Thrombocytopenia in pregnancy

What every physician needs to know:

Thrombocytopenia can be seen in 5 to 10% of all pregnancies. There is an expected physiologic drop in the platelet count during pregnancy so that a platelet count below the lower limit of normal is usually not concerning until the platelet count reaches 100,000 to 115,00/ul. Determining the etiology of thrombocytopenia is important so that appropriate care and management can be given to improve maternal outcomes and decrease bleeding risks, especially the risk of maternal hemorrhage at the time of delivery. Generally, platelet counts less than 100,000/ul require further evaluation by a hematologist.

The most common etiology of thrombocytopenia in pregnancy is gestational thrombocytopenia, occurring in 70% of cases. The remaining 30% of cases are due to a variety of diagnoses, including idiopathic thrombocytopenic purpura (ITP) and other autoimmune associated causes of thrombocytopenia (for example, systemic lupus erythematosus [SLE]).

Disorders associated with microangiopathic changes and thrombocytopenia include:

Continue Reading

  • pre-eclampsia/eclampsia,
  • HELLP (hemolysis, elevated liver enzymes, low platelets) syndrome,
  • thrombotic thrombocytopenic purpura – hemolytic uremic syndrome (TTP/HUS).

Some of these disorders are unique to pregnancy, such as pre-eclampsia or HELLP syndrome. Some, such as TTP, appear to occur more frequently in pregnancy. Others are not pregnancy specific and can occur with equal frequency in pregnant and non-pregnant patients. Rare inherited platelet disorders associated with thrombocytopenia, such as May-Hegglin anomaly, may come to light during pregnancy, as pregnancy may be the first time the patient has a complete blood count (CBC) performed.

The first step in evaluating the pregnant woman with thrombocytopenia is to determine the urgency required to identify the diagnosis. Symptomatic, moderate or severe bleeding, platelet count less than 20,000/ul, or signs or symptoms of microangiopathic processes (such as TTP/HUS, HELLP, or pre-eclampsia) require immediate evaluation. The work-up of an asymptomatic patient with incidental finding of mild thrombocytopenia on routine CBC with no signs of bleeding, bruising, or petechiae, and without associated hypertension (HTN) and proteinuria, can proceed at a slower pace.

Patients with platelet counts lower than 10,000 to 20,000/ul or signs or symptoms of bleeding need to be assessed as soon as the platelet count is known. Stage of pregnancy, prior history of thrombocytopenia and treatments (if any), CBC and reticulocyte count, PT (prothrombin time), PTT (partial thromboplastin time), fibrinogen, and comprehensive chemistry panel that includes transaminases, alkaline phosphatase, lactate dehydrogenase (LDH), and total bilirubin, along with blood pressure and urine analysis for protein, should be obtained. The history and these test results direct the next steps in evaluation and management of the patient.

The most difficult diagnostic dilemma is usually between idiopathic thrombocytopenic purpura (ITP) and gestational thrombocytopenia, as other causes of thrombocytopenia are often obvious, based on physical exam and other lab findings.

Primary ITP is diagnosed by excluding all other causes of thrombocytopenia, including neoplastic diseases and secondary autoimmune thrombocytopenia associated with HIV, hepatitis C, Helicobacter pylori infection, or rheumatologic disorders. Gestational thrombocytopenia is similar; other etiologies must be excluded and the platelet count observed over time. Often it is only the level to which the platelet count drops that distinguishes between ITP and gestational thrombocytopenia. In patients with gestational thrombocytopenia, the platelet count usually does not drop below 70,000/ul.

A past history of thrombocytopenia, either during or unassociated with pregnancy, guides evaluation and management.

In gestational thrombocytopenia, the platelet count returns to normal soon after delivery without other treatment. It can recur in subsequent pregnancies.

Patients with ITP will often have low platelet counts prior to pregnancy or thrombocytopenia that persists postpartum, and may have required treatment for ITP. Collagen vascular disorders, such as lupus or rheumatoid arthritis, can be associated with thrombocytopenia. Circulating antiphospholipid antibodies are also associated with thrombocytopenia in a small number of patients who may have a history of lupus anticoagulant or elevated PTT.

Rheumatological diseases, such as lupus, can flare during pregnancy resulting in exacerbation of pre-existing, or development of new, cytopenias (including thrombocytopenia).

Determining the etiology of thrombocytopenia is important to minimize potential bleeding risks for mother and fetus during gestation, labor, and delivery. Some disorders do not require treatment, such as gestational thrombocytopenia or May-Hegglin anomaly, whereas in other disorders, such as TTP, rapid institution of appropriate therapy can be life-saving and have a significant impact on outcome.

What features of the presentation will guide me toward possible causes and next treatment steps?

The stage of pregnancy, and both the rate of decline and the magnitude of drop in platelet count, can be helpful in determining diagnosis.

Gestational thrombocytopenia usually presents in the third trimester. ITP can occur during any trimester and is generally the only cause of new thrombocytopenia in the first trimester. Congenital thrombocytopenias that have gone unidentified can also be discovered in the first trimester.

