Corticosteroids Plus Intravenous Immunoglobulin Compared with Corticosteroids for the Treatment of Immune Thrombocytopenia in Pregnancy: An Observational Multicenter Study in China

Immune thrombocytopenia (ITP) affects 1-10 in 10,000 pregnancies (American College of Obstetricians and Gynecologists, et al. Int J Gynaecol Obstet. 1999). In contrast to ITP in nonpregnant individuals, treatment of ITP in pregnancy is limited and no consensus has been reached regarding treatment. Although corticosteroids and intravenous immunoglobulin (IVIg) are acceptable treatments for ITP in pregnancy, the evidence from large-scale clinical trials is still insufficient. Here we retrospectively analyzed the data from 5 collaborative centers in China to compare the effectiveness of corticosteroids plus IVIg with that of corticosteroids alone for the treatment of immune thrombocytopenia in pregnancy.

Between January 2001 and July 2017, 521 women (572 pregnancies) were enrolled in this retrospective study. The diagnosis of ITP was confirmed in women previously diagnosed with ITP. Patients whose platelet counts remained above 70×10⁹/L during pregnancy and normalized in the postpartum period were deemed to have gestational thrombocytopenia and were excluded (Sun D, et al. Blood 2016). The median initial prednisone dose was 15 mg/d (interquartile range [IQR] 10-30 mg/d) in both the combined therapy group and the monotherapy group. IVIg was given concomitantly at a daily dose of 20g during the first 3-5 days in the combined therapy group. Another 3 cycles of IVIg were given every 2-4 weeks if maternal platelet response was observed (platelets ≥30×10⁹/L). Withdrawal of IVIg could be performed when platelets rose above 300×10⁹/L in <14 days. A total of 2 cycles of IVIg could be given for no platelet response (platelets <30×10⁹/L).

The intervention was not introduced in 340 pregnancies (340/572, 59%). Among the remaining 232 women with pregnancies, 105 (45%) were given prednisone plus IVIg as the initial treatment and 127 were given only prednisone. The maternal characteristics, including age, maternal pretreatment platelet count, bleeding score and more, did not differ between the two groups. No differences between the two groups were observed with regard to the maternal response to the initial treatment (including complete response [CR] and partial response [PR], prednisone plus IVIg 48% vs. prednisone 36%, p=0.08), nor were differences observed with regard to the proportion of maternal complications, including bleeding episodes in the antenatal period, platelet count at delivery, pre-delivery platelet transfusion, mode of delivery (vaginal delivery or cesarean section), abortion and premature delivery. For patients with an initial response, there was a significant difference between the time to response (TTR) of the corticosteroids plus IVIg group (median and range, 3 and 1-14 days) and the TTR of the corticosteroid group (7 and 2-18 days) (p <0.01). Notably, 62 of 81 pregnancies with no response to prednisone were given IVIg, and the response to the treatment was 34% (21/62). Moreover, of 324 women (357 pregnancies) with a history of ITP, the majority of pregnancies (255/357, 71%) did not require treatment. A review of 197 women (215 pregnancies) with newly diagnosed ITP during pregnancy revealed that most ITP (161/215, 74.8%) was first recognized in the first trimester. With regard to the neonatal characteristics, 599 neonates (572 pregnancies) were live births and none were intrauterine death. Among 389 neonates (389/599, 65%) whose cord blood platelet counts were measured, platelet counts<100×10⁹/L were observed in 118/389 (30%) neonates, and specifically in 18/155 neonates (12%) whose mothers were not treated. No fatal or severe neonatal hemorrhage was observed. The proportion of neonates with an Apgar score <7 at 5 minutes and neonatal platelet counts at birth did not differ between the groups. Neonatal cord platelet count compared with maternal platelet count at delivery indicated a lack of linear relation. Adverse events, including hyperglycemia requiring treatment, infection, and others, were reported in 19/105 (18%) of the prednisone plus IVIg group and in 31/127 (24%) of the prednisone group (p=0.28).

In conclusion, no better maternal platelet count response but a shorter TTR were shown in the prednisone plus IVIg group compared with the prednisone group in this observational study of ITP patients in pregnancy. Prospective studies are needed to better identify the effectiveness and safety of these regimens and to explore new therapeutic options.