Thrombocytopenia in pregnancy

INTRODUCTION — Evaluation and management of thrombocytopenia during pregnancy and postpartum may be challenging because there are many potential causes, some directly related to the pregnancy and some unrelated. For many of the causes, there are no diagnostic laboratory tests. Management options may have the potential for serious complications for both mother and fetus and may require urgent decisions about delivery, and there may be concerns about fetal thrombocytopenia.

This topic reviews our approaches to determining the cause of thrombocytopenia in a pregnant patient and to management during pregnancy and delivery, which in some cases cannot wait for full diagnostic evaluation.

Additional information on the diagnosis and management of specific conditions that cause thrombocytopenia is presented in more detail separately:

●Immune thrombocytopenia (ITP) – (See "Immune thrombocytopenia (ITP) in adults: Clinical manifestations and diagnosis".)

●Drug-induced thrombocytopenia – (See "Drug-induced immune thrombocytopenia".)

●Preeclampsia – (See "Preeclampsia: Clinical features and diagnosis".)

●HELLP syndrome – (See "HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets)".)

●Disseminated intravascular coagulation – (See "Disseminated intravascular coagulation (DIC) during pregnancy: Clinical findings, etiology, and diagnosis".)

●Acquired, autoimmune thrombotic thrombocytopenic purpura (TTP) – (See "Diagnosis of immune TTP".)

●Hereditary TTP – (See "Hereditary thrombotic thrombocytopenic purpura (hTTP)".)

●Complement-mediated thrombotic microangiopathy (CM-TMA) – (See "Complement-mediated hemolytic uremic syndrome in children".)

DEFINITION AND INCIDENCE — Thrombocytopenia is defined as a platelet count below the lower limit of the normal range (typically, <150,000/microL). In most uncomplicated pregnancies, platelet counts remain within the normal range (150,000 to 450,000/microL). Platelet counts are slightly lower in twin compared with singleton pregnancies [1].

The normal changes in platelet counts during pregnancy were illustrated in a series of 7351 individuals at the Oklahoma University Medical Center for whom data were available on platelet counts during pregnancy and at delivery from 2011 to 2014 [1]. This represented approximately one-half of the deliveries at this center. Only one pregnancy from each person was included. Findings were as follows:

●Mean platelet counts in uncomplicated singleton pregnancies decreased progressively throughout pregnancy and increased after delivery. Platelet counts in the first trimester were already significantly lower than platelet counts of nonpregnant people, which came from data in the National Health and Nutrition Examination Survey (NHANES):

•Nonpregnant – 273,000/microL

•First trimester – 251,000/microL

•Second trimester – 230,000/microL

•Third trimester – 225,000/microL

•Delivery – 217,000/microL

•Postpartum (approximately seven weeks) – 264,000/microL

Mean platelet counts in uncomplicated twin pregnancies were lower, with a similar pattern (first, second, third trimester and delivery: 240,000, 221,000, 217,000, and 202,000/microL, respectively); however, there were fewer than 100 twin pregnancies. These findings are summarized in the table (table 1).

●At delivery, the distribution of platelet counts in uncomplicated singleton pregnancies was as follows:

•≥150,000/microL – 4115 (90 percent)

•125,000 to 149,000 – 293 (6.4 percent)

•100,000 to 124,000 – 126 (2.7 percent)

•80,000 to 99,000 – 27 (0.6 percent)

•60,000 to 79,000 – 4 (0.1 percent)

•<60,000 – 0

These data suggest that mild decreases in platelet count occur during all uncomplicated pregnancies. The mechanism may involve pooling of platelets in the splenic and placental circulation (see 'Gestational thrombocytopenia (GT)' below). Platelet counts <100,000/microL occur in <1 percent of uncomplicated pregnancies [1]. Therefore:

●Platelet counts <100,000/microL should be investigated for an etiology other than gestational thrombocytopenia (GT; ie, platelet counts <100,000/microL should not be attributed to GT unless alternative etiologies have been excluded). (See 'Determining the likely cause(s)' below.)

●Platelet counts between 100,000 and 150,000/microL at any time during pregnancy, including the first trimester, are likely to be GT and may need no evaluation (table 2), unless other factors are present (see 'Gestational thrombocytopenia (GT)' below). Examples of other factors that would suggest the need for additional evaluation include thrombocytopenia prior to pregnancy or other diagnoses such as preeclampsia [1].

LIST OF CAUSES — The causes of thrombocytopenia during pregnancy vary with the duration of gestation, from early in the first trimester through delivery, the severity of thrombocytopenia, and the patient's clinical status [2,3].

GT and other causes of thrombocytopenia in pregnant patients are discussed in detail on the following sections, and selected causes other than GT are listed in the table (table 3).

Gestational thrombocytopenia (GT) — Gestational thrombocytopenia (GT), also called incidental thrombocytopenia of pregnancy, is a self-limited condition that requires no additional evaluation or treatment [4-6]. GT accounts for the vast majority of cases of thrombocytopenia discovered during pregnancy, and almost all cases of thrombocytopenia in uncomplicated pregnancies. GT may occur during the first trimester, but it becomes more common as gestation progresses, with the highest frequency at the time of delivery, when the frequency is 5 to 10 percent [1,3,7,8].

GT is typically characterized by the following [1,9]:

●Most common at delivery, but can occur at any time during pregnancy.

●Mild thrombocytopenia. (In 99 percent of people, the platelet count is ≥100,000 /microL.)

●No increased bleeding or bruising.

●No associated abnormalities on complete blood count (CBC).

●No fetal or neonatal thrombocytopenia.

GT is a diagnosis of exclusion. The diagnosis of GT is accepted if the person has mild thrombocytopenia (platelet count 100,000 to 150,000/microL), especially during late pregnancy and at delivery, with no other associated findings on CBC or physical examination. In our experience, platelet counts below 100,000/microL are extremely unlikely to be due to GT [1]. The likelihood of GT decreases dramatically with lower platelet counts, as illustrated in the table (table 2).

GT resolves postpartum, but in some instances, the return to a normal platelet count requires more than six weeks. A normal prepregnancy platelet count is helpful but not always available. A history of mild thrombocytopenia during a previous pregnancy supports the diagnosis of GT because the risk of recurrent GT is 14-fold greater among individuals who have had previous GT than among those who have not had previous GT [1].

GT requires no treatment and no change of normal prenatal care and management of delivery. No diagnostic testing is necessary because a platelet count >100,000/microL causes no risk for the mother or the fetus.

The mechanism(s) of GT has not been documented, but it may be assumed to be a physiologic adaptation of pregnancy related to the increased plasma volume, pooling or consumption of platelets in the placenta, or other physiologic changes that occur in uncomplicated pregnancies [1]. The placenta has many vascular characteristics in common with the spleen, a major site of physiologic platelet sequestration. An analysis of placental histology following 40 scheduled cesarean deliveries found that platelets were present in many areas in the perivillous fibrinoid (picture 1), supporting the concept that platelet sequestration and consumption in the placenta plays a role in GT [10].

Immune thrombocytopenia (ITP) — Immune thrombocytopenia (ITP) occurs in approximately 1 to 3 in 10,000 pregnancies; platelet counts <50,000/microL only affect a subset of these [11]. This is approximately 10-fold greater than the incidence of ITP in the general population, estimated to be 3 in 100,000 adults [12]. The increased frequency of ITP diagnosed during pregnancy may reflect more frequent CBC testing during pregnancy, the increased incidence of autoimmune disorders in females, and possibly decreased platelet counts due to the GT [1]. (See "Immune thrombocytopenia (ITP) in adults: Clinical manifestations and diagnosis", section on 'Epidemiology'.)

ITP may occur during any trimester, or the diagnosis may be known prior to the pregnancy. The severity of thrombocytopenia is variable, and the platelet count will decrease during the pregnancy, as it does in all pregnancies as part of normal physiology. (See 'Definition and incidence' above.)

For individuals with a prior history of ITP, more severe thrombocytopenia during pregnancy has been reported and attributed to the greater severity of the ITP [13]. However platelet counts of individuals with ITP decrease at the same rate as in uncomplicated pregnancies, and their platelet counts are 15 to 20 percent lower at delivery, similar to uncomplicated pregnancies [1]. Therefore, a decreased platelet count during pregnancy may not be related to greater severity of the ITP.

As in nonpregnant individuals, the risk of bleeding is greater with platelet counts <20,000 to 30,000/microL, although there is no absolute platelet count threshold above which bleeding does not occur. This was illustrated in a long-term study involving 119 pregnancies in 92 individuals over the course of 11 years [14]. Most pregnancies were uneventful, but there was moderate bleeding in 21 patients (18 percent; described as epistaxis or mucous membrane bleeding) and severe bleeding in 4 patients (3 percent; described as hematuria or gastrointestinal). Most deliveries were vaginal, and one-fourth of the infants had thrombocytopenia. (See 'Neonatal testing' below.)