The platelet count in gestational thrombocytopenia stays in the same general range, whereas in ITP, there can be an extremely rapid drop from low/normal (around 120,000/ul) to between 10,000 and 20,000/ul, over the course of a week or two.

In a patient with completely normal physical findings, normal white blood cells (WBCs), normal WBC morphology and normal hematocrit, the differential diagnosis is limited to gestational thrombocytopenia versus ITP, once viral and collagen vascular etiologies are excluded. Patients with gestational thrombocytopenia will usually have platelet counts in the 80,000 to 90,000/ul range. Patients with ITP will have platelet counts that drop below 70,000/ul and often significantly lower.

Patients with gestational thrombocytopenia do not experience bleeding. Patients with ITP will experience at least minor bleeding (petechiae, mild trauma related bruising, and epistaxis) as the platelet count falls below 30,000/ul. The extent of bleeding directly correlates with the degree of thrombocytopenia in ITP. Patients with moderate or more severe bleeding and platelet counts above 30,000/ul should be evaluated for intrinsic bone marrow disorders, disseminated intravascular coagulation (DIC) and the other microangiopathic disorders, platelet function disorders, and liver disease.

In the pregnant patient with leukopenia, anemia, and thrombocytopenia, the differential diagnosis is the same as in a non-pregnant patient and includes viral marrow suppression, collagen vascular disorders, aplastic anemia, and hematologic malignancies such as leukemia, lymphoma, myelodysplastic syndromes (MDS), myelofibrosis, and other neoplasms.

Thrombocytopenia and anemia with evidence of hemolysis point to pre-eclampsia, HELLP, and TTP/HUS as possible diagnoses. There are other associated physical signs and symptoms in these disorders, such as:

  • HTN and proteinuria in pre-eclampsia,
  • right upper quadrant pain (RUQ) and abnormal LFTS in HELLP and acute fatty liver of pregnancy (AFLP),
  • and fever, neurologic changes, and renal insufficiency in TTP and HUS.

Most of these disorders do not present until late in the second or third trimester.

HTN and proteinuria are required criteria for the diagnosis of pre-eclampsia. HTN is defined as systolic BP greater than or equal to 140 mmHg or diastolic 90 mmHg occurring after 20 weeks gestation. Proteinuria needs to be 1+ or higher on dipstick, confirmed with 0.3 gms or greater in 24 hour urine. Thrombocytopenia occurs in 50% of cases of pre-eclampsia and can sometimes be the first sign of the disorder.

Clinical and lab findings suggestive of liver disease can also indicate viral hepatitis, cirrhosis, or other acute or chronic liver disorders that can be found in both pregnant and non-pregnant patients and can be associated with thrombocytopenia.

All pregnant patients have an increase in procoagulant factors, including von Willebrand factor (vWF), fibrinogen and FVIII, with a decrease in activity of protein S.

What laboratory studies should you order to help make the diagnosis and how should you interpret the results?

Laboratory studies (first level):
  • CBC
  • Reticulocyte count
  • Peripheral smear
  • PT
  • PTT
  • Fibrinogen
  • Chemistry panel: blood, urea, nitrogen (BUN), creatinine, and electrolytes
  • Transaminases
  • Alkaline phosphatase
  • LDH
  • Total bilirubin
  • Urine protein analysis

These tests are the first level of laboratory testing used to determine the etiology of thrombocytopenia. In addition to the platelet count, other values included in the CBC can help point to underlying etiology. Abnormalities in WBC, hemoglobin and hematocrit, and often mean corpuscular volume (MCV) suggest a global marrow process or systemic disorder such as collagen vascular disease. Abnormal WBC morphology or differential can indicate malignancy, intrinsic marrow process, or viral infection.

Although anemia is common in pregnancy, due to increase in plasma volume often accompanied by iron deficiency, it is important to evaluate anemia that occurs in the pregnant patient with thrombocytopenia. Abnormalities in the MCV, red cell distribution volume (RDW), and reticulocyte count point to increased or decreased red cell production. Increased reticulocytes accompany hemolytic processes, and decreased reticulocytes can be seen in nutrient deficiency or intrinsic marrow abnormalities.

Platelet clumping on the peripheral smear indicates pseudo thrombocytopenia.

Schistocytes and fragmented red cells together with thrombocytopenia suggest DIC, TTP, HUS, pre-eclampsia or HELLP syndrome.

Elevated LDH and transaminases and low haptoglobin support the diagnosis of these microangiopathic hemolytic processes, as do elevated PT/PTT and decreased fibrinogen in DIC.

Elevated creatinine in addition to thrombocytopenia and schistocytes suggests TTP or HUS.

Proteinuria with HTN points to pre-eclampsia.

Larger platelets on peripheral smear can be seen in ITP, as well as in MDS, myeloproliferative disorders and myelofibrosis.

