Objective. To evaluate the efficacy and safety of cordocentesis for pr
enatal diagnosis of various fetal diseases. Patients and Methods. Feta
l blood sampling was performed in 30 pregnancies from gestational week
19 to 36 for various prenatal diagnoses. We used Aloka SSD-650 real-t
ime ultrasound scanner with 3.5 mHz curvilinear transducer during the
procedure. A regular 22-gauge spinal needle was used for fetal blood s
ampling. Results. Twenty-five cordocenteses were successful at the fir
st attempt. A second attempt of cordocentesis succeeded in three cases
. In the remaining two cases, intrahepatic vein sampling and cardiocen
tesis were employed respectively. The duration of procedure was shorte
r than 15 minutes in 25 cases. 1-6 ml of fetal blood was obtained from
each case. Fetal blood was diluted with amniotic fluid in four sample
s. Maternal blood contamination never happened. No lethal complication
occurred in our series. Conclusion. It is believed that cordocentesis
is a safe and useful method for prenatal diagnosis, and intrahepatic
vein sampling is an alternative choice when cordocentesis has failed.