A 2023 report described the course of 131 pregnant patients with ITP compared with 131 nonpregnant females with ITP who were matched by history of splenectomy, ITP status, and duration of disease and found no increased risk of bleeding in the pregnant versus the nonpregnant group [15].

The diagnosis of ITP is based only on the exclusion of other causes of thrombocytopenia. Therefore, in a pregnant patient with mild thrombocytopenia (platelet count 100,000 to 150,000/microL), GT and ITP cannot be distinguished. The diagnosis of GT is much more likely than ITP in such patients because the frequency of GT is 100-fold greater than the frequency of ITP during pregnancy. (See "Immune thrombocytopenia (ITP) in adults: Clinical manifestations and diagnosis".)

Preeclampsia with severe features/HELLP (PE/HELLP) — The terminology for "preeclampsia with severe features" and "HELLP syndrome" (co-occurrence of microangiopathic hemolytic anemia, elevated liver function tests, and low platelet count) is evolving.

●Preeclampsia – Preeclampsia is common, occurring in approximately 5 percent of pregnancies; it is manifested by new onset hypertension and proteinuria or new-onset hypertension and end-organ dysfunction with or without proteinuria after 20 weeks of gestation (mid-second trimester) in a previously normotensive individual (table 4). Preeclampsia is associated with a platelet count <100,000/microL in only 7 percent of cases, and with severe thrombocytopenia (platelet count <60,000/microL) in 3 percent (table 5) [16]. (See "Preeclampsia: Clinical features and diagnosis", section on 'Definitions/diagnostic criteria'.)

●Preeclampsia with severe features – Preeclampsia with severe features is defined in the table (table 6). Severe features include more severe hypertension, severe headache and/or visual symptoms, liver function abnormalities and epigastric pain, and thrombocytopenia (platelet count <100,000/microL); this affects 1 percent of individuals with preeclampsia [8,17]. Seizures change the diagnosis to eclampsia. (See "Eclampsia".)

Preeclampsia with severe features is also associated with microangiopathic hemolytic anemia (MAHA) with one or more of the following: schistocytes on the blood smear, increased lactate dehydrogenase (LDH), increased bilirubin, and decreased haptoglobin; these features also occur in thrombotic microangiopathy syndromes such as thrombotic thrombocytopenic purpura (TTP). (See 'Thrombotic microangiopathy (TMA)' below.)

Management is discussed separately. (See "Preeclampsia with severe features: Delaying delivery in pregnancies remote from term" and "Preeclampsia: Antepartum management and timing of delivery".)

●HELLP – The name HELLP syndrome has been used for decades; this syndrome may not require the presence of hypertension or proteinuria. By definition, HELLP syndrome includes thrombocytopenia and evidence of microangiopathic hemolytic anemia [18]. HELLP is generally managed with delivery as soon as possible. (See "HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets)".)

●PE/HELLP – Together, "preeclampsia with severe features" and "HELLP" describe an acutely ill pregnant individual with thrombocytopenia who requires delivery to halt the disease process. There is substantial overlap between these syndromes, but it is possible to have HELLP without hypertension (thus, by definition, the patient does not have preeclampsia) and it is also possible to have preeclampsia with severe features without all of the manifestations of HELLP.

Because preeclampsia with severe features, eclampsia, and the HELLP syndrome have overlapping clinical features, we use the combined term "preeclampsia with severe features/HELLP" (PE/HELLP) to include preeclampsia with severe features, eclampsia, and HELLP syndrome [19]. This combined term is helpful when evaluating patients with platelet counts <100,000/microL. Distinguishing these three syndromes provides no clinical benefit for a hematologist focused on diagnosing the etiology of thrombocytopenia. In contrast, the distinction of these syndromes has significance for obstetric care (particularly timing of delivery).

Although 20 weeks of gestation is the established time when PE/HELLP can occur, most occurrences of PE/HELLP are much later during pregnancy (table 7) [16]. Rarely, the diagnosis is made in the first 24 to 48 hours after delivery. Among 818 occurrences of PE/HELLP at the Oklahoma University Medical Center in which gestational age was reported, 99 percent occurred at ≥25 weeks of gestation, 93 percent occurred at ≥30 weeks of gestation and 75 percent occurred at ≥35 weeks of gestation (table 7) [16].

Because the characteristic clinical features of HELLP syndrome (microangiopathic hemolytic anemia, thrombocytopenia) are also the characteristic clinical features of TTP, measurement of ADAMTS13 activity may be important to exclude the diagnosis of TTP and support the diagnosis of PE/HELLP. Measurement of ADAMTS13 activity is especially important if the gestational age is <25 to 30 weeks, if symptoms of neurologic ischemia are prominent, or if the symptoms do not resolve promptly following delivery.

It is necessary to gauge the relative likelihood of these conditions that are often treated by delivery versus other conditions for which delivery does not result in improvement and other urgent treatments are required, such as sepsis related to a non-obstetric infection, TTP, or complement-mediated thrombotic microangiopathy (CM-TMA). (See 'Management decisions' below.)

DIC — Disseminated intravascular coagulation (DIC) is a systemic process in which coagulation and fibrinolysis become activated within the vasculature, often massively. This can lead to depletion of clotting factors and platelets, with severe bleeding as well as increased risk of thrombosis. There is always an underlying cause that initiates systemic activation of the clotting cascade. Causes of DIC in pregnancy include placental abruption, retained dead fetus, amniotic fluid embolism, septic abortion, and others. Patients may have severe hemorrhage and/or diffuse oozing. There may be MAHA with schistocytes on the blood smear. Typically, the prothrombin time (PT) and activated partial thromboplastin time (aPTT) are prolonged, the fibrinogen is low, and the plasma D-dimer is elevated.

Management of DIC involves identifying and treating the underlying cause; in some cases, this may require delivery (eg, retained dead fetus, abruption), whereas in others, the primary treatment may involve antibiotics (eg, sepsis from a non-obstetric infection) or other therapies (eg, therapy for a malignancy). Transfusions may be needed while bleeding is being controlled. (See "Disseminated intravascular coagulation (DIC) during pregnancy: Management and prognosis".)

Acute fatty liver of pregnancy — Acute fatty liver of pregnancy (AFLP) is an uncommon form of liver injury that typically occurs in the third trimester. The major clinical findings relate to fatty infiltration of the liver and include nausea, vomiting, and abdominal pain. The platelet count may be decreased. If liver function is severely impaired, the PT and aPTT will be prolonged, and the fibrinogen may be low. AFLP is discussed in more detail separately. (See "Acute fatty liver of pregnancy".)

Thrombotic microangiopathy (TMA) — Thrombotic microangiopathies (TMAs) include a number of conditions in which platelet microthrombi form in small vessels and lead to organ damage.

There are three principal primary TMAs:

●Thrombotic thrombocytopenic purpura (TTP)

●Complement-mediated TMA (CM-TMA)

●Shiga toxin-mediated hemolytic uremic syndrome (ST-HUS).

TTP and CM-TMA can be hereditary or acquired and their frequency may be increased with pregnancy. ST-HUS is an infectious disease that occurs primarily in children. Its frequency is not associated with pregnancy. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)".)

Almost all patients with hereditary TTP have an acute, severe episode of TTP with pregnancy [20,21]. The table illustrates differences in the timing of presentation between immune TTP versus hereditary TTP (table 7). Immune TTP occurs equally across all gestational ages and postpartum, whereas hereditary TTP most often occurs later in the pregnancy. [20]. For patients in remission from immune TTP, pregnancy may also increase the risk of relapse, but the frequency is very low [22]. For patients with CM-TMA, pregnancy and the postpartum period are also risks for triggering an acute episode of severe acute kidney injury [23].

The only defining clinical features of the TMA syndromes are MAHA, which is inferred from the presence of schistocytes on the peripheral blood smear (picture 2), and thrombocytopenia, which can be severe (table 7). Severe acute kidney injury is the most important presenting abnormality in patients with CM-TMA. Other findings that may be present in all TMA syndromes include neurologic symptoms ranging from mild headache to seizures and transient focal abnormalities. Although fever has been described as a manifestation of TTP, this was most commonly seen in the era prior to the use of therapeutic plasma exchange (TPE); in the post TPE-era, fever almost always indicates a systemic infection [24].

Persistence or worsening of thrombocytopenia, MAHA, and ongoing end-organ injury for more than three days after delivery is considered by some experts to be strongly suggestive of a TMA, since preeclampsia with severe features almost always begins to recover before this time.

Despite the overlapping clinical presentations, different TMAs have differing pathophysiologies and require different treatments. TMAs are potentially life-threatening, so it is important to make the most accurate diagnosis possible and, in many cases, to treat presumptively for the most serious of these while pursuing diagnostic testing. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Overview of primary TMA syndromes'.)

Many pregnancies complicated by TMA result in the birth of healthy term infants. However, intrauterine fetal death may occur from placental infarction caused by thrombosis of the decidual arterioles. These complications are discussed separately. (See "Immune TTP: Initial treatment", section on 'Immune TTP during pregnancy' and "Hereditary thrombotic thrombocytopenic purpura (hTTP)", section on 'Management of pregnancy'.)

