VTE in pregnancy

Venous Thromboembolism (VTE) in Pregnancy

1. What every clinician should know

Clinical features and incidence

Venous thromboembolism (VTE) occurs in approximately 0.5-2 women per 1000 in the antepartum period or the first 6 weeks postpartum and thus should be considered in the differential diagnosis when a pregnant or postpartum woman presents with characteristic symptoms. A sizeable majority of cases of deep vein thrombosis (DVT) associated with pregnancy occur in the left lower extremity, particularly in the iliac or femoral system. This predisposition has widely been attributed to compression of the left common iliac vein by the right iliac artery and enlarging gravid uterus.

Pregnancy itself poses a substantial increase in risk for developing VTE. The antepartum risk of VTE is at least five-fold that of the non-pregnant state, and the postpartum risk is even higher. Pulmonary embolus (PE) is more likely to be diagnosed in the postpartum period.

Risk factors

There are numerous risk factors for VTE in pregnancy. The most important ones are previous VTE (provoked or unprovoked), known thrombophilia, medical comorbidities (e.g., SLE), and surgical procedure in pregnancy or in the first 6 weeks postpartum. As a general rule, individual pregnant patients ought to undergo VTE risk assessment to determine the best course of care during the antepartum and postpartum periods.

Thrombus other than DVT or PE

A. Superficial venous thrombosis (SVT)

Superficial venous thrombosis presents as thrombosis and inflammation in a superficial vein. It involves the great saphenous vein in up to 70% of cases, the small saphenous vein in 15% of cases and upper extremity veins in a small percentage of cases. It is widely thought that it usually occurs due to local venous inflammation, with varicose veins probably the most common precursor.

Though the diagnosis is clinical, it is prudent to obtain a duplex ultrasound in patients in whom apparent great saphenous superficial venous thrombosis is anything other than small or relatively symptom free. In particular, the status of the sapheno-femoral junction (SFJ) and sapheno-popliteal junctions (SPJ) should be determined with regard to their relative distance from the thrombosed segment, possible extension of the clot into the deep system and whether the junctions are incompetent (i.e. demonstrate reflux).

The extent of superficial thrombosis should be documented, and evaluation for a concomitant DVT, present in over 25% of cases, must be completed. Patients with superficial venous thrombosis in close proximity to the SFJ are generally anticoagulated, even though the evidence for progression into the deep venous system is weak. The treatment of superficial venous thrombosis in close proximity to the SPJ is more controversial. The decision to treat with anticoagulants also is influenced by factors such as first episode vs. recurrence, presence or absence of varicose veins and history of hypercoagulable disorders.

For those cases of superficial venous thrombosis in which anticoagulant treatment is judged appropriate during pregnancy, experts recommend prophylactic or intermediate doses of LMWH or intermediate doses of UFH for at least 4 weeks.

B. Cerebral Venous Thrombosis

Thromboses within the venous system draining the brain may involve the midline sagittal sinus, the transverse sinuses or related smaller venous structures. Collectively called cerebral venous thrombosis, the condition occurs in approximately 1-11/10,000 deliveries. It is most commonly diagnosed postpartum. The most frequent presenting symptom is severe headache. 60-80% of patients will present with seizures, and 29-50% will have some evidence of paresis. The gold standard for diagnosis is MR venography, which is safe in pregnancy. Patients must be evaluated for evidence of elevated intracranial pressure (ICP), including fundoscopic examination to evaluate for papilledema.

Treatment includes anticoagulation, hydration and control of symptoms, including anti-epileptics if seizures are present. If the patient is postpartum, mannitol may be used to decrease cerebral edema; however, its use during pregnancy is controversial because of concerns for uteroplacental insufficiency. In the pregnant patient other options are available, including dexamethasone. It is important to note that anticoagulation is indicated in these patients even in the presence of cerebral hemorrhage. Consultation with a neurologist is an important component of the care of these patients.

