Genetic counselling in a dizygotic twin pregnancy is complicated by the large number of possible pregnancy outcomes and by the conceivable differences in the parental valuation of these outcomes. We present the probability distributions of the pregnancy outcomes in dizygotic twin pregnancies for women from 35 to 45 years old without prenatal diagnosis and with transabdominal chorionic villus sampling (TA-CVS) or amniocentesis (AC), using data from the literature. TA-CVS always gives a higher probability of a favourable pregnancy outcome (the birth of one or two infants with a normal karyotype) than AC. For a 35-year-old woman, a 0.7 per cent risk of an unfavourable pregnancy outcome without prenatal diagnosis has to be weighed against the 2.1 per cent excess risk of loss of the entire pregnancy after TA-CVS. For a 45-year-old woman, a 10.2 per cent risk of an unfavourable pregnancy outcome without TA-CVS has to be balanced against a 4.4 per cent excess risk of pregnancy loss after TA-CVS. This study provides a quantitative tool for the support of individual parents with respect to the decision to undergo prenatal diagnosis in a dizygotic twin pregnancy.