Macro-thrombocytes are characteristic of many of the congenital thrombocytopenic disorders, including the MHY9 mutation-associated disorders. Döhle bodies in granulocytes and extremely large platelets in addition to thrombocytopenia may be due to congenital May-Hegglin anomaly.

Elevated transaminases and LDH and mild hyperbilirubinemia are seen in HELLP and AFLP.

Further laboratory tests
  • HIV (human immunodeficiency virus)
  • hepatitis C
  • ANA (antinuclear antibody)
  • rheumatoid factor
  • antiphospholipid antibodies

A positive result from one of the above tests can be an indication of a disorder that results in secondary ITP. Both HIV and hepatitis C infections can have associated antibody-mediated platelet destruction. Lupus and rheumatoid arthritis can have cytopenias that may or may not be antibody-mediated, including thrombocytopenia. Antiphospholipid antibodies can be associated with thrombocytopenia by binding to and clearing platelets from the circulation.

  • von Willebrand antigen
  • FVIII activity
  • ristocetin cofactor activity
  • vWF (von Willebrand factor) multimer analysis

Patients who have a history of recurrent mucosal bleeding, menorrhagia, or family history of this type of bleeding, and mild thrombocytopenia, should be evaluated for type II B von Willebrand disease (vWD).

Type II B vWD is caused by a gain-of-function mutation in the vWF molecule that results in increased affinity for platelets. Large platelet–vWF aggregates are formed and are cleared from the circulation, resulting in thrombocytopenia. During pregnancy, the estrogen-dependent increase in circulating vWF can result in even lower platelet counts. Although vWF antigen levels rise in pregnancy, the increased sensitivity to low dose ristocetin that characterizes type II B vWD does not change. The largest plasma vWF multimers are also absent in type II B vWD and do not change in pregnancy.

What conditions can underlie thrombocytopenia in pregnancy:


When do you need to get more aggressive tests?

Patients with unexplained abnormalities in the CBC (such as leukocytosis, leukopenia, or anemia) in addition to thrombocytopenia may require bone marrow aspiration/biopsy. Patients with abnormal WBC morphology on peripheral smear, elevated MCV not due to B12 or folate deficiency, or those suspected of having hematologic malignancy, may also require bone marrow aspiration/biopsy.

Results of screening tests for von Willebrand disease (vWD) in a pregnant patient with a history of bleeding, bruising, or menorrhagia prior to pregnancy, or a family history of these symptoms, can be difficult to interpret, due to estrogen-induced increase in vWF antigen.

What imaging studies (if any) will be helpful?

Ultrasound (US) of liver and spleen in early stages of pregnancy should be obtained if liver disease or splenomegaly with splenic sequestration is considered an etiology for thrombocytopenia. In later stages of pregnancy, US can be technically difficult, but can be attempted in patients with right upper quadrant (RUQ) pain to aid in evaluation of HELLP, versus amplified fragment length polymorphism (AFLP), versus other etiologies. CT scan to evaluate the liver may be required, but there are risks of radiation exposure to the mother and fetus.

What therapies should you initiate immediately and under what circumstances – even if root cause is unidentified?

It is important to note that for most of the disorders resulting in thrombocytopenia, there are almost no randomized controlled trial (RCT) data for treatment in pregnant patients. Treatment is based on clinical experience, extrapolation from RCTs in non-pregnant patients, and expert consensus.

Thrombocytopenia is a concern in any patient because of the risk of bleeding, and the broad heterogeneity of underlying diagnoses. Anxiety over thrombocytopenia in the pregnant patient is heightened because of the hemostatic stress of labor and delivery, concern for uterine or placental bleeding that could result in adverse outcomes for both mother and fetus at any point in pregnancy, and possibility of neonatal thrombocytopenia and bleeding. Thrombocytopenia in pregnancy is also associated with disorders such as pre-eclampsia or TTP that can have significant maternal morbidity and mortality.

In the thrombocytopenic pregnant patient the first question to ask is – “Is there bleeding?” If yes, how severe? Moderate to life-threatening bleeding requires immediate evaluation and treatment, using standard approach and supportive care.

Patients who have moderate to life-threatening bleeding

Rapid evaluation for pre-eclampsia/eclampsia, HELLP, TTP and HUS is required before transfusing platelets or instituting other therapies. If findings are consistent with any of these diagnoses, appropriate therapy should be started as quickly as possible.

Plasma exchange therapy for treatment of presumed TTP should be instituted in any patient with thrombocytopenia and microangiopathic hemolysis without clear-cut other diagnosis, and should be continued until the diagnosis is established. Plasma exchange is the primary therapy for TTP/HUS in non-pregnant and pregnant patients. TTP/HUS usually does not resolve with delivery, and therapy must continue postpartum until resolution of TTP/HUS. Patients require aggressive supportive care to manage other complications, including red cell transfusions, as needed.

Definitive treatment for pre-eclampsia and HELLP syndrome is delivery of the fetus. The gestational age of the fetus, however, significantly impacts decisions regarding delivery or termination of pregnancy.