Two obstetric concerns may arise in a pregnant patient with cerebral vein thrombosis – mode of delivery and regional anesthesia. Patients with cerebral venous thrombosis can have a vaginal birth, in the absence of other obstetric contraindications, with strong consideration of scheduled induction of labor. Neuraxial anesthesia is not contraindicated in patients with cerebral venous thrombosis other than the risks associated with anticoagulation (see section below on neuraxial anesthesia). If there is evidence of elevated intracranial pressure (ICP) at the time of labor, the patient should have adequate anesthesia to suppress involuntary pushing and should have an assisted second stage of labor in order to avoid further increases in ICP caused by Valsalva.

What is the cause of the disease?

All elements of Virchow’s Triad – stasis, vascular trauma and hypercoagulability – are present during pregnancy and the postpartum period. Lower extremity venous stasis, as reflected by decreased venous flow velocity, worsens as pregnancy advances. Additionally, pregnancy-related venous distention may result in endothelial damage and prothrombotic changes in the endothelium. These vascular changes do not normalize until nearly 6 weeks postpartum.

Vascular trauma in the form of endothelial damage may occur due to venous distention during pregnancy but is certainly present in the setting of preeclampsia and the associated vascular endothelial activation. During normal delivery, venous compression may occur. Operative and assisted deliveries are thought to contribute to vascular trauma, also possibly contributing to the risk of thrombosis in the postpartum period.

Changes in the coagulation cascade

Normal pregnancy is accompanied by dramatic changes in the hemostatic system that one would characterize as “hypercoagulable.” A reasonable assumption is that these evolved for the prevention of hemorrhage at the time of delivery. Most clotting factors increase, some anticoagulants decrease and fibrinolytic activity decreases.

With regard to specific factors, factors II, VII, VIII, IX, XII and von Willebrand factor increase throughout pregnancy. Fibrinogen levels increase to almost twice that of the non-pregnant state. Anticoagulant changes include decreased free and total protein S antigen levels, as well as decreased activity, all of which are present very early in pregnancy. Though protein C levels remain unchanged, an overall increase in activated protein C resistance is present, with the degree of resistance dependent on several modifiers, including the presence of the factor V Leiden mutation and thrombin generation. Fibrinolysis is decreased, predominantly due to diminished tissue plasminogen activator activity and increases in plasminogen activator inhibitor-1 and -2 and thrombin activatable fibrinolysis inhibitor.


A. Hereditary Thrombophilias

The role of hereditary thrombophilias in VTE in pregnancy has seen considerable emphasis within the last two decades, and it is now recognized that up to 50% of pregnancy-associated VTEs occur in women with hereditary thrombophilias. The VTE risk per pregnancy is probably highest for individuals homozygous for the prothrombin 20210 mutation or factor V Leiden, individuals heterozygous for both the prothrombin mutation and factor V Leiden, or for women with antithrombin deficiency. The VTE risk per pregnancy is lower for protein C and S deficiency and heterozygosity for factor V Leiden mutation or the prothrombin mutation. Despite the understanding that heritable thrombophilias are clearly associated with pregnancy-associated VTE, there is insufficient evidence to support screening all pregnant women for thrombophilias.

Depending upon the population being studied, up to 15% of unselected individuals may be found to carry a recognized heritable thrombophilia. One-third or more of individuals with their first acute DVT carry a heritable thrombophilia, and the presence of a heritable thrombophilia poses an increased risk for VTE in pregnancy. While it may seem logical to test individuals presenting with a first-time VTE for heritable thrombophilia, the practice of doing so in all such patients is debatable for several reasons. First, the most common heritable thrombophilias do not substantially increase the risk of VTE recurrence. Second, the likelihood of discovering a more serious thrombophilia that would require long-term anticoagulation is small in the absence of compelling family history. And finally, testing is costly.

Recently however, and with specific reference to pregnancy, the American College of Obstetricians and Gynecologists has recommended consideration of thrombophilia screening (to include tests for factor V Leiden, prothrombin mutation, protein C deficiency, protein S deficiency, and antithrombin deficiency) in women with:

  • a personal history of VTE that was associated with a nonrecurrent risk factor (e.g. a fracture or surgery).