TTP — Thrombotic thrombocytopenic purpura (TTP) is caused by severely reduced activity of ADAMTS13, a protease that cleaves the very large von Willebrand factor (VWF) multimers secreted by endothelial cells. ADAMTS13 may be reduced because of a neutralizing autoantibody (acquired, autoimmune TTP) or an inherited mutation in the ADAMTS13 gene (hereditary TTP). (See "Diagnosis of immune TTP" and "Hereditary thrombotic thrombocytopenic purpura (hTTP)".)

A significant proportion of patients with hereditary TTP have their first presentation of disease during pregnancy, but acquired TTP is more common than hereditary TTP and thus more likely in a pregnant patient without a family history of TTP. A 2012 series of 42 patients with a first episode of TTP during pregnancy found three-fourths were due to acquired TTP and one-fourth to hereditary TTP [20]. A 2014 series of 35 individuals from the United Kingdom TTP registry who presented with new-onset TTP during pregnancy documented acquired TTP in 12 (34 percent) and hereditary TTP in 23 (66 percent) [25].

Features suggestive of TTP include thrombocytopenia and schistocytes combined with severe neurologic findings (although one-half of patients with TTP have no or only minor neurologic abnormalities) and absence of features of DIC (eg, absence of coagulation abnormalities). Acute kidney injury (AKI) rarely occurs in patients with acquired TTP. Acquired TTP occurs equally during all trimesters and postpartum (table 7) [20,26].

The diagnosis of TTP relies on a thorough clinical assessment combined with a finding of ADAMTS13 activity <10 percent. Rarely, patients may present with ADAMTS13 activity ≥10 percent [27,28]. As in nonpregnant patients, the clinical assessment is vital to the diagnosis, because results of ADAMTS13 activity testing often are not immediately available. In contrast to severe ADAMTS13 deficiency, moderately reduced ADAMTS13 activity (eg, activity between 10 and 50 percent) can be seen in many systemic disorders and is not specific for (or indicative of) TTP. ADAMTS13 activity decreases during pregnancy, related to the increased levels of VWF, and may fall below 50 percent in many pregnant patients near term (but not below 10 percent, which is typical of TTP) [29]. These issues are discussed in more detail separately. (See "Diagnosis of immune TTP", section on 'Reduced ADAMTS13 activity' and "Diagnosis of immune TTP", section on 'Evaluation and diagnosis' and "Pathophysiology of TTP and other primary thrombotic microangiopathies (TMAs)", section on 'TTP pathogenesis'.)

The clinical course of TTP is not affected by delivery, intravenous immune globulin (IVIG), or eculizumab. The most important treatment is urgent therapeutic plasma exchange (TPE), which removes the autoantibody to ADAMTS13 and supplies functional ADAMTS13 in the replacement plasma. For patients known to have hereditary TTP, plasma infusion is sufficient for treatment because it supplies ADAMTS13.

For patients with hereditary TTP, we begin regular plasma infusions when pregnancy is first documented, typically at gestational age 5 to 10 weeks. The amount of plasma and frequency of infusions is adjusted to maintain the patients' normal platelet count. The frequency of infusions is typically increased during the late second and third trimester. We continue plasma infusions for six weeks postpartum. (See "Hereditary thrombotic thrombocytopenic purpura (hTTP)", section on 'Management'.)

Recombinant ADAMTS13, when available, will overcome the complex logistics of frequent plasma infusions and challenges of large volume administration that are required to maintain the target ADAMTS13 activity during pregnancy.

Platelet transfusion should be reserved for treatment of severe bleeding in a patient with TTP due to the potential increased risk of thrombosis, but platelets should not be withheld in a bleeding patient due to concerns about this risk. (See "Hereditary thrombotic thrombocytopenic purpura (hTTP)", section on 'Management of pregnancy' and "Immune TTP: Initial treatment", section on 'Therapeutic plasma exchange'.)

With appropriate monitoring and therapy, outcomes for hereditary TTP are almost always successful. In a review of cases from the United Kingdom TTP registry, 10 individuals known to have hereditary TTP who were actively monitored and treated during pregnancy had 15 live births, and six individuals with acquired TTP had six pregnancies with live births; there were no maternal or fetal deaths, and all deliveries were in the third trimester [25].

Complement-mediated TMA — Complement-mediated TMA (CM-TMA), also called complement-mediated HUS or "atypical HUS," is a disorder in which patients have increased activation of complement on endothelial cells. We avoid the term atypical HUS because it is nonspecific. Patients with dysregulated complement can develop microthrombi in small vessels throughout the vasculature; the kidney is often affected. Complement may be dysregulated because of a neutralizing autoantibody or an inherited mutation in a complement regulatory gene such as complement factor H (CFH). (See "Acute kidney injury in pregnancy" and "Complement-mediated hemolytic uremic syndrome in children", section on 'Genetic variants' and "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS", section on 'Causes of CM-TMA'.)

Patients with hereditary CM-TMA may have their first presentation of disease during pregnancy or postpartum. CM-TMA may present similarly to HELLP syndrome and TTP, and the differentiation among CM-TMA, HELLP, and TTP can often be challenging. Key diagnostic features include microangiopathic hemolytic anemia, thrombocytopenia, and increasing serum creatinine.

●Most patients with CM-TMA have severe kidney injury requiring dialysis. (See "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS".)

●Severe acute kidney injury rarely occurs in patients PE/HELLP [30] or TTP [31].

●ADAMTS13 activity is not severely deficient in CM-TMA (activity is ≥10 percent) and stool studies are negative for Shiga toxin-producing organisms.

●Patients with CM-TMA may be thought to have HELLP syndrome until kidney function worsens or fails to improve following delivery.

●Limited data suggest that the platelet count is not severely decreased in CM-TMA, and overt neurologic abnormalities are rare.

The clinical course of CM-TMA is not affected by delivery. In fact, the postpartum period may be the time of greatest risk for the occurrence of CM-TMA [23,32]. The most important treatment is anti-complement therapy (eg, eculizumab) to stop the process, along with supportive care for kidney failure that may include dialysis. Anti-complement therapy should be initiated promptly if CM-TMA is considered the most likely diagnosis because end-stage kidney disease may occur rapidly if the process goes unchecked. Additional information about anti-complement therapies, including therapies under development, is presented separately. (See "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS", section on 'Terminal complement blockade' and "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS", section on 'Pregnancy or postpartum'.)

Fetal effects of anti-complement therapy in CM-TMA have not been well studied. However, experience with the use of eculizumab in paroxysmal nocturnal hemoglobinuria (PNH) does not appear to show evidence of increased fetal risks. (See "Treatment and prognosis of paroxysmal nocturnal hemoglobinuria", section on 'Pregnancy'.)

ST-HUS — Shiga toxin-mediated hemolytic uremic syndrome (ST-HUS) is a disorder in which enteric infection with an organism that produces the toxin (eg, enterohemorrhagic Escherichia coli, Shigella); the kidneys are commonly affected. The mechanism(s) by which the toxin produces a TMA are incompletely understood. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Overview of primary TMA syndromes'.)

Patients with ST-HUS typically have severe abdominal pain and diarrhea, which is typically overtly bloody, as their major symptom. There may be a history of eating improperly prepared food or a recent outbreak, but almost all cases are sporadic. MAHA and thrombocytopenia are present and often accompanied by renal insufficiency. Fever and neurologic symptoms may also be present but are not always seen. These features are similar to those seen in children, in whom ST-HUS is much more common.

The diagnosis of ST-HUS is made based on clinical features, MAHA, thrombocytopenia, and demonstration of an implicated diarrheal organism. Stool testing for Shiga toxin-secreting bacteria (eg, E. coli 0157:H7, Shigella dysenteriae) and for Shiga toxin itself should be performed. Most patients will have been tested for and found to have normal or only mildly reduced ADAMTS13 activity and normal or nonspecifically reduced complement levels. Management involves supportive care, similar to children. (See "Treatment and prognosis of Shiga toxin-producing Escherichia coli (STEC) hemolytic uremic syndrome in children".)

Other causes of thrombocytopenia — Some conditions associated with thrombocytopenia are discovered incidentally during the pregnancy. These include systemic lupus erythematosus (SLE), medications that cause immune thrombocytopenia, heparin-induced thrombocytopenia (HIT), cancer, infection, liver disease, hypersplenism, deficiency of vitamin B12/folate/copper, or inherited platelet disorders. Most of these causes of thrombocytopenia are rare during pregnancy, with the exception of SLE, which may be more common than several of the conditions discussed above such as TMAs. These are discussed in separate topic reviews:

●SLE – Pregnancy is associated with risk for acute "flares" of SLE. These flares may be associated with thrombocytopenia as well as other cytopenias and/or manifestations of SLE. (See "Hematologic manifestations of systemic lupus erythematosus", section on 'Thrombocytopenia'.)