Moderate to severe bleeding with thrombocytopenia and no findings of microangiopathic hemolytic anemia can be treated with platelet transfusions as evaluation for underlying etiology of thrombocytopenia continues. Bleeding is not a feature of gestational thrombocytopenia, as the platelet count does not drop low enough to result in bleeding. Mild trauma-induced bruising and minor mucosal bleeding can occur in ITP, but moderate to severe bleeding with platelet counts greater than 30,000/ul is uncommon in ITP.

Patients who do not have clinically significant bleeding

Even in patients who do not have clinically significant bleeding, the platelet count serves as a guide for urgency of institution of therapy. Patients with platelet counts less than 10,000/ul with minimal clinical bleeding and no other symptoms likely have ITP and should be treated rapidly to increase the platelet count, especially if in the third trimester of pregnancy. IV IgG (one gram/ kg) is the initial treatment of choice for pregnant patients with ITP and platelet counts below 10,000/ul, an increased platelet count can be seen within 12 hours of administration. The platelet count will continue to rise over the next few days.

The question of whether to admit the pregnant patient with an extremely low platelet count to the hospital for treatment can arise. In a reliable patient with a known diagnosis of ITP, with no clinical signs or symptoms of bleeding, a normally progressing pregnancy, and ability to return the following day for evaluation and CBC, admission for IV IgG is not necessary. If however, there is any doubt about the diagnosis, or the ability of the patient to return if bleeding develops, or if labor is imminent, admission is warranted.

IV IgG is a treatment that only temporarily raises the platelet count. The duration of response to one dose of IV IgG is variable, lasting from 1 to 6 weeks. Further treatment should be planned, as described below.

Cesarean section (C-section) can be performed if necessary, and epidural or spinal anesthesia used if desired or necessary. These treatment plans vary somewhat depending on whether labor and delivery will occur spontaneously or will be a scheduled induction of labor or elective C-section. The minimum platelet count for C-section is 50,000/ul; however institutional guidelines for epidural or spinal anesthesia often require a higher value, usually between 70,000-100,000, therefore the target platelet count at 35 weeks gestation and later in pregnancy should be 100,000/ul.

Patients who have no bleeding

For patients with no bleeding, no other clinical findings, and platelet count greater than 30,000/ul but less than 70,000/ul, the presumptive diagnosis is ITP. In the early stages of pregnancy there is generally no need to treat ITP unless the platelet count drops below 10,000-20,000/ul, or there is associated minor bleeding such as epistaxis and bruising. The threshold platelet count for treatment is variable. The rapidity with which the platelet count falls is often a more significant factor in determining need for therapy in ITP than the actual number.

In a pregnant patient with newly diagnosed ITP and rapidly falling platelet count, there is no need to wait until the count falls precisely below 10,000 or 20,000/ul. In a patient with chronic ITP and a relatively steady platelet count, even at 15,000-20,000/ul, treatment in early pregnancy can be deferred until a further drop or signs or symptoms of bleeding occur. Platelet counts can be checked once a month in early pregnancy as long as they are stable and as long as the patient is aware that she should have a platelet count with the onset of any new bleeding (epistaxis, gingival bleeding, bruising).

As pregnancy progresses, the platelet count should be obtained every 2 weeks starting in the third trimester, and then once a week after 36 weeks. A platelet count of 30,000/ul is sufficient to prevent maternal bleeding in uncomplicated delivery, but for C-section a minimum of 50,000/ul is required. The platelet threshold for neuraxial anesthesia varies from institution to institution but is generally in the range of 70,000-100,000/ul. If treatment for ITP is given in preparation for delivery, the target platelet count should be roughly 100,000/ul as it is difficult to achieve a precise number.

Glucocorticoids are first line therapy in patients without clinical bleeding and platelet count above 10,000/ul. The response rate is 60-80%. It takes longer to see an increase in the platelet count with corticosteroids than with IV IgG, with an increase in platelet count often not observed until after 5-10 days of treatment.

Early in pregnancy during organogenesis there is concern for an association between glucocorticoids and oral clefting in animals; in humans the relationship is not clear but cannot be excluded. Corticosteroids are also associated with gestational diabetes and hypertension, and less commonly premature rupture of membranes and placental abruption.

Both prednisone and dexamethasone can be used to treat ITP. Recent data suggest equivalent efficacy in the short term.

The standard dosing regimen for dexamethasone is 40 mg po qd for 4 days, repeated every 14 days for 4 cycles. The standard prednisone dose is 1 mg/kg po qd until platelet count reaches acceptable level, followed by slow taper. Many hematologists will start all patients on 60 mg of prednisone po qd regardless of weight. In pregnant women early in gestation the starting dose of prednisone can be lower, e.g., 20 mg a day, with the goal of raising the platelet count to 30,000-50,000/ul. If there is no response after 4 weeks, steroids should be tapered.

Most obstetricians prefer prednisone, as fewer metabolites cross the placenta compared to dexamethasone; however, the actual difference may be not be significant because of differences in dose and schedule of administration.

