We have attempted to determine the accuracy of greyscale and color Doppler ultrasound in the differentiation of adnexal masses in pregnancy. The studied group included 2245 pregnant women from low risk population. Following criteria were evaluated: maximal diameter and volume of the tumor, echogenicity, presence of septa and papillary projections in grey scale sonography. Color Doppler analysis included blood vessel presence and arrangement and blood flow characteristics with the use of pulsatility (PI), resistive (RI) and systolic/diastolic (S/D) indices. Preoperative CA-125 serum levels were available in 11 patients. In 66 (2.94%) patients adnexal tumors were detected during routine ultrasound scan at the end of the first trimester. Twenty-seven masses (1.2%) persisted beyond 16 weeks of gestation and were subsequently surgically removed. Pathological diagnosis confirmed 19 serous cystadenomas, 4 endometriomas and 2 dermoids, one pedunculated myoma and one fibrothecoma. Mean size of the tumors was 79 Jmm (range: 43-245 mm), mean volume 166. lml (range: 30-1320 ml). Doppler indices values presented as mean, SD and range were as follows: PI = 1.26 +/- 0.71 (range: 0.57-3.84); RI = 0.61 +/- 0.15 (range: 0.33-0.89) and S/D = 2.62 +/- 0.98 (range: 1.17-4.91). Median serum concentration of CA-125 was 17 IU/ml (range: 8.4-1247 IU/ml). Only 3 of these women had elevated (> 35 IU/ml) levels: 2 endometriomas (344 IU/ml and 1247 IU/ml) and one myoma (37 IU/ml), respectively. Based on the sonographic findings two solid tumors were incorrectly classified as probably malignant (fibrothecoma and subserous myoma). Negative predictive value of ultrasound diagnosis in the studied population was therefore 92.6% (25 of 27). We conclude that although prenatal sonography has the potential to correctly classify most of adnexal masses, caution in risk assessment is needed especially when persistent solid tumor is found.