a gonadal dose above 10 milligrays had a hazard ratio of pregnancy of 1.44 (95% confidence interval, 0.73–2.86, n = 31).
Our study provides no evidence of a long-term detrimental effect of male gonadal dose of ionizing radiation delivered during radiodiagnostic on the monthly probability of pregnancy during the year following discontinuation of contraceptive use. Classification errors due to the retrospective assessment of male gonadal exposure may have limited the statistical power of our study.