Patients with presumed gestational thrombocytopenia

Patients with presumed gestational thrombocytopenia do not require treatment, but platelet counts need to be followed over time in case the diagnosis is, in fact, ITP. Since gestational thrombocytopenia usually presents in late pregnancy, when visits to the obstetrician usually occur every 2 weeks, the platelet count can be checked at the same time.

If there is no significant change over this time period, then concern for further decrease at time of labor and delivery is lessened. Standard labor and delivery practices can be followed, including the use of neuraxial anesthesia, although “cut-off” values for platelet counts to receive neuraxial anesthesia vary among institutions. A short trial of steroids or IV IgG can be given late in pregnancy (usually after 35-36 weeks) for patients with borderline platelet counts around 70,000-80,000/ul, in case the patient may have ITP or that there might be an increase in platelet count sufficient to allow neuraxial anesthesia.

What other therapies are helpful for reducing complications?

Patients with ITP can lose responsiveness to steroids or IV IgG, despite increasing the dose. In this case the next step is to use the combination of steroids and IV IgG, especially in the third trimester. Often high dose IV methylprednisolone (1000 mg) can be used to achieve an increase in platelet count.

If a pregnant patient with ITP is refractory to steroids and IV IgG early in pregnancy, then splenectomy is an option. Splenectomy is safe for both mother and fetus prior to 26 weeks gestation. After that time the size of the pregnant uterus makes it technically difficult. Response rates of 75% have been reported with duration of response usually sufficient to last through the remainder of the pregnancy.

IV anti-RhD has been used safely in pregnancy, although in one study additional treatment with steroids or IV IgG was required to obtain a platelet count greater than 50,000/ul for delivery. A mild decrease in hemoglobin occurs, usually less than 2 gm/dl. There were some early reports of anti-RhD causing acute renal failure due to hemoglobinemia or hemaglobinuria in 0.04% of non-pregnant patients. Patients must be Rhesus (Rh) positive in order for therapy to be effective. Newborns should be evaluated for jaundice. The advantage of anti-RhD is short administration time compared to IV IgG. Although complications are rare, it is infrequently used in pregnant patients who may already be anemic.

The use of rituximab, an anti-CD 20 monoclonal antibody, in pregnancy has been limited to patients with lymphoma during the second and third trimesters. There is one case report of the use of rituximab to treat ITP in the third trimester of pregnancy. In this case, the lymphocyte count of the infant was low at birth, but normalized by 6 months of age. Other case reports also note no obvious long-term adverse effects in lymphocyte populations, hematopoiesis or immunity in infants born to mothers treated with rituximab during pregnancy; however, data are limited and routine use in pregnancy cannot be recommended.

Thrombopoietin (TPO) mimetics bind to the thrombopoietin receptor and result in increased platelet production by initiating signal transduction through the JAK 2 tyrosine kinase pathway. No data are available for use in pregnancy; however, data will be collected from those patients who become pregnant while on TPO mimetics. The effects of TPO mimetics on the developing fetal bone marrow are unknown. There is also an increased risk of thrombosis in non-pregnant patients treated with TPO mimetics that could be potentially higher in the pregnant patient given the hypercoagulable state of pregnancy.

Azathioprine, cyclophosphamide, cyclosporine, vincristine and other cytotoxic or immunosuppressive agents have been used to treat ITP after the first trimester when organogenesis is complete. Although many report the safe use of these agents in pregnancy after the first trimester, use in pregnancy is generally avoided if possible. For severe refractory cases of ITP, use of other treatments including Campath, autologous and even allogeneic stem cell transplant have been reported; these can be considered for use after delivery.

Patients with chronic liver disease and thrombocytopenia from portal hypertension and resultant splenic sequestration benefit from platelet transfusions. Other ancillary therapies, such as lV IgG or steroids, have no impact on the platelet count in this situation.

What should you tell the patient and the family about prognosis?

Gestational thrombocytopenia

Gestational thrombocytopenia will resolve after delivery, with no sequelae. There is no associated risk of thrombocytopenia in the fetus. Platelet count should return to normal range after delivery, but gestational thrombocytopenia may recur in 20% of women who have subsequent pregnancies.

Idiopathic thrombocytopenic purpura

In ITP, the platelet count can decrease to extremely low levels requiring treatment during pregnancy. Patients will require close monitoring, especially in the third trimester as delivery approaches. Collaboration and coordination between hematologist, obstetrician, and anethesiologist is required to ensure safe delivery. ITP can resolve after delivery but may persist, however after delivery the therapeutic options are greater. The overall prognosis is good for patients with ITP, with excellent response to therapy. Up to 80% of patients treated for ITP relapse after the first course of treatment and require additional therapy. In many cases, ITP is a “chronic disease,” but is rarely life-threatening with appropriate treatment and care.