●Antiphospholipid syndrome – Antiphospholipid syndrome (APS) is an autoimmune disorder characterized by arterial and venous thromboembolism and pregnancy morbidity. A small subset of these individuals has catastrophic APS (CAPS), with widespread thrombotic disease. (See "Catastrophic antiphospholipid syndrome (CAPS)" and "Clinical manifestations of antiphospholipid syndrome".)

●Medications

•Drug-induced ITP – (See "Drug-induced immune thrombocytopenia", section on 'Commonly implicated drugs'.)

•HIT – (See "Use of anticoagulants during pregnancy and postpartum", section on 'Suspected HIT' and "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Evaluation'.)

•Drug-induced TMA (eg, from quinine) – (See "Drug-induced thrombotic microangiopathy (DITMA)".)

●Cancer – (See "Diagnostic approach to thrombocytopenia in adults", section on 'Postoperative or hospitalized'.)

●Infections

•Hepatitis C virus (HCV) – (See "Extrahepatic manifestations of hepatitis C virus infection".)

•Human immunodeficiency virus (HIV) – (See "HIV-associated cytopenias".)

•Other infectious organisms – (See "Diagnostic approach to thrombocytopenia in adults", section on 'Postoperative or hospitalized'.)

●Liver disease – (See "Hemostatic abnormalities in patients with liver disease", section on 'Thrombocytopenia and platelet dysfunction' and "Acute fatty liver of pregnancy" and "HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets)".)

●Deficiency of vitamin B12, copper, or (extremely rarely) folate – (See "Treatment of vitamin B12 and folate deficiencies" and "Clinical manifestations and diagnosis of vitamin B12 and folate deficiency".)

●Inherited disorders

•Inherited platelet disorders – (See "Inherited platelet function disorders (IPFDs)".)

•Type 2B von Willebrand disease (VWD) – In type 2B VWD, the normal increase in von Willebrand factor (VWF) that occurs during pregnancy can lower the platelet count, since the dysfunctional VWF binds to platelets. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Abnormalities in the CBC and coagulation tests'.)

DETERMINING THE LIKELY CAUSE(S)

Overview of the evaluation — In many cases, pregnancy imposes additional urgency on determining the cause of thrombocytopenia and additional management decisions related to the potential for complications that may affect the patient and the fetus. The diagnostic evaluation and interventions for the most likely cause(s) of the findings must often be performed simultaneously. Early involvement of experts in maternal-fetal medicine, hematology, and other appropriate specialists (eg, nephrology when renal failure is present; infectious diseases when sepsis is suspected) is advised.

Our approach to distinguishing among these causes, as described in the following sections, is consistent with a practice guideline on immune thrombocytopenia (ITP) from the American Society of Hematology [33]. Additional information concerning drug-induced thrombocytopenia and drug-induced thrombotic microangiopathy can be found on a website maintained and updated by Dr. James N George at the University of Oklahoma Health Sciences Center: www.ouhsc.edu/platelets [34].

As noted above, our approach to the evaluation takes into account the severity of thrombocytopenia, clinical presentation, and trimester (table 7). Helpful information includes the following:

●Course of the pregnancy so far, including presence or absence of complications

●Symptoms of infection such as fever and chills

●New daily medications within the past three weeks, or occasional medications taken immediately before symptoms occurred

●Personal or family history of excessive bleeding, bruising, pregnancy complications, or known thrombotic microangiopathy (TMA) syndrome

●Systemic lupus erythematosus (SLE) or other autoimmune disorder

●History of liver disease

●Timing of the drop in platelet count (which trimester, how rapidly)

●Presence of anemia more severe than expected for the stage of pregnancy

●Abnormalities of the peripheral blood smear, such as abnormal white blood cells or nucleated red blood cells

Isolated, asymptomatic thrombocytopenia; platelets >100,000/microL — Almost all pregnancies with relatively mild or incidentally discovered thrombocytopenia (eg, platelet count between 100,000 and 150,000/microL) without other cytopenias or major clinical findings will be attributable to gestational thrombocytopenia (GT). The only other common cause to consider may be ITP; however, ITP with mild thrombocytopenia requires no treatment and therefore the management is observation, as for GT.

Platelet counts <100,000/microL should be investigated for an etiology other than GT. Other, uncommon causes of isolated thrombocytopenia include conditions that may have preceded the pregnancy or developed during the pregnancy but remained clinically silent, such as SLE, inherited thrombocytopenia, bone marrow disorders such as myelodysplasia, HIV, liver disease, hypersplenism, nutrient deficiencies (eg, vitamin B12, folate, copper), and type 2 von Willebrand disease (VWD). These conditions account for <1 percent of cases. (See 'Other causes of thrombocytopenia' above.)

Evaluation for these conditions includes a personal and family history for bleeding disorders; review of the complete blood count (CBC) for other cytopenias and morphologic abnormalities in white blood cells, red blood cells, and platelets.

Pseudothrombocytopenia due to in vitro platelet clumping can be misinterpreted as true thrombocytopenia. Pseudothrombocytopenia should be excluded by reviewing the blood smear and, if necessary, repeating the platelet count using a tube containing citrate or heparin rather than EDTA as the anticoagulant. This and other testing for causes of thrombocytopenia not directly related to pregnancy are discussed separately. (See "Diagnostic approach to thrombocytopenia in adults".)

HIV testing is appropriate for all patients if not done already. Testing early in each pregnancy is routine in the United States. (See "Prenatal care: Initial assessment", section on 'HIV' and "Screening and diagnostic testing for HIV infection", section on 'Routine screening'.)

As noted above, it is not possible (or necessary) to distinguish between GT and ITP during pregnancy as long as the thrombocytopenia is mild (platelet count between 100,000 and 150,000/microL). Both of these conditions are diagnosed by excluding other causes of thrombocytopenia; there are no laboratory tests for either condition. GT can present at any time during pregnancy, but because CBCs are more likely to be performed later in the pregnancy, GT is more often recognized near term or at delivery. GT resolves after delivery, but it may require more than six weeks. ITP is often present before, during, and after the pregnancy, and the degree of thrombocytopenia is variable. If the platelet count drops below 100,000/microL, then ITP becomes the more likely diagnosis. If the history and/or examination suggest a cause other than GT or ITP, appropriate testing should be obtained. (See 'List of causes' above.)

After delivery, GT will resolve and ITP typically will persist, allowing a presumptive diagnosis to be made in most cases. The value of this is that patients with GT do not require any additional monitoring until or unless they become pregnant again. GT typically recurs with subsequent pregnancies, so it should be anticipated. If thrombocytopenia persists after pregnancy (eg, at the six-week follow-up), then ITP becomes the likely diagnosis. A less common condition such as an inherited platelet disorder may also be responsible. The evaluation, differential diagnosis, and management of ITP are discussed in detail separately. (See "Diagnostic approach to thrombocytopenia in adults" and "Immune thrombocytopenia (ITP) in adults: Clinical manifestations and diagnosis" and "Initial treatment of immune thrombocytopenia (ITP) in adults".)

Monitoring of mild thrombocytopenia, management as delivery nears, and neonatal platelet count testing are discussed below. (See 'Management decisions' below.)

Symptomatic, platelets <100,000/microL, bleeding, thrombosis, or other major findings — The major causes of pregnancy-associated thrombocytopenia that present with acute systemic illness and/or severe thrombocytopenia or bleeding include preeclampsia with severe clinical features, disseminated intravascular coagulation (DIC), and thrombotic microangiopathies (TMAs; eg, thrombotic thrombocytopenic purpura [TTP] or complement-mediated TMA [CM-TMA]). Distinguishing features are summarized in the table (table 7).

Other possible diagnoses include conditions not specifically related to the pregnancy such as systemic lupus erythematosus (SLE), sepsis, severe ITP, drug-induced thrombocytopenia (DITP), catastrophic antiphospholipid syndrome (CAPS) or, rarely, other conditions such as malignancy or severe liver disease.

Preeclampsia (PE) with severe features and HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets), together designated PE/HELLP, is significantly more common than DIC, and DIC is significantly more common than the TMAs (approximate incidences, 1 in 100, 1 in 1000, and 1 in 100,000, respectively) [19]. (See "Disseminated intravascular coagulation (DIC) during pregnancy: Clinical findings, etiology, and diagnosis", section on 'Prevalence'.)

In a patient with SLE, it can be challenging (and may not be possible) to distinguish a flare of SLE from preeclampsia with severe features. This evaluation is presented separately. (See "Pregnancy in women with systemic lupus erythematosus", section on 'Preeclampsia versus lupus nephritis'.)

The evaluation to determine the cause includes review of the CBC to assess for other cytopenias (eg, leukopenia, anemia out of proportion to the stage of pregnancy) and the red blood cell indices; review of the peripheral blood smear to detect schistocytes or abnormal white blood cells, prothrombin time (PT) and activated partial thromboplastin time (aPTT), and fibrinogen level; metabolic panel with renal function and hepatic function tests; urinalysis, lactate dehydrogenase (LDH), and bilirubin to assess hemolysis; and, if the PT, aPTT, and/or fibrinogen level are abnormal, D-dimer. At least one additional platelet count measurement should be obtained in order to identify a declining trend (and its tempo).

