Nonstress fetal heart rate (FHR) recording was used as a primary test to detect fetal distress in 145 pregnant women with insulin-dependent diabetes. Testing was performed every second day beginning with the 32nd week of pregnancy and daily after the 34th week until delivery. One hundred eighteen (81.4%) had normal, nine (6.2%) suspicious, and 18 12.4%) pathologic FHR recordings. Poor metabolic control of diabetes was observed in 25 (17.2%) of the 145 pregnancies during the last trimester of pregnancy. Nine of these 25 women (35%) with poor metabolic control had a suspicious or pathologic FHR recording, which was significantly more frequent (P<.02) than in women with good metabolic control (18 of 120,15%). The mean value (± SD) of hemoglobin Arcduring the last trimester in diabetic women with pathologic FHR records was 7.63 ± 0.87%, which was significantly higher (P<.02) than in diabetic women with normal FHR records (6.91 ± 0.83%). None of the 145 fetuses monitored died in utero. It was concluded that no obvious iatrogenic morbidity was caused by early intervention in cases with pathologic FHR recordings.