Idiopathic thrombocytopenic purpura and neonatal thrombocytopenia

In patients with ITP, there is anxiety about neonatal thrombocytopenia and subsequent hemorrhage, especially intracranial hemorrhage (ICH), due to birth trauma at delivery. The actual risks for bleeding in the neonate are low. Many studies of the platelet count in the fetus have found that the sampling methods, such as percutaneous umbilical cord blood sampling or fetal scalp blood sampling, are technically difficult, can give erroneous results, and have associated fetal bleeding risks that are equal to or greater than the risk of ICH.

The risk of thrombocytopenia in the baby is low. Less than 10% of neonates born to mothers with ITP have a platelet count less than 50,000/ul; less than 4% have a platelet count less than 20,000/ul. Intracranial hemorrhage in the newborn is rare and occurs in less than 1% of cases. Platelet counts in the newborn can be normal, but must be checked in the first few days after birth because the platelet count can rapidly decline due to circulating maternal antiplatelet antibodies (nadir up to 5 days after birth). Thrombocytopenia due to maternal ITP can persist in the newborn for 4 to 6 weeks after birth.

The only predictor of thrombocytopenia in the newborn is a prior history of thrombocytopenia at birth in an older sibling. Maternal platelet count, history of prior treatment including splenectomy, and other factors have been found not to correlate with thrombocytopenia in the newborn. Current guidelines suggest that vaginal delivery is safe for the infant. Maternal thrombocytopenia does not mandate C-section to decrease risk of ICH in the infant.

Secondary idiopathic thrombocytopenic purpura

Secondary ITP due to other autoimmune or viral processes has a prognosis directly related to the underlying disorder. Secondary ITP usually responds to treatment with steroids and IV IgG with generally similar good results as in patients with primary ITP. Treatment of the underlying disorder, however, often restores the platelet count to normal.

Thrombocytopenia due to processes associated with microangiopathic hemolytic anemia

Patients with thrombocytopenia due to processes associated with microangiopathic hemolytic anemia, including TTP, HUS, pre-eclampsia and HELLP syndrome, or AFLP can have poor prognosis with devastating maternal and fetal outcomes.

“What if” scenarios.

1. What if the pregnant patient with a diagnosis of ideopathic thrombocytopenic purpura does not respond to initial therapy?

If a patient with presumptive diagnosis of ITP and platelet count less than 70,000/ul fails to have any response to IV IgG or steroids, diagnosis of congenital thrombocytopenic disorder should be considered. Platelet and WBC morphology should be carefully examined, and genetic analysis for MPL, WAS, and RUNX1 mutations may be performed. Past history of thrombocytopenia or a family history of thrombocytopenia is helpful, as is history of bleeding.

Patients with congenital thrombocytopenia can have more pronounced thrombocytopenia during pregnancy. Patients with ITP can fail to respond to first line therapy, and more aggressive therapy with IV IgG (1gram/kg per day for two days) or high dose IV methylprednisolone (1 gram per day for two days), or combination of both, can be used. Duration of response can be short, in which case splenectomy can be considered prior to 26 weeks gestation.

2. What if a patient presents in labor with no history or prior records and is thrombocytopenic?

If delivery is occurring there will be no time to worry about platelet threshold for epidural anesthesia, or even whether to administer it. The focus will then be on preventing postpartum hemorrhage. Blood products, including platelets, can be used, as long as a diagnosis of TTP or HUS has been ruled out. Evaluation for the microangiopathic disorders must occur rapidly so that appropriate therapy can be instituted if necessary.

In a thrombocytopenic patient whose labor is progressing more slowly, there is more time to make a diagnosis; however, if there is significant bleeding or platelet count is less than 30,000/ul, platelet transfusions should be given. Patients with ITP may respond transiently to platelet transfusions.

3. What if a patient with ideopathic thrombocytopenic purpura presents in labor and platelet count is low?

The generally accepted threshold for “safe” platelet count for labor and delivery is above 50,000/ul. This is based on expert consensus, with concern for possible C-section. There are reports of pregnant women delivering with lower platelet counts and no significant bleeding. General anesthesia can be used in place of neuraxial anesthesia if the patient requires C-section and the platelet count is above 50,000/ul, but below the institutional threshold for neuraxial anesthesia. If the platelet count is above 30,000/ul, the patient can be observed for excess bleeding and platelet transfusions given if bleeding occurs. A platelet count lower than 30,000/ul may require platelet transfusion.

4. What should the platelet count be for neuraxial anesthesia?

Epidural hematoma is a complication of neuraxial anesthesia that occurs rarely, but can cause serious complications. The incidence in the obstetric patient population is estimated to be 1:200,000, with the risk of permanent neurologic deficits approximately 1:237,000.

Thrombocytopenia increases the risk of epidural hematoma. There is less risk in thrombocytopenic patients with normal platelet function, such as in ITP or May-Hegglin anomaly. Patients with coagulopathies may have a higher bleeding risk and, hence, risk for epidural hematoma – especially when thrombocytopenia is also present. A platelet count of 100,000/ul with no other bleeding risks is accepted as being adequate for neuraxial anesthesia. There has been a trend over the last two decades to accept an even lower threshold, between 50,000–100,000/ul; however, institutional practices vary widely.

