Growth discordancy in twin pregnancies: a risk factor not detected by measurements of biparietal diameter.

Among 460 twin pregnancies delivered at the University Central Hospital of Turku from 1970 to 1981, there were 41 (8.9%) with a weight difference of 25% or more between twins when calculated from the weight of the larger twin. The perinatal death rate in the first group (9.7%) was significantly higher (P less than .01) than the perinatal death rate (3.7%) in the group with the weight difference of less than 25%. The intrauterine mortality rate, in particular, was significantly increased (P less than .001) in the group with 25% or more difference being 6.5-fold when compared with the more difference being 6.5-fold when compared with the group with the lower weight difference. Among 271 twin pregnancies examined by ultrasound one to two weeks before delivery, there were 31 (11.4%) pairs of twins with a 3-mm or more difference in biparietal diameter, 11 (4.1%) with a 4-mm or more difference, and seven (2.6%) with a 5-mm or more difference. The sensitivity of measurements of biparietal diameter to detect the growth discordancy was 9 to 35%, the specificity 90 to 98%, and the positive predictivity 23 to 29%. This study indicates that a divergent growth pattern in twin pregnancy carries an elevated risk of intrauterine death, especially for the smaller twin. Measurement of biparietal diameter is not a method sensitive enough to detect these high-risk twin pregnancies.