  • a first degree relative with a high-risk thrombophilia or VTE before age 50 years in the absence of other risk factors.

In pregnancy, it is notable that neither methylene tetrahydrofolate reductase (MTHFR) gene polymorphisms nor the finding of elevated homocysteine levels are significantly associated with VTE; therefore, testing for either of these is not recommended.

See below under risk assessment for further discussion regarding risk of thrombosis and suggested treatment regimens.

In women with two or more VTE events, thrombophilia testing is unnecessary for pregnancy considerations because these women should be treated during the antepartum and postpartum periods regardless of the etiology, with some consideration given to whether or not they are on long-term anticoagulation outside of pregnancy. In women with a single prior VTE associated with a nonrecurrent risk factor such as orthopedic surgery who would prefer to avoid antepartum anticoagulation, testing might be considered in order to better assess their risk in the current pregnancy.

In pregnancy, testing for genetic mutations (Factor V Leiden and prothrombin gene mutation) can be done successfully, as well as antibody testing for evaluation of APS. Other tests, such as those for antithrombin deficiency and protein C and S deficiencies, may be less reliable in pregnancy or while on certain anticoagulants and are best deferred until after 6 weeks postpartum.

B. Acquired thrombophilias

There are numerous acquired thrombophilias, the most important of which is antiphospholipid syndrome (APS), an autoimmune condition associated with venous and arterial thrombosis, fetal loss and early delivery for preeclampsia or severe placental insufficiency. The diagnosis is made when an individual with a clinical feature of APS repeatedly tests positive for the antibody markers of APS, either lupus anticoagulant, anticardiolipin antibodies or anti-β2-glycoprotein I antibodies.

It is prudent to investigate women with VTE for APS because those confirmed to have the condition are candidates for thromboprophylaxis during pregnancy, regardless of the nature of the VTE event. In addition, APS is associated with complications of pregnancy that require modification of usual care, such as serial fetal sonography and fetal surveillance testing.

2. Diagnosis and differential diagnosis

Establishing the diagnosis

The diagnosis of DVT in non-pregnant patients is usually objectively made by venous compression ultrasonography (CUS), a popular imaging choice in pregnancy because it is non-invasive and has acceptably high sensitivity and specificity as judged by comparison to venography in symptomatic patients. Though there are no large, prospective studies in pregnant women suspected of having DVT, a clearly abnormal compression ultrasonography in a pregnant woman is sufficient for the diagnosis of DVT in the popliteal, femoral or iliac vein and for initiation of treatment.

A negative compression ultrasonography, however, is probably not as reassuring in the pregnant patient as in the non-pregnant patient because compression ultrasonography does not perform well in the detection of isolated iliac thrombosis, a diagnosis thought to be more common in pregnant patients. Pregnant women suspected of having DVT in whom the initial CUS is negative and in whom there is clinical suspicion of iliac vein thrombosis should undergo either duplex Doppler, MRI or venography. Those in whom iliac vein thrombosis is not suspected should undergo repeat CUS on days 6-8 after presentation or sooner if clinical suspicion is high to evaluate for proximal extension of distal thrombosis.

With regard to the diagnosis of PE in a patient in whom it is suspected, none of the usual diagnostic tools have been well studied in pregnant women. Depending upon the availability of equipment and expertise, three inital approaches are acceptable: CTPA, VQ scan or bilateral lower extremity compression ultrasonography. Each approach has advantages and disadvantages.

VQ scanning delivers higher radiation doses to the fetus than does CTPA, but CTPA delivers a higher dose of radiation to the mother, increasing the estimated lifetime risk of breast cancer. VQ scanning and bilateral compression ultrasonography require no further testing when high probability or positive, respectively; however, any other result from either test requires additional testing to confirm or refute the diagnosis. CTPA is associated with a very low number of non-diagnostic results.

The algorithm in Figure 1 provides a useful approach to diagnosis.

Figure 1.