5. What are the risks of thrombocytopenia for the baby?

Thrombocytopenia in the newborn and increased risk of bleeding, especially intracranial hemorrhage (ICH), is of concern when the mother has thrombocytopenia due to antibody-mediated platelet destruction and trans-placental passage of antibodies in ITP. There is a 0-1% risk of ICH, with approximately 4% chance of platelet count less than 20,000/ul, 10% chance less than 50,000/ul, and 15% chance less than 100,000/ul.

The infant’s platelet count can be normal at birth, but then drop during the first week after delivery due to persistent circulating maternal anti-platelet antibodies. It is important that the infant be monitored closely during this time period. A diagnosis of neonatal alloimmune thrombocytopenia (NAIT) should be considered in any newborn found to have thrombocytopenia if the mother has a normal platelet count.


1. Thrombocytopenia due to a microangiopathic condition requires immediate attention and treatment. Significant advances in understanding the pathophysiology of these processes have been made in the last 10-20 years. The etiology of the underlying microangiopathic process depends on the type of process: antibody to ADAMTS 13 in acquired TTP, inherited abnormality with decreased levels of ADAMS TS 13 in congenital TTP (Upshaw-Shülman syndrome), abnormal complement regulatory proteins such as factor H, factor I, or membrane cofactor protein, either from mutations or autoantibodies, in congenital or acquired diarrhea-associated HUS, shiga toxin in acquired HUS.

Pre-eclampsia/eclampsia and HELLP syndromes are disorders of maternal/placental vascular endothelium and placental function that are beginning to be elucidated. Abnormalities in placental development lead to placental hypoperfusion and ischemia with resultant changes in inflammatory cytokines, placenta-derived factors, and growth factors. Recent data suggest roles for endoglin (endothelial cell-derived member of TGF-Beta receptor family), sFlt1 (soluble VEGF receptor type 1), and placenta derived growth factor (PLGF). These molecules affect vascular endothelial cells, ultimately resulting in platelet activation and formation of micro-vascular thrombi. Testing for any of these factors to make a diagnosis is not practical, however, as therapy needs to be started as soon as possible. Turnaround times are slow for many of these tests, and many are not even available at this time for clinical use.

2. The etiology of gestational thrombocytopenia is not clear. It is generally accepted that there is a dilutional component to thrombocytopenia in pregnancy. There are some data to support increased activation and clearance of platelets that is not due to an anti-platelet antibody-mediated process. Since gestational thrombocytopenia resolves after delivery, it must be due to some pregnancy-specific factors.

3. In lupus anticoagulant(LAC)-associated thrombocytopenia, antiphospholipid antibodies bind to platelet phospholipids and increase platelet clearance. LAC can occur in patients with SLE or in patients without other underlying autoimmune disorders. Clinical manifestations attributed to LAC include venous or arterial thrombosis, or pregnancy loss either early or late in gestation. There are some data that suggest that antiphospholipid antibodies predispose to pre-eclampsia.

4. The underlying pathophysiology of acute fatty liver of pregnancy in some cases is due to abnormalities in intra-mitochondrial fatty acid beta-oxidation associated with both maternal and fetal mutations in long chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD). Excess hepatotoxic fatty acids accumulate, resulting in maternal liver and vascular damage with DIC (thrombocytopenia, consumptive coagulopathy, and bleeding).

What other clinical manifestations may help me to diagnose thrombocytopenia in pregnancy?

It is important to determine if thrombocytopenia existed prior to pregnancy, or if there is a history of prior bleeding.

Signs of bleeding in ITP, such as petechiae or ecchymoses, correlate with platelet count. Manifestations of bleeding at higher platelet counts point to other diagnoses, including ones with associated coagulopathies. Gestational thrombocytopenia, the most common etiology of thrombocytopenia in pregnancy, should have no unusual signs or findings on physical exam. Splenomegaly or lymphadenopathy suggest that thrombocytopenia is secondary to the primary diagnosis.

What other additional laboratory studies may be ordered?

Testing for congenital thrombocytopenic disorders and inherited marrow failure syndromes that can first manifest as thrombocytopenia include genetic analysis for MPL, WAS, and RUNX1 mutations that can be obtained after initial testing has excluded other etiologies of thrombocytopenia in a patient with a suggestive family history. More detailed platelet analysis include aggregation studies, electron microscopy, or flow cytometry for expression of platelet surface proteins (e.g., GP Ib-IX-V for Bernard -Soulier syndrome; GP IIb-IIIa for Glanzmann’s thrombasthenia) when considering a diagnosis of an inherited platelet disorder.

Measurements of global coagulation status have been used to try to assess bleeding risk in a wide variety of surgical patients, including pregnant patients, prior to neuraxial anesthesia. Thromboelastography (TEG) and platelet function analyzer (PFA-100) have been studied in the pregnant patient. It has been proposed that normal TEG results are a more important determinant for risk of bleeding in the pregnant patient with ITP than actual platelet number. While results are intriguing, data are limited and use of these tests for clinical decision making is not recommended at this time.

