What to expect when an ITP patient is expecting

In this issue of Blood,Guillet et al¹ prospectively studied the hematologic outcomes in pregnant patients with preexisting immune thrombocytopenia (ITP) and their nonpregnant matched controls.

“How will pregnancy affect my ITP?” As a practicing hematologist at the intersection of blood disorders and pregnancy, it is a common question asked by patients and driven by a combination of excitement and fear. People with chronic, stable ITP tend to gradually gain a sense of comfort when their condition has not required treatment for years; however, it is difficult for them to forget the worry and anxiety they felt with the initial diagnosis and during disease relapses leading to bleeding. These patients are concerned not only about the effect ITP has on their pregnancy outcomes, but about long-term implications that pregnancy can have in the prognosis of their ITP. As their treating hematologist, I share these concerns. The constant search for answers is typically met with even more questions. The art of my practice is carefully blending the best (available) evidence with patient expectations and preferences to allow for informed and shared decision-making. These are the times that I most regret taking the large randomized controlled trial data I learned for other conditions during medical training for granted!

Much of what is known about the outcomes of pregnant patients with ITP comes from retrospective studies which include individuals with a broad range of clinical characteristics. The reported risk of worsening ITP during pregnancy is approximately 30% and the risk of neonatal thrombocytopenia is of similar magnitude.^2-5 The heterogeneity in ITP duration, prior treatment, and platelet count at the time of pregnancy all pose challenges to the external validity of the studies. Additionally, although they provide a general framework to approach these patients clinically, prior studies do not address the question of whether a pregnancy (in a patient with known ITP) modifies the natural course of disease and changes their clinical outcomes.

In this study, the exposed cohort was matched by multiple characteristics including duration of ITP, status of their ITP at the time of pregnancy or study enrollment, and prior treatment (including rates of splenectomy in both groups). The non-exposed prospectively describes the natural history of ITP for nonpregnant females of reproductive age and provides us with the first reference cohort reported in the literature.

The authors focus on the primary end point of a worsening ITP event, defined as the combination of new bleeding, presence of severe thrombocytopenia, or initiation/modification of treatment (other than for delivery planning) during the study period. These outcomes are highly clinically relevant as they determine the time points at which patients require intervention to reduce risk of death or permanent disability. They show that pregnant patients, overall, do not differ from their nonpregnant counterparts in terms of worsening ITP around the time of gestation (including postpartum follow-up), supporting the notion that the biology of a patient’s ITP is not significantly modified by pregnancy. Although the presence of recurrent severe thrombocytopenia and increased frequency of treatments was higher in the exposed cohort, this difference could be related to the closer monitoring of patients during pregnancy along with variability in the clinician’s threshold to treat. However, bleeding events were not different.

The rates of pregnancy loss and preterm delivery in this study were not different from those of the general population. They also corroborate data from retrospective studies which indicate that 30% of patients required initiation or intensification of ITP treatment during pregnancy. Guillet et al found that 39% of the population studied received treatment before delivery. Treatment with corticosteroids and immunoglobulin in anticipation of delivery was successful in raising platelet counts which allowed for patients to have the option to receive neuraxial anesthesia. Beyond achieving adequate pain management, the ability to be awake for delivery is meaningful for patients and their families. This opportunity can be extended to most patients with appropriately timed treatment to optimize platelet counts before delivery.

Neonatal thrombocytopenia was present (as expected) in 27.2% of cases; it was associated with platelet counts less than <50 × 10⁹/L in the 3 months preceding birth and neonatal thrombocytopenia with a prior pregnancy. This study differed from prior reports that suggested prior splenectomy, but not severe thrombocytopenia, is associated with neonatal thrombocytopenia. The discrepancies serve as a reminder of the large knowledge gap that remains in this often forgotten field.

In summary, yes, there are multiple shortcomings in this exposed/non-exposed cohort study. The population size is unsurprisingly small and there are other confounding factors such that may introduce bias. However, I choose to see the glass half full instead of half empty. Guillet et al help us take a small but firm step in our ability to better care for patients with less common disorders. Their investigation is especially meaningful as it includes a population that has been historically excluded from clinical research and deserves equitable access to safe and effective medical care.

Conflict-of-interest disclosure: Juliana Perez Botero declares no competing financial interests.