Features that help to narrow down to the most likely (or most dangerous) cause in the symptomatic or severely thrombocytopenic patient include the following:

●Isolated thrombocytopenia – Severe ITP, drug-induced ITP

●Fever, chills – Infection, DIC

●Severe hypertension – PE/HELLP, possible TMA

●Hypotension – Bleeding, DIC

●Neurologic findings – Possible central nervous system bleeding, preeclampsia, DIC, TTP

●Bloody diarrhea – Shiga toxin-mediated hemolytic uremic syndrome (ST-HUS), possible TMA

●Hemolytic anemia (drop in hemoglobin, increased LDH and bilirubin) – PE/HELLP, DIC, TTP

●Schistocytes on the blood smear – TTP, CM-TMA, PE/HELLP, possibly DIC

●Leukopenia or leukocytosis – Infection, possibly DIC

●Prolonged PT and aPTT, low fibrinogen – DIC, severe liver disease

●Rapidly increasing creatinine – CM-TMA, ST-HUS, DIC

●Elevated liver function tests – PE/HELLP, acute fatty liver of pregnancy (AFLP), infection (eg, viral illness)

●Hypoglycemia – AFLP

●Proteinuria – PE/HELLP

The table summarizes distinguishing features between PE/HELLP and TTP (table 8).

The most important diagnostic distinctions to make are between conditions that are treated by delivery and those that are not, especially if the patient is not at term (algorithm 1). Diagnostic criteria for these conditions are summarized in the table (table 5). If a TMA is considered likely, the major decisions are whether to initiate plasma exchange or anti-complement therapies; a TMA will not resolve with delivery alone.

Early communication between the obstetrician, anesthesiologist, and consulting hematologist is advised when there are questions about the diagnosis or possible need for urgent interventions.

MANAGEMENT DECISIONS

Treatment of bleeding or severe thrombocytopenia

Platelet transfusions — The risk of severe bleeding due to thrombocytopenia only increases substantially with platelet counts below 10,000 to 20,000/microL. With platelet counts of 20,000 to 100,000/microL, increased bleeding may occur with invasive procedures but will not occur spontaneously. For platelet counts <20,000 and severe bleeding (bleeding into a closed space, bleeding requiring transfusion, bleeding that will not stop) or bleeding that is expected to become severe, platelet transfusion should be given immediately, regardless of the underlying cause of thrombocytopenia. The increase in platelet count for patients with consumptive disorders such as immune thrombocytopenia (ITP) and thrombotic thrombocytopenic purpura (TTP) may be lower than expected. (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'Actively bleeding patient'.)

Platelet transfusions are not appropriate for isolated thrombocytopenia without active bleeding, unless surgery and/or delivery is imminent.

The platelet count threshold for a non-bleeding pregnant patient nearing delivery or a procedure depends on the expected mode of delivery or type of procedure. In the absence of bleeding, we use the following thresholds:

●Vaginal delivery – Transfuse to a platelet count of 30,000/microL

●Cesarean delivery – Transfuse to a platelet count of 50,000/microL

●Neuraxial anesthesia – (See 'Neuraxial anesthesia' below.)

●Invasive procedure – (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'Preparation for an invasive procedure'.)

The safety of these thresholds is supported by data from a review of 119 pregnancies associated with immune thrombocytopenia (ITP), 17 of 110 individuals (15 percent) had platelet counts <50,000/microL at delivery [14]. Hemorrhagic complications were uncommon and did not correlate with the platelet count. None of the individuals with platelet counts <50,000/microL had blood loss >1 liter; the greatest blood loss was in four individuals with platelet counts between 54,000 and 321,000/microL. Individuals with other causes of thrombocytopenia that are also associated with platelet dysfunction may require higher platelet counts, and clinical judgment is required to incorporate the cause of thrombocytopenia and the bleeding risk for the specific patient.

The indications for platelet transfusions in patients with thrombotic thrombocytopenic purpura (TTP) or heparin-induced thrombocytopenia (HIT) do not differ from other thrombocytopenic disorders; platelet transfusions should be reserved for bleeding that is clinically important or for prevention of bleeding with an invasive procedure or delivery. (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'TTP or HIT'.)

Management of delivery — Conditions treated by delivery are discussed below. (See 'Conditions treated by delivery' below.)

There are no data comparing vaginal delivery versus cesarean delivery in ITP, and we reserve cesarean delivery for standard obstetrical considerations [6]. This practice is consistent with guidelines published by the American Society of Hematology, the British Society of Haematology, the American College of Obstetricians and Gynecologists, and an international consensus report [6,35-37]. (See "Cesarean birth: Preoperative planning and patient preparation", section on 'Indications'.)

Both forceps and vacuum-assisted delivery are relatively contraindicated in the setting of severe maternal thrombocytopenia. However, if operative vaginal delivery is performed, we use forceps rather than vacuum-assisted delivery because the potential harms of vacuum-assisted delivery are greater for the fetus than those with forceps if there is fetal thrombocytopenia. The complications of operative vaginal delivery are described in more detail separately.

We do not attempt to measure the fetal platelet count prior to delivery, and we avoid placing a fetal scalp electrode if the maternal platelet count is <80,000/microL. Measurement of the neonatal platelet count may be appropriate in some cases, as described in the next section. (See 'Neonatal testing' below.)

Frequency of platelet count monitoring — The frequency of platelet count monitoring is greater for those with more severe thrombocytopenia, with the specific interval individualized for the patient and altered as needed depending on the platelet count trend.

The following provides a general guide:

●Severely ill patients (eg, those with HELLP) may require platelet count monitoring as often as every four to six hours.

●Daily monitoring is appropriate for most patients who are hospitalized due to maternal illness, although gestational thrombocytopenia (GT) and mild, stable thrombocytopenia during hospitalization for an unrelated illness such as hyperemesis or asthma may not need daily platelet counts. Sometimes the frequency can be decreased depending on the patient's level of stability.

●For mild to moderate thrombocytopenia, which is likely due to GT, or less commonly to ITP, we generally monitor the platelet count once per trimester or once per month (depending on the absolute platelet count and platelet count trend), and assess the platelet count at 36 to 37 weeks to plan for a possible need for treatment near delivery.

●Routine obstetrical management is appropriate for platelet counts >100,000/microL.

Role of urgent/emergency delivery — The need for delivery depends on the underlying cause, which may not be definitively determined (algorithm 1). The most challenging cases are those in which the cause is not entirely clear.

In a preterm gestation, as long as mother and fetus appear stable, completing a thorough evaluation before making decisions for delivery is important, as several of the etiologies can be addressed, allowing the pregnancy to safely progress to term.

If the mother and/or fetus is unstable, and steps taken for stabilization do not allow a detailed evaluation, then delivery may be the best choice. In such cases, antenatal glucocorticoids should be given prior to delivery if possible. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery".)

At term, delivery may be warranted to facilitate evaluation of the underlying cause more efficiently.

Conditions treated by delivery — Thrombocytopenic conditions that are treated by delivery include the following, management of which is discussed in detail separately:

●Preeclampsia with severe features or HELLP syndrome – (See "Preeclampsia: Antepartum management and timing of delivery" and "HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets)", section on 'Management'.)

●Disseminated intravascular coagulation (DIC; when due to retained dead fetus or intra-amniotic infection) – (See "Disseminated intravascular coagulation (DIC) during pregnancy: Management and prognosis".)

Conditions not treated by delivery — Conditions that are not treated by delivery include:

●Thrombotic thrombocytopenic purpura (TTP) – (See "Immune TTP: Initial treatment" and "Hereditary thrombotic thrombocytopenic purpura (hTTP)", section on 'Management'.)

●Complement-mediated thrombotic microangiopathy (C-TMA) – (See "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS", section on 'Pregnancy or postpartum'.)

●Drug-induced thrombocytopenia – (See "Drug-induced immune thrombocytopenia", section on 'Management' and "Management of heparin-induced thrombocytopenia".)

●DIC (when due to a non-obstetric cause such as malignancy or extrauterine infection) – (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults", section on 'Treatment'.)

Therapies directed at the cause — In conditions in which platelets are being destroyed, the increase in platelet count with transfusion will only be temporary, and additional interventions directed at the underlying disorder are also needed.

●ITP – Glucocorticoids or IVIG. (See 'ITP therapies' below.)

●TTP – Therapeutic plasma exchange (TPE) for immune TTP; plasma infusion for hereditary TTP. (See 'Therapy for TTP or CM-TMA' below.)

●CM-TMA – Anti-complement therapy. (See 'Therapy for TTP or CM-TMA' below.)

●HIT – Cessation of all heparin exposure and anticoagulation with a non-heparin anticoagulant (eg, fondaparinux). (See 'Anticoagulation for HIT' below.)