What’s the evidence?

Boehlen, F, Hohlfeld, H, Extermann, P, Perneger, TV, de, Moerloose. “Platelet count at term pregnancy: a reappraisal of the threshold”. Obstet Gynecol. vol. 95. 2000. pp. 29-33. [Prevalence of thrombocytopenia and correlation with maternal and fetal morbidity in late pregnancy assessed in 6770 women with definition of new lower limit of normal platelet count in pregnant women.]

Burrows, RF, Kelton, JG. “Thrombocytopenia at delivery: a prospective survey of 6715 deliveries”. Am J Obstet Gynecol. vol. 162. 1990. pp. 731-734. [Documents the incidence of all types of thrombocytopenia at delivery in 6715 cases.]

Samuels, P, Bussel, JB, Braitman, LE. “Estimation of the risk of thrombocytopenia in the offspring of pregnant women with presumed immune thrombocytopenic purpura”. N Engl J Med. vol. 323. 1990. pp. 229-235. [Prospective study of outcomes of 162 consecutive pregnancies affected by maternal ITP with calculation of risk of neonatal thrombocytopenia, and associated maternal characteristics that might predict severity of neonatal thrombocytopenia.]

Burrows, RF, Kelton, JG. “Fetal thrombocytopenia and its relation to maternal thrombocytopenia”. N Engl J Med. vol. 329. 1993. pp. 1463-1466. [Evaluation of platelet count in 15,932 newborns and correlation with maternal count to document incidence of thrombocytopenia in newborns, correlation with diagnoses on mothers, and risk for intracranial hemorrhage in the newborn.]

Ruggeri, M, Schiavotto, C, Castaman, G, Tosetto, A, Rodeghiero, F. “Gestational Thrombocytopenia: A prospective study”. Haematologica. vol. 82. 1997. pp. 341-34. [Prospective study of 37 consecutive cases of gestational thrombocytopenia with outcomes in 41 pregnancies.]

Webert, KE, Mittal, R, Siguoin, C, Heddle, NM, Kelton, JG. “A retrospective 11 year analysis of obstetric patients with idiopathic thrombocytopenic purpura”. Blood. vol. 102. 2003. pp. 4306-4311. [Retrospective review of morbidity associated with ITP during 119 pregnancies on 92 women over 11 years.]

Veneri, D, Franchini, M, Raffaelli, R. “Idiopathic thrombocytopenic purpura in pregnancy: analysis of 43 consecutive cases followed at a single Italian institution”. Ann Hematol. vol. 85. 2006. pp. 552-554. [Review of treatment and outcome of 43 consecutive cases of ITP during pregnancy.]

Sibai, B, Dekker, G, Kupferminic, M. “Pre-eclampsia”. Lancet. vol. 365. 2005. pp. 785-799. [Review of findings on the diagnosis, risk factors, and pathogenesis of pre-eclampsia and the present status of its prediction, prevention, and management.]

George, JN. “The thrombotic thrombocytopenic purpura and hemolytic uremic syndromes: evaluation, management, and long-term outcomes experience of the Oklahoma TTP-HUS Registry, 1989-2007”. Kidney Int. vol. 112. Supp. 2009. pp. S8-S10. [Review of 382 consecutive patients with TTP or HUS from a community perspective with report on cases of pregnancy-associated TTP.]

Martin, JN, Bailey, AP, Rehberg, JF. “Thrombotic thrombocytopenic purpura in 166 pregnancies: 1955-2006”. Am J Obstet Gynecol. vol. 199. 2008. pp. 98-104. [Review of reports on TTP in pregnancy with review of time of presentation during pregnancy, prevalence of abnormal lab findings, and incidence of mortality.]

Fujimura, Y, Matsumoto, M, Kokame, K. “Pregnancy-induced thrombocytopenia and TTP, and the risk of fetal death, in Upshaw-Schulman syndrome: a series of 15 pregnancies in 9 genotyped patients”. Br J Haemtol. vol. 144. 2009. pp. 742-754. [Small number of women with known ADAMTS13 mutations prospectively followed with report of pregnancy outcomes.]

Ibdah, JA, Bennett, MJ, Rinaldo, P. “A fetal fatty-acid oxidation disorder as a cause of liver disease in pregnant women”. N Engl J Med. vol. 340. 1999. pp. 1723-1731. [High proportion of women with AFLP or HELLP have mutation in long-chain hydroxyacyl-CoA dehydrogenase.]

Ruppen, W, Derry, S. “Incidence of epidural hematoma, infection, and neurological injury in obstetric patients with epidural analgesia/anaesthesia”. Anesthesiology. vol. 105. 2006. pp. 394-399. [Review of 27 recent studies reporting on adverse events for epidural use for childbirth in 1.37 million women; estimates of risk for various complications were calculated.]