Abstract 1327 Poster Session IV, Tuesday, 5/4 (poster 224)

Thrombocytopenia is a common problem in neonatal intensive care units, but the kinetic mechanism responsible for thrombocytopenia in these patients is often unknown. In thrombocytopenic adults, it has been reported that circulating thrombopoietin (Tpo) concentrations are inversely related to the megakaryocyte mass, and therefore are useful in evaluating the responsible mechanism. To assess the utility of Tpo levels among neonates, we measured serial plasma Tpo by ELISA in 43 thrombocytopenic neonates. Their mean gestational age was 31±6 weeks, and their age at onset of thrombocytopenia was 8±15 days; 18 were thrombocytopenic at birth. The platelet count at study entry was 64,000±21,000/uL. Initial Tpo concentrations ranged from 59 to >2000 pg/mL, with a mean ±SD of 375 ±398 pg/mL (we previously found that 50 normal control neonates had Tpo <162 pg/mL). We found no correlation between Tpo concentration and gestational age, age at onset of thrombocytopenia, platelet count, or duration of the thrombocytopenia. However, the combination of a WBC <5000/uL plus thrombocytopenia, which was not associated with pregnancy-induced hypertension (PIH), was predictive of a poor prognosis. Of 8 infants with this combination, 2 died shortly after enrollment, and the others had a duration of thrombocytopenia significantly longer than non-neutropenic neonates (32±14 vs. 10±10 days, p=0.014). We categorized the neonates according to diagnosis: PIH (n=6), sepsis (n=9), alloimmune (n=5), DIC (n=3), and unexplained (n=20). 39 had initial Tpo levels that were either normal or only moderately elevated (<800 pg/mL). The other 4 (2 with sepsis, 1 with alloimmune, and 1 with unexplained thrombocytopenia) had levels > 1000 pg/mL, suggesting a reduced megakaryocyte mass. To directly compare the Tpo levels with marrow megakaryocyte numbers, we obtained marrow aspirates or biopsies on 14 of these thrombocytopenic neonates. Marrow megakaryocytes were normal in 5, increased in 2, and decreased in 7. The Tpo concentration among those with decreased megakaryocytes was 409±425 pg/mL, compared with 202 ±107 pg/mL among those with normal or increased megakaryocytes, but much overlap between groups was evident. We followed serial Tpo concentrations in 33 patients until resolution of the thrombocytopenia. Among the other 10, 6 expired and 4 left the hospital while still thrombocytopenic. In 84% of the patients followed, Tpo fell as the platelet count increased (p=0.0001). In conclusion, our results suggest that in thrombocytopenic neonates: 1) Tpo levels do not correlate with platelet count or gestational age, 2)Tpo levels decrease as the platelet counts increase, 3) The combination of neutropenia plus thrombocytopenia is highly predictive of a prolonged duration of thrombocytopenia, and 4)Tpo levels do not clearly indicate the underlying kinetic mechanism responsible for the thrombocytopenia.