The purpose of this investigation is to determine whether fetal thrombocytopenia is a risk factor for puncture site bleeding. Three groups of fetuses either with or at known risk for thrombocytopenia were identified from a prospectively maintained data base of 1,100 procedures: alloimmune thrombocytopenia (ATP, 29 cordocenteses); unexpected thrombocytopenia (53 cordoncenteses), and intravascular transfusion for fetal hemolytic anemia (194 transfusions). A fourth group (58 cordocenteses) included as a normal control consisted of all appropriately grown fetuses tested within the same gestational age range as those with ATP. In total, 276 fetal blood sampling procedures were included, of which 134 (49%) yielded a platelet count of < 120 x 10(3)/microliters and 38 (14%) a platelet count of < 50 x 10(3)/microliters. The first platelet count obtained from fetuses with ATP ranged from 1 x 10(3) to 159 x 10(3)/microliters. There was no correlation between the platelet count and bleeding time whether the analysis was limited to only pretreatment procedures or included all. Each fetus with unexpected thrombocytopenia was systemically ill. There was no correlation between platelet count and the duration of bleeding from the cord puncture site. Intravascular transfusion produced a significant decline in the platelet count (238 +/- 66 x 10(3) vs. 153 +/- 56 x 10(3)/microliters, p < 0.001). Twenty-nine percent of the post-transfusion platelet counts were below 120 x 10(3)/microliters. There was a significant negative correlation between the final platelet count and the duration of puncture site bleeding (r = -0.178, p = 0.03) independent of either the presence of hydrops, the initial or the increase in the umbilical venous pressure during transfusion.(ABSTRACT TRUNCATED AT 250 WORDS)