●Other drug-induced thrombocytopenias – Cessation of drug exposure. (See "Drug-induced immune thrombocytopenia", section on 'Management'.)

ITP therapies

●Indications for ITP therapy – The management of ITP during pregnancy is generally the same as ITP in a nonpregnant patient, in that the goal of therapy is to reduce the risk of bleeding, not to normalize the platelet count. In practice, we institute treatment (glucocorticoids or IVIG) in individuals who are not bleeding only if the platelet count is below 30,000/microL or if a higher count is needed for an invasive procedure (typically below 50,000/microL if planned cesarean delivery). (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'Preparation for an invasive procedure'.)

Therapy may also be appropriate at a higher platelet count if the patient has a history of bleeding at a higher count, or if there are other factors that increase the risk of bleeding [35,36,38]. In contrast, some patients with ITP who have persistent platelet counts less than 30,000/microL may not require treatment during pregnancy if they were not receiving it prior to conception, except in preparation for delivery and/or neuraxial anesthesia. (See 'Treatment of bleeding or severe thrombocytopenia' above and 'Neuraxial anesthesia' below.)

If therapy is needed to raise the platelet count in ITP, glucocorticoids or intravenous immune globulin (IVIG) are given approximately one week prior to a scheduled delivery (algorithm 1). If the platelet count is <20,000 to 30,000/microL at the time of delivery and therapy is needed to raise the platelet count, platelet transfusions are considered. (See 'Platelet transfusions' above.)

Decisions about the need for ITP therapy during pregnancy should be based on maternal indications, as there is no evidence that administration of ITP therapies to the mother increases the fetal platelet count or improves neonatal outcomes. This lack of effect on fetal/neonatal platelet counts is illustrated by the following:

•A 2012 Cochrane review identified only a single trial from 1990 that evaluated neonatal platelet counts in pregnancies associated with maternal ITP in which the patients were randomly assigned to receive or not receive ITP therapy [39]. This trial initially included 41 pregnancies in 38 patients randomly assigned to receive betamethasone 1.5 mg daily for two weeks, but only 28 were evaluable [40]. Of these, there were no significant differences in the frequency of neonatal thrombocytopenia in the betamethasone-treated patients versus controls (seven and six infants; 64 versus 57 percent) or in the frequency of neonatal bleeding (three and two infants; 21 versus 14 percent). All of the bleeding was mild (eg, cephalohematoma associated with forceps use).

•In a retrospective review of 98 pregnancies in which therapy was administered for ITP, approximately one-half were treated with glucocorticoids and one-half with intravenous immune globulin (IVIG) [41]. Outcomes were similar with both therapies, including maternal platelet count increases (seen in approximately 40 percent in both groups) and bleeding risks, neonatal platelet counts or bleeding, and method of delivery. There were no maternal deaths or critical maternal bleeding.

●Agent selection

•Glucocorticoids – When therapy is indicated, we usually use a glucocorticoid as initial therapy. High-dose dexamethasone (typical dose, 40 mg per day for four days) and prednisone (typical dose, 1 mg/kg per day for two weeks followed by a gradual taper) are both effective; dexamethasone may have fewer adverse effects. (See "Initial treatment of immune thrombocytopenia (ITP) in adults", section on 'Choice of glucocorticoid and dosing'.)

The choice of glucocorticoid in pregnancy thus depends on whether the goal is to minimize fetal exposure or to treat both the mother and the fetus:

-When delivery is not imminent (eg, during early pregnancy and/or when preterm birth is not expected), prednisone is preferable because there is less fetal exposure [42]. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Glucocorticoids'.)

-If the patient is a candidate for therapy to improve neonatal outcomes related to preterm birth, then dexamethasone or betamethasone given antenatally can serve to treat both preterm birth and ITP. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery".)

•IVIG – IVIG may also be appropriate, either in addition to or instead of a glucocorticoid, especially if there is a need to raise the platelet count more rapidly. If therapy is needed to raise the platelet count prior to delivery or neuraxial anesthesia, it should be initiated approximately one week in advance if possible, to allow time for maximal efficacy and platelet count retesting. The increase in platelet count with IVIG is usually temporary.

Additional information about the administration and dosing of glucocorticoids and IVIG for ITP is discussed separately. (See "Initial treatment of immune thrombocytopenia (ITP) in adults", section on 'Minor bleeding and severe thrombocytopenia without bleeding'.)

●Expected efficacy and time course of response – Typical times for these therapies to take effect are as follows [36,43]:

•IVIG – 1 to 3 days for initial response, 2 to 7 days to peak response.

•Dexamethasone – 2 to 14 days for initial response, 4 to 28 days to peak response.

•Prednisone – 4 to 14 days for initial response, 7 to 28 days to peak response.

Outcomes in pregnancies with ITP have been described in retrospective studies. In general, the use of any type of therapy to raise the platelet count is reported in 30 to 40 percent of pregnancies, and serious complications are rarely seen [14,41]. The pediatrician should be informed about the possibility of neonatal thrombocytopenia, which may develop several days after delivery, as discussed below. (See 'Neonatal testing' below.)

●Adverse effects – As discussed in more detail separately, glucocorticoids have a possible small risk of causing oral clefts when administered in the first trimester of pregnancy; however, administration of glucocorticoids to facilitate delivery would be well outside this developmental window. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Glucocorticoids'.)

The fetal safety of IVIG administered to the mother are inferred from studies in the setting of neonatal alloimmune thrombocytopenia (NAIT) [44,45]. (See "Fetal and neonatal alloimmune thrombocytopenia: Parental evaluation and pregnancy management".)

●Second-line therapies (rituximab, splenectomy, TPO-RAs) – We generally reserve rituximab, splenectomy, and thrombopoietin receptor agonists (TPO-RAs) during pregnancy for recalcitrant disease, due to concerns about unknown maternal and fetal adverse effects of these interventions [33]. However, we have used both rituximab and TPO-RAs and have recommended splenectomy in pregnant individuals; these therapies may be appropriate in selected cases.

•Rituximab – Antibodies do not cross the placenta until the second trimester, with a linear increase in the amount of transfer as pregnancy progresses [46,47]. Thus, a monoclonal antibody therapy such as rituximab therapy in the first trimester should be without adverse effects.

Review of a global rituximab drug safety database identified 153 pregnancies associated with rituximab exposure for which outcomes were known. Of these, 90 (59 percent) resulted in live births; 22 (14 percent) were associated with premature birth; 11 neonates (7 percent) had hematologic abnormalities; one neonate died at age six weeks; and two infants had congenital malformations [48]. Most of these pregnancies were confounded by concomitant use of teratogenic agents and other maternal conditions (eg, lymphoma). (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Rituximab'.)

•Splenectomy – Splenectomy may be required for a patient with severe thrombocytopenia and bleeding that is unresponsive to glucocorticoids, IVIG, rituximab, and maximal doses of a TPO-RA, but splenectomy later during pregnancy becomes a greater risk as the uterus becomes larger. If required, splenectomy is ideally performed in the first half of pregnancy when other options have not worked.

•TPO-RAs – Data on the safety of TPO-RAs such as romiplostim and eltrombopag during pregnancy are limited, but emerging information suggests they do not cause a high rate of adverse events. In a review of 92 pregnancies with romiplostim exposure (mostly from 20 days before pregnancy and the first trimester) and known birth outcome, the rates of adverse pregnancy outcomes were similar to the general population [49]. The rate of spontaneous miscarriage was 13 percent. Of 42 infants with first trimester exposure, one had an inguinal hernia and another had a single umbilical artery; of 31 infants with third trimester exposure, one had growth restriction. There were no chromosomal abnormalities attributed to romiplostim. Other clinical experience is limited to case reports [50-52].

TPO-RAs stimulate platelet production by activating the platelet thrombopoietin receptor; they also may stimulate other hematopoietic cells. Registries have been established by the manufacturers to document pregnancy outcomes when these agents are inadvertently administered during pregnancy. (See "Second-line and subsequent therapies for immune thrombocytopenia (ITP) in adults", section on 'TPO receptor agonists' and "Clinical applications of thrombopoietic growth factors".)

A form of recombinant human thrombopoietin (rhTPO) is available in China. When administered to 31 pregnant individuals with ITP who had platelet counts <30,000/microL that could not be increased by glucocorticoids, IVIG, or platelet transfusions, this rhTPO was well tolerated and was not associated with adverse effects in the newborns [53]. Platelet counts normalized in 10 and increased in an additional 13 (overall response rate, 74 percent).

If an individual taking rituximab or a TPO-RA becomes pregnant, discussion regarding the necessity of continuing the medication in the setting of limited data on fetal risks versus the use of alternative medications should be individualized. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Rituximab'.)

Anticoagulation for HIT — Heparin-induced thrombocytopenia (HIT) is extremely rare in pregnancy. If HIT is suspected or confirmed, all heparin exposure should be eliminated and presumptive therapy with a non-heparin anticoagulant should be initiated. The evaluation and management of HIT, and options for anticoagulation during pregnancy, are discussed in detail separately. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia" and "Management of heparin-induced thrombocytopenia" and "Use of anticoagulants during pregnancy and postpartum", section on 'Alternatives to heparin'.)