A useful approach to diagnosis

Differential diagnosis

The differential diagnosis of DVT includes benign lower extremity edema and cellulitis. The diagnosis of DVT is usually suspected when there is asymmetry in the edema, such that one extremity has a larger circumference than the other. In a pregnant woman presenting with new onset edema of one extremity, erythema and/or a palpable firm, tender mass consistent in shape and location with a venous structure, the evaluation of DVT should be done with compression ultrasonography in order to first rule out the most potentially dangerous diagnosis.

The differential diagnosis of a pregnant woman presenting with chest pain is broad; however, PE should always be in the forefront of the physician’s mind. Other items in the differential diagnosis include gastroesophageal reflux disease, pericarditis, acute MI, pneumothorax, pneumonia and musculoskeletal pain. As always, a thorough history and physical exam should precede any laboratory or radiologic evaluation.

The role of D-dimer in diagnosis

D-dimer, a degradation product of cross-linked fibrin, is found in increased concentration in the circulation of patients with VTE. Using an appropriate threshold for the specific laboratory test being used, modern D-dimer tests have very good sensitivity and specificity with regard to the diagnosis of VTE in non-pregnant patients with low or intermediate probability of disease. However, circulating D-dimer increases in normal pregnant women with advancing gestational age, limiting the utility of this test in pregnancy until appropriate thresholds are established.

3. Management

Medical treatment of acute VTE

The vast majority of cases of acute VTE in pregnancy (proximal DVT and uncomplicated PE) are managed according to well-defined protocols using unfractionated heparin (UFH) or low molecular weight heparin (LMWH); these are available through most hospital pharmacies. Accumulated experience in a variety of non-pregnant patients, as well as the overall clinical experience in pregnant women over the last decade, support the use of LMWH in pregnant patients, and it should be the first line of treatment in pregnant women.

Compared to LMWH, UFH has the disadvantage of requiring hospitalization for at least several days in order to monitor the aPTT and adjust the dose of UFH. However, UFH has one major advantage over other anticoagulants: it is fully reversible with protamine sulfate, making UFH a wise anticoagulant choice in some clinical settings in pregnancy.

Unfractionated heparin

There are two recognized regimens for treatment using a continuous intravenous infusion of unfractionated heparin. For both regimens, it is imperative to get an aPTT prior to starting heparin.

A small proportion of patients treated with UFH demonstrate heparin resistance, likely due to increased heparin clearance, increased heparin-binding proteins or increased levels of factor VIII. Though rare, antithrombin deficiency also results in apparent lack of the usual heparin effect on the aPTT. As a general rule, patients who require more than 35,000 U of UFH per day to elevate their aPTT should be monitored based on anti-Xa levels, aiming to achieve a target within the range of 0.35 to 0.7 U/mL.

Unfractionated heparin regimen A

Table I. Heparin dose-adjustment nomogram for Regimen A.

Table I.
aPTT, seconds Repeat Bolus Dose, U Stop Infusion, mins Change Rate of Infusion mL/hr at 40 U/mL (U per 24 hours) Time of Next aPTT, hrs
<50 5,000 U bolus 0 +3 (+2,880) 6
50-59 0 +3 (+2,880) 6
60-85 0 0 (0) 12
86-95 0 -2 (-1,920) 12
96-120 30 -2 (-1,920) 6
>120 60 -4 (3,840) 6

Unfractionated heparin regimen B

Table II. Heparin dose-adjustment nomogram for Regimen B.

Table II.
Variable Adjustment
Initial dose 80 U/kg bolus, then 18 U/kg/hr
aPTT <35 secs 80 U/kg bolus, then increase 4 U/kg/hr
aPTT 35-45 secs 40 U/kg bolus, then increase 2 U/kg/hr
aPTT 46-70 secs No change
aPTT 71-90 secs Decrease infusion rate by 2 U/kg/hr
aPTT >90 secs Hold infusion for 1 hr, then decreaseinfusion rate by 3 U/kg/hr

Low molecular weight heparin

LMWH has largely replaced UFH in most clinical settings, including pregnancy. LMWH has excellent bioavailability and a more predictable anticoagulant response than UFH. The LMWH fragments have reduced binding to proteins and cells, a feature that in part explains the more predictable dose-response of LMWH. LMWH also has reduced activity with regard to thrombin inactivation, leading to a less profound effect on the aPTT. The major anticoagulant effect of LMWH lies in its enhancing antithrombin mediated inactivation of factor Xa, an effect measurable only via anti-Xa assays. Dosing is weight based, and the route of treatment is subcutaneous (SC), even for acute VTE.