Therapy for TTP or CM-TMA — Therapeutic plasma exchange (TPE) is urgent initial treatment for immune TTP; plasma infusion is the treatment for hereditary TTP; and anti-complement therapy is urgent initial treatment indicated for complement-mediated TMA (CM-TMA) (algorithm 1).

TTP and CM-TMA can usually be distinguished by the degree of kidney injury. Acute kidney injury (AKI) with an increasing serum creatinine causing anticipation of dialysis is typical of CM-TMA and very rarely occurs in TTP. (See "Immune TTP: Initial treatment" and "Hereditary thrombotic thrombocytopenic purpura (hTTP)" and "Thrombotic microangiopathies (TMAs) with acute kidney injury (AKI) in adults: CM-TMA and ST-HUS".)

TPE and anti-complement monoclonal antibodies cannot be initiated simultaneously, unless the monoclonal antibody is repeated daily (instead of weekly) following each TPE procedure, because TPE removes the antibody.

TPE and anti-complement therapy carry risks of potentially serious adverse effects and are costly. Thus, we reserve these therapies for individuals with strong supporting evidence for the underlying condition [54]. This includes the following features (see "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Initial evaluation (all patients)'):

●Thrombocytopenia and microangiopathic hemolytic anemia (MAHA)

●Lack of evidence of an alternative cause of thrombocytopenia and MAHA

•No obvious malignancy by physical examination

•No obvious DIC by coagulation testing

•No obvious drug-induced cause by history

•No obvious acute liver disease or autoimmune disease by examination and laboratory testing

•No previous hypotension as a cause of acute tubular necrosis and AKI

MAHA and thrombocytopenia are characteristic features of preeclampsia with severe features, which typically resolves within three days following delivery. Persistence of MAHA and thrombocytopenia for more than three days postpartum is also consistent with TTP or CM-TMA, but delivery should not be performed solely to establish this finding, and therapy should not be delayed until after delivery if one of these TMAs is strongly suspected.

Decisions regarding the likelihood of TTP and CM-TMA may be complex and may require the involvement of multiple specialists. We advise early involvement of the consulting hematologist, nephrologist (if appropriate), and laboratory personnel, as well as the anesthesiologist should these therapies be ineffective and emergency delivery required. These considerations are discussed separately. (See "Diagnostic approach to suspected TTP, HUS, or other thrombotic microangiopathy (TMA)", section on 'Key distinguishing features among the primary TMA syndromes'.)

Management of hereditary TTP (due to biallelic pathogenic variants in the ADAMTS13 gene) during pregnancy is discussed separately. (See "Hereditary thrombotic thrombocytopenic purpura (hTTP)", section on 'Management of pregnancy'.)

Antibiotics — Broad spectrum antibiotics are appropriate for patients with a presumed infectious cause of thrombocytopenia (severe bacterial infection or DIC).

The choice of antibiotic depends on the site of infection and likely organisms, as discussed in separate topic reviews. As examples:

●Pyelonephritis – (See "Urinary tract infections and asymptomatic bacteriuria in pregnancy", section on 'Acute pyelonephritis'.)

●Pneumonia – (See "Approach to the pregnant patient with a respiratory infection", section on 'Community-acquired pneumonia'.)

●Septic thrombophlebitis – (See "Septic pelvic thrombophlebitis", section on 'Treatment'.)

●Intraamniotic infection – (See "Clinical chorioamnionitis", section on 'Antibiotic therapy'.)

Neuraxial anesthesia — High-quality data are not available to determine what constitutes a safe platelet count for placement of a neuraxial catheter, and the threshold may differ for different patients and different thrombocytopenic disorders. It is also worth noting that of the numerous pregnancies in which epidural anesthesia has been administered without platelet count testing, some will predictably be mildly thrombocytopenic and not counted in case series [55].Epidural anesthesia is generally considered safe if the platelet count is above 50,000 to 80,000/microL, provided the platelet count is stable, platelet function is normal, the patient is not taking antiplatelet or anticoagulant drugs, and they have no other acquired or inherited coagulation disorder [37,56-58]. A 2021 consensus from the Society for Obstetric Anesthesia stated that the best evidence suggests the risk of spinal epidural hematoma is very low with a platelet count >70,000/microL [59]. Preferences and practice patterns vary among individual anesthesiologists; anesthesiologists will make the final decision as they are performing the procedure. It is important to discuss options for neuraxial analgesia or anesthesia with the consulting anesthesiologist prior to delivery in patients with thrombocytopenia. Obstetricians should be familiar with policies at their institution and preferences of the anesthesiologist(s) who will perform their patient's neuraxial procedure. This subject is discussed in more detail separately. (See "Adverse effects of neuraxial analgesia and anesthesia for obstetrics", section on 'Neuraxial analgesia and low platelets'.)

Depending on the cause and severity of thrombocytopenia, platelet transfusion and/or other therapies may be indicated two or more days prior to placement of a neuraxial catheter. (See 'ITP therapies' above.)

For individuals with a presumed diagnosis of ITP who are near term and have a platelet count just below the "acceptable" limit for regional anesthesia, a brief treatment course with a glucocorticoid such as dexamethasone, 40 mg daily for up to four days may be used, with the understanding that there is little evidence to support this practice. Compared with prednisone, dexamethasone has better efficacy in raising the platelet count and fewer adverse effects. In contrast to prednisone, dexamethasone crosses the placenta in its active form, which may improve the fetal platelet count (if low) and aid in lung maturation, although either dexamethasone or prednisone can be used. Clinicians and their patients should decide the risks and benefits of giving glucocorticoids to raise the platelet count prior to delivery on a case-by-case basis. (See "Initial treatment of immune thrombocytopenia (ITP) in adults", section on 'Choice of glucocorticoid and dosing' and 'ITP therapies' above.)

The range of outcomes from placement of a neuraxial catheter in pregnancies with ITP was illustrated in a retrospective series of 119 pregnant individuals with ITP over a 10-year period, for whom data on analgesia were available for all but one [14]. Of the 42 individuals who received epidural anesthesia, none had a complication related to the catheter placement. Platelet counts were <100,000/microL in 19 (45 percent), <75,000/microL in six (14 percent), and <50,000/microL in one (2 percent).

Neonatal testing — Settings in which testing of the neonatal platelet count is prudent include the following:

●Maternal ITP.

●Neonatal thrombocytopenia in a previous pregnancy. (See "Neonatal thrombocytopenia: Etiology" and "Fetal and neonatal alloimmune thrombocytopenia: Parental evaluation and pregnancy management".)

●Indications unrelated to maternal platelet count, such as:

•Bleeding or petechiae/ecchymoses in the infant

•Congenital anomalies associated with thrombocytopenia

•Trisomy 21, 18, or 13

•Neonatal infections such as cytomegalovirus or rubella.

These are discussed separately. (See "Neonatal thrombocytopenia: Clinical manifestations, evaluation, and management".)

The risk of neonatal thrombocytopenia in pregnancies with gestational thrombocytopenia (GT) is difficult to determine, as there are no tests that can distinguish GT from ITP, and individuals in some series may have been misclassified [60]. If we are confident in the diagnosis of GT (if previous pregnancies were associated with maternal thrombocytopenia that resolved after delivery), then we do not routinely obtain a neonatal platelet count. If the diagnosis of GT versus ITP is unclear, it may be reasonable to obtain a neonatal platelet count.

The overall risk of thrombocytopenia in infants born to patients with ITP has been estimated to be in the range of 10 to 15 percent; severe thrombocytopenia is possible but much less common. In the series of 119 pregnancies associated with maternal ITP, 31 of 109 infants (28 percent) had platelet counts <150,000/microL [14]. Of these, six (6 percent) had counts <20,000/microL, five had counts between 20,000 and 50,000, and the remainder had counts >50,000/microL. One infant had an intracerebral hemorrhage (ICH) that was not clearly due to thrombocytopenia; the infant was born at 29 weeks of a twin gestation, and the platelet count nadir was 135,000/microL on the day after delivery. There were two fetal deaths, one unrelated to thrombocytopenia and one a stillbirth at 27 weeks associated with extensive fetal hemorrhage and severe maternal thrombocytopenia. Smaller series have found similar rates of thrombocytopenia and similar platelet count distributions [61-69].

There does not appear to be a strong correlation between maternal platelet count and neonatal platelet count in ITP [14]. Risk factors for neonatal thrombocytopenia include a previous history of neonatal thrombocytopenia, prior splenectomy, severe thrombocytopenia (<50,000/microL) at some point during the pregnancy, and possibly maternal platelet count <100,000/microL at the time of delivery [14,61,64,65,70,71]. There is no evidence that ITP therapy for the mother raises the fetal platelet count. (See 'ITP therapies' above.)