Pregnancy likely increases renal clearance of LMWH, and other as yet poorly characterized pregnancy-related alterations in LMWH metabolism may be at play. Depending upon the risk of VTE, monitoring of LMWH action should be performed fairly frequently in pregnancy, probably at least once every 2-4 weeks in most cases. Many experts recommend monitoring peak LMWH anti-Xa levels in pregnant women to achieve levels of 0.6-1.2 U/mL at 3-3.5 hours after SC injection. Monitoring LMWH activity is particularly important in pregnant women with impaired renal or hepatic function and for women who weigh less than 50kg or more than 100kg.

There are several LMWH products available in the US, and these each have slightly different pharmacokinetic properties. It is best to become familiar with and use one LMWH product.Enoxaparin is perhaps the most widely used of these in pregnancy in the U.S, and it is dosed at 1mg/kg SC BID. One study in pregnant patients concluded that twice daily dosing of enoxaparin is superior to once daily dosing in terms of maintaining the targeted circulating levels of LMWH activity.

Duration of treatment

There is no evidence-based approach regarding the duration of heparin treatment for acute VTE in pregnancy. Two points of detail deserve discussion. First, if the patient has an acute VTE in early pregnancy and is successfully treated for 20 weeks or more, it is reasonable to consider reducing the heparin dose to thromboprophylactic levels in the late third trimester in light of the potential risk for delivery-related hemorrhage and with an eye towards the potential need for neuraxial anesthesia.

A second important caveat is that postpartum anticoagulation is important regardless of how early in pregnancy the patient had her acute VTE. Thus, even if the patient was managed on a reduced dose of heparin in the late third trimester, she should be therapeutically anticoagulated for 6 weeks postpartum.

Other treatment modalities for acute VTE


Thrombolytic therapy is gaining favor in non-pregnant patients with PE. Tissue plasminogen activator does not cross the placenta; however, hemorrhage is a major risk associated with thrombolysis, with an overall major bleeding rate of approximately 10-15% and an intracranial hemorrhage rate of 2-5%. Thrombolysis at the time of cesarean or delivery would pose considerable risk for hemorrhage and should be avoided. Other than at the time of delivery, it appears that thrombolytic therapy is a treatment option in cases of life-threatening PE in pregnancy.

Physical treatment

A. Early ambulation

In the past, patients with acute DVT were placed on bed rest for as long as 1 week or more due to concerns that vigorous movement of the lower extremity might result in disruption of the thrombus and increased risk of PE. Recent evidence suggests there is no difference in risk between ambulation and bed rest for development of PE, progression of existing DVT or development of a new DVT.

B. Compression hose

Some evidence suggests that the long-term use of graduated elastic compression stockings prevents post thrombotic syndrome after symptomatic proximal DVT. These stockings are below the knee devices to be worn during the day (when up and about) or longer.

Invasive treatment

A. Vena cava filter

Caval filters are indicated in patients at high risk for PE or recurrent PE in spite of anticoagulant treatment or in whom anticoagulation is contraindicated. Thus, they may be indicated in the pregnant patient who has recently had a VTE, especially a PE, and who is undergoing delivery such that reversal of anticoagulation is indicated. They can be placed via the jugular or femoral vein. In general, removable filters are preferred because the presence of long-standing filters is associated with an increased subsequent risk of lower extremity DVT without an increase in survival rate.

B. Embolectomy

Embolectomy for massive PE may rarely be used in life-threatening cases, though thrombolysis would seem a good first approach in all but the most unusual case. The need for cardiopulmonary bypass in the setting of a grave maternal status is associated with a high stillbirth rate.

Thromboprophylaxis in pregnancy

Women at risk for VTE in pregnancy should be managed with heparin thromboprophylaxis beginning as early in pregnancy as is practical. Heparin is the anticoagulant drug of choice for VTE prophylaxis during pregnancy. Neither UFH nor LMWH heparin cross the placenta. The risk and safety profile of LMWH makes it the preferred agent, though cost considerations have resulted in the frequent use of UFH in some parts ofthe U.S.

Those not on long-term anticoagulation should begin heparin after the demonstration of a normally rising hCG or after identification of a nearly intrauterine pregnancy. Women being treated with warfarin for long-term anticoagulation and who are attempting pregnancy should be monitored frequently for pregnancy using sensitive hCG tests. One approach is cycle day 25-27 serum hCG screens. Once very early pregnancy is diagnosed, warfarin should be discontinued and heparin started; this should be accomplished no later than 6 weeks gestation to avoid the risk of warfarin embryopathy. Of course, it is prudent to establish that the pregnancy is likely normal by appropriately rising hCG levels or ultrasound confirmation of an early intrauterine pregnancy.

Risk assessment

All women seeking medical advice about pregnancy or presenting for a first prenatal visit should be assessed for VTE risk.

Women with:

  • acute VTE in the current pregnancy

  • recent acute VTE (within several months)

  • previous recurrent VTE (two or more events)

will require adjusted-dose (therapeutic) heparin during pregnancy, and anticoagulation with heparin or warfarin for at least 6 weeks postpartum; management should be individualized after 6 weeks.

Other women will present with a constellation of risk factors and their individual risk should be assessed. The risk assessment tool below combines features of the risk assessment recommended by the Royal College of Obstetricians and Gynaecologists and the American College of Obstetricians and Gynecologists and other expert opinion. To use the tool, calculate the sum of the risk factors present in your patient and treat according to the recommendations in Table III.

Table III.
Pre-existing risk factors Score
Previous VTE – unprovoked or estrogen related 4
Previous VTE – provoked, with known thrombophilia 4
Known high risk thrombophilia without previous VTE* 4
Previous VTE – single episode, provoked, without thrombophilia 3
Medical comorbidities** 2
Obesity, BMI >= 40kg/m2 (based on BMI in early pregnancy) 2
Known low risk thrombophilia without previous VTE*** 1
Family history of VTE (1st degree relative with event <age 50 years) 1
Age > 35 years 1
Obesity, BMI 31-39 kg/m2 (based on BMI in early pregnancy) 1
Parity > 2 1
Current smoker 1
Gross varicose veins 1
Obstetric risk factors in current pregnancy Score
Cesarean in labor 2
Preeclampsia 1
Dehydration/hyperemesis/OHSS 1
Multiple pregnancy or ART 1
Elective cesarean 1
Mid-pelvis forceps or rotational forceps 1
Prolonged labor (>24 hours) 1
Peripartum hemorrhage (>1 liter) 1
Transient risk factors
Surgical procedure in the pregnancy or <= 6 weeks postpartum 2
Current systemic infection 1
Immobility 1

* includes homozygosity or compound heterozygosity for Factor V Leiden or prothrombin gene mutation (20210), persistently positive lupus anticoagulant or antithrombin deficiency

** e.g., SLE, current cancer, nephrotic syndrome

*** includes heterozygosity for Factor V Leiden or prothrombin gene mutation (20210), protein C deficiency and protein S deficiency

Table IV. Management Recommendation Based on Risk Score

Table IV.
Total Score Antepartum Management Postpartum Management
>= 4 Prophylactic or intermediate-dose heparin during pregnancy Intermediate-dose heparin or warfarin anticoagulation for 6 weeks postpartum
3 Surveillance without anticoagulation or prophylactic dose heparin Surveillance acceptable if no previous VTE; if previous VTE intermediate-dose heparin or warfarin anticoagulation for 6 weeks postpartum
1-2 Surveillance without anticoagulation or prophylactic dose heparin Surveillance without anticoagulation therapy
1 or more If postoperative or on bedrest: SCDs or prophylactic dose heparin. Encourage early ambulation when appropriate. If postoperative or on bedrest: SCDs or prophylactic dose heparin. Encourage early ambulation when appropriate.

Specific comments regarding selected risk factors

A. Single prior VTE episode, and not on long-term anticoagulants

Patients whose single, prior VTE occurred without provocation or was associated with estrogen treatment or pregnancy should receive either prophylactic or intermediate-dose UFH or LMWH during pregnancy. Those whose single, prior VTE was associated with a transient risk factor and who do not have a thrombophilia are candidates to avoid treatment during pregnancy but must be cautioned regarding the signs and symptoms of VTE and what measures to take to decrease risk. In addition, the physician must take the entire clinical picture, e.g. obesity or bedrest, into account. All patients with a single, prior VTE should be given postpartum thromboprophylaxis.

B. Antiphospholipid syndrome

Women without prior VTE and diagnosed with antiphospholipid syndrome because of pregnancy morbidity should receive either prophylactic or intermediate-dose UFH or LMWH during pregnancy. In the setting of definite APS, we and others have suggested prophylactic UFH be 7500 to 10000 units SC q 12 hours and LMWH be dosed in a q 12 hour regimen. Following delivery, postpartum thromboprophylaxis with warfarin or LMWH is indicated.

C. Inherited thrombophilia

Though uncommon, antithrombin deficiency, homozygosity for Factor V Leiden or prothrombin gene mutation, and compound heterozygosity for FVLM and PGM are considered as high risk for thrombosis during pregnancy even if the patient has not previously had a VTE. We favor either intermediate-dose or adjusted-dose UFH or LMWH with antithrombin deficiency (and monitoring of anticoagulant effect); for the other mutations we recommend prophylactic or intermediate dose UFH or LMWH during pregnancy and postpartum thromboprophylaxis.

Women without a prior VTE who are heterozygous for Factor V Leiden or prothrombin gene mutation or who have protein C or S deficiency can be managed without antepartum thromboprophylaxis if the remainder of their individualized risk assessment proves acceptable. Postpartum care may be similarly individualized.

Anticoagulant dosing

The language used in the literature to describe heparin thromboprophylaxis can be somewhat confusing. Depending on the source, some guidelines distinguish between “prophylactic,” “mini dose prophylactic” and “intermediate-dose” heparin, with the latter referring to an UFH regimen given subcutaneously every 12 hours adjusted to an anti-Xa level of 0.1-0.3 U/mL or a LMWH regimen every 12 hours and adjusted for weight. Full anticoagulation is referred to variously as “adjusted-dose,” “therapeutic,” or “treatment.”

Unfractionated heparin thromboprophylaxis dosing

Table V. UFH Thromboprophylaxis Regimens. SC = Subcutaneous

Table V.
“Minidose” prophylactic UFH UFH 5000 units SC q 12 hrs
Prophylactic UFH UFH 5000-10000 units SC q 12 hrs
Intermediate-dose UFH UFH SC q12 hrs, adjust dose to target anti-Xa level of 0.1-0.3 U/mL
Adjusted-dose UFH UFH SC q12 hrs, adjust dose to target a mid-interval aPTT in therapeutic range

Low molecular weight heparin thromboprophylaxis dosing

Table VI. Prophylactic dose LMWH.

Table VI.
Weight Enoxaparin Dalteparin Tinzaparin
<50kg 20 mg daily 2500 units daily 3500 units daily
50-90kg 40 mg daily 5000 units daily 4500 units daily
91-130kg 60 mg daily* 7500 units daily* 7000 units daily*
131-170kg 80 mg daily* 10000 units daily* 9000 units daily*
>170kg 0.6 mg/kg/day* 75 units/kg/day* 75 units/kg/day*

*may be divided into twice daily dosing

Table VII. Intermediate dose LMWH.

Table VII.
Enoxaparin Dalteparin Tinzaparin
40 mg q 12 hours 5000 units q 12 hours 4500 units q 12 hours

Table VIII. Adjusted dose LMWH

Table VIII.
Enoxaparin Dalteparin Tinzaparin
1 mg/kg q 12 hours 100 units/kg q 12 hours 175 units once daily

All regimens use subcutaneous administration.