When neonatal testing is performed, it should be done after delivery, typically by cord blood platelet count using clean venipuncture of a cord vessel rather than draining blood from the cord [35]. Platelet count testing by scalp blood sampling or cordocentesis is not recommended because these procedures are associated with adverse events and have not been demonstrated to improve outcomes. Percutaneous umbilical cord blood sampling carries an increased risk of fetal hemorrhage that is considered to be greater than the risk of severe neonatal ICH from thrombocytopenia (approximately 2 versus 1 percent) [6,14,35,61,72]. Fetal scalp blood sampling has been abandoned because it is technically difficult and frequently results in spuriously low platelet counts [6,35,38,61].

Based on our experience, if the neonate's platelet count is <150,000/microL, a repeat platelet count the next day is appropriate to be confident that the platelet count is stable or increasing. The neonatal platelet count may decrease up to several days after birth (typical nadir at day 2 to 5 postpartum) [63]. The mechanism is thought to involve passage of maternal antiplatelet antibodies to the fetal circulation, with gradual acquisition of infant splenic function within the first few days after birth leading to destruction of antibody-sensitized platelets. Management of the thrombocytopenic neonate is discussed separately. (See "Neonatal thrombocytopenia: Clinical manifestations, evaluation, and management", section on 'Management'.)

Postpartum — The postpartum course depends on the cause of thrombocytopenia. If the diagnosis was unclear, the postpartum course is helpful in clarifying the diagnosis, and the diagnosis in turn guides postpartum management.

●Thrombocytopenia is expected to resolve postpartum in GT, PE/HELLP syndrome, and DIC for which the cause has been corrected.

•In GT, the maternal platelet count may return to normal within two to four weeks following delivery but may not be normal until six to eight weeks [1,5,73]. In our 2018 study of platelet counts in pregnancy, the rate of reoccurrence of GT in a subsequent pregnancy was 63 percent [1]. When GT recurs during subsequent pregnancies, the platelet count decrease is typically similar to the decrease in prior pregnancies [5,60]. For individuals with presumed GT, we check a platelet count at the six-week postpartum visit and repeat the count in another one to two weeks only if the six-week count remains low.

•Diagnosis of new onset thrombocytopenia in PE/HELLP in the postpartum period accounts for approximately five percent of PE/HELLP cases. When it occurs, it should be evaluated and treated appropriately. In most cases, maternal platelet count recovery begins within one to three days postpartum. We initially check the platelet count daily until it begins rising and then again at the six-week postpartum visit. For HELLP, we check the platelet count every 6 to 12 hours until it begins to stabilize or improve. Failure to stabilize or resolve should prompt evaluation for an alternative diagnosis, such as TTP or CM-TMA.

•If DIC is present at term, recovery from thrombocytopenia is related to correction of the underlying cause of the DIC. Most causes associated with pregnancy, such as placental abruption, will be corrected by delivery, although others may not be corrected. Sepsis requires appropriate identification of the infectious organism, treatment with antibiotics, and control of the source, with possible surgical removal. We check the platelet count daily (or more often, if clinical features warrant) until it begins to stabilize or improve. The causes of DIC in pregnancy, their management, and expected postpartum course are discussed in more detail separately. (See "Disseminated intravascular coagulation (DIC) during pregnancy: Clinical findings, etiology, and diagnosis".)

●Thrombocytopenia caused by a condition unrelated to the pregnancy such as an inherited platelet disorder or ITP will persist after delivery.

For stable or worsening thrombocytopenia following delivery, diagnostic evaluation should be pursued as done for nonpregnant individuals. (See "Diagnostic approach to thrombocytopenia in adults".)

Indications for platelet count testing in the neonate are described above. (See 'Neonatal testing' above.)

Recurrence in future pregnancies — The recurrence of thrombocytopenia in future pregnancies depends on the underlying etiology. As examples:

●GT recurs in most subsequent pregnancies (63 percent) [1]. The risk for recurrence was 14-fold greater than the risk for GT in individuals without a history of previous GT.

●Immune TTP has a low but possible risk of recurrence with a subsequent pregnancy [22]. We follow these individuals with frequent blood counts throughout their pregnancy but do not use prophylactic treatment. (See "Immune TTP: Management following recovery from an acute episode and during remission", section on 'Pregnancy after an episode of TTP'.)

●Hereditary TTP predictably recurs during subsequent pregnancies unless appropriate prophylaxis with plasma infusion is begun when pregnancy is first confirmed [20,21,74]. (See "Hereditary thrombotic thrombocytopenic purpura (hTTP)", section on 'Prophylaxis/prevention of future exacerbations'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Immune thrombocytopenia (ITP) and other platelet disorders".)

PATIENT PERSPECTIVE TOPIC — Patient perspectives are provided for selected disorders to help clinicians better understand the patient experience and patient concerns. These narratives may offer insights into patient values and preferences not included in other UpToDate topics. (See "Patient perspective: Thrombotic thrombocytopenic purpura (TTP)".)

SUMMARY AND RECOMMENDATIONS

●Incidence – Platelet counts decrease 15 to 20 percent during uncomplicated pregnancies, starting in the first trimester, but generally remain in the normal range (150,000 to 450,000/microL) (table 1). Mild thrombocytopenia (100,000 to 149,000/microL) is usually due to gestational thrombocytopenia (GT) and does not require further evaluation (table 2). (See 'Definition and incidence' above.)

•GT – GT is a physiologic condition that requires no evaluation or treatment. It is the most common cause of thrombocytopenia in pregnancy and is the presumptive diagnosis with a platelet count of 100,000 to 149,000/microL, provided there are no other abnormal findings. It is a diagnosis of exclusion. GT affects 5 to 10 percent of pregnancies; only 1 percent of these have a platelet count <100,000/microL (table 2). (See 'Gestational thrombocytopenia (GT)' above.)

•ITP – Immune thrombocytopenia (ITP) is an autoimmune condition that can precede pregnancy or can occur during pregnancy or postpartum. ITP is a diagnosis of exclusion. (See 'Immune thrombocytopenia (ITP)' above and "Immune thrombocytopenia (ITP) in adults: Clinical manifestations and diagnosis".)

•Less common causes – The table lists conditions associated with severe thrombocytopenia and acute systemic illness (table 3); these include preeclampsia with severe features and HELLP (hemolysis, elevated liver enzymes, and low platelets), disseminated intravascular coagulation (DIC), and thrombotic microangiopathies (TMAs). (See 'Preeclampsia with severe features/HELLP (PE/HELLP)' above and 'DIC' above and 'Thrombotic microangiopathy (TMA)' above.)

•Unrelated to pregnancy – Other rare causes of thrombocytopenia may occur or be discovered during the pregnancy, including systemic lupus erythematosus (SLE), catastrophic antiphospholipid syndrome (CAPS), drug-induced thrombocytopenia, cancer, infection, liver disease, hypersplenism, deficiency of vitamin B12/folate/copper, or hereditary platelet disorders (table 3). (See 'Other causes of thrombocytopenia' above.)

●Evaluation – Most pregnancies with mild thrombocytopenia without other cytopenias or major clinical findings will have GT. It is not possible or necessary to distinguish GT from mild ITP; both are diagnoses of exclusion and neither requires therapy. In individuals who are acutely ill, PE with severe features is significantly more common than DIC, and DIC is significantly more common than the TMAs (table 7). Laboratory testing is discussed above; the table lists diagnostic criteria (table 5). (See 'Determining the likely cause(s)' above.)

●Platelet transfusions – Platelet transfusion may be required for severe thrombocytopenia with bleeding, invasive procedures, or delivery. Thresholds are provided above. (See 'Treatment of bleeding or severe thrombocytopenia' above.)

●Management – Management depends on the underlying cause (algorithm 1):

•Preeclampsia with severe features, HELLP, DIC/sepsis due to retained dead fetus or amniotic infection – Delivery. (See 'Role of urgent/emergency delivery' above.)

•TMAs – Immune thrombotic thrombocytopenic purpura (TTP) is treated with therapeutic plasma exchange (TPE). Complement mediated (CM)-TMA is treated with anticomplement therapy. Hereditary TTP is treated with plasma infusions from confirmed pregnancy until six weeks postpartum. (See 'Therapy for TTP or CM-TMA' above.)

•ITP – Not all patients with ITP require treatment to raise the platelet count. Glucocorticoids or intravenous immune globulin (IVIG) are used if needed. (See 'ITP therapies' above.)

•Heparin-induced thrombocytopenia (HIT) – Heparin discontinuation and anticoagulation with a nonheparin agent. (See 'Anticoagulation for HIT' above.)

•Infection/sepsis – Appropriate antibiotics and sepsis protocols. (See 'Antibiotics' above.)

•Neuraxial anesthesia, delivery, neonatal testing, and postpartum care – (See 'Neuraxial anesthesia' above and 'Neonatal testing' above and 'Postpartum' above and 'Recurrence in future pregnancies' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Eric J Knudtson, MD, deceased, who contributed to an earlier version of this topic review.