Peripartum management of thromboprophylaxis

We recommend changing patients on LMWH to UFH at 36-37 weeks gestation for two reasons. First, because of its shorter half-life, the use of UFH is more likely to afford an acceptable opportunity for neuraxial anesthesia if labor occurs. Second, UFH is reversible with protamine sulfate. Patients should be advised that if they suspect spontaneous labor, heparin should be discontinued.

For induction or scheduled cesarean, adjusted-dose and intermediate-dose LMWH should be discontinued 24 hours prior to the scheduled admission. Prophylactic heparin should be discontinued at least 12 hours prior. For high-risk patients, e.g. those with a recent VTE, reasonable options include reducing the heparin dose to 5000 units SC BID or utilizing a continuous infusion of heparin during labor (1 unit/kg/hr) with discontinuation when delivery is estimated to be 1-2 hours away.

In most cases, heparin should be restarted 6-8 hours following delivery or cesarean section. For high-risk patients, continuous infusion should be restarted after delivery when the risk of bleeding has decreased (usually 2-4 hours after delivery).

Reversing the anticoagulant effect of UFH or LMWH

Protamine sulfate binds to heparin to form a stable salt, and 1mg of protamine will neutralize the anticoagulant effect of 100 units of UFH. Because the half-life of intravenously administered heparin is only 60-90 minutes, relatively small doses of protamine can be expected to reverse UFH in patients on a continuous infusion (approximately 25-30mg for a patient on 1000-1250 units/hour). Larger doses are required for reversal soon after a bolus of UFH and a prolonged infusion may be required for the patient who has received subcutaneous UFH.

Protamine sulfate neutralizes the anti-IIa (activated thrombin) activity of LMWH, but has less effect on the anti-Xa activity. One suggested dosing guideline is if LMWH was administered subcutaneously within 8 hours, 1 mg of protamine should be given for every 1mg of enoxaparin (or for every 100 anti-Xa units). A second dose of half as much (0.5mg protamine/mg of enoxaparin) should be given if dangerous bleeding persists.

Anticoagulation and regional anesthesia

The American Society of Regional Anesthesia and Pain Medicine has made recommendations regarding anticoagulation and regional anesthesia.

Regarding UFH:

  • In patients receiving prophylaxis with subcutaneous UFH with dosing regimens of 5000 units twice daily, there is no contraindication to neuraxial techniques.

  • Subcutaneous heparin dosed twice daily with a total dose less than 10000 units of UFH per day is not a contraindication to neuraxial anesthesia. The safety of neuraxial blockade in patients receiving doses greater than 10000 units of UFH daily or more than twice-daily dosing of UFH has not been established.

  • Intravenous heparin can be initiated 1 hour following neuraxial anesthesia with catheter removal 2-4 hours after the last heparin dose.

Regarding LMWH:

  • The anti-Xa level is not predictive of the risk of bleeding and therefore is not a useful parameter for determining risk of neuraxial anesthesia.

  • For prophylactic LMWH, regional anesthesia can be placed 10-12 hours duration from the last dose of LMWH.

  • Regional anesthesia is contraindicated in patients less than 24 hours from their last dose of twice-daily LMWH.

  • The neuraxial catheter should be removed 2 hours prior to the first LMWH dose.

Management of Acute VTE at Term or in Labor

In hemodynamically stable patients, a temporary vena cava filter should be placed once the diagnosis has been confirmed. As soon as the patient goes into active labor or a cesarean section is considered, heparin should be stopped and reversed with protamine if necessary. A cesarean section should not be performed while the patient is in a fully anticoagulated state due to the risk of uncontrolled bleeding and maternal death. The care of the pregnant patient who has massive pulmonary embolism either at term or when suspicion of compromised fetal status would call for expeditious cesarean delivery is complex, and requires a coordinated treatment strategy by the obstetrician, intensivist, cardiothoracic surgeon, anesthesiologist and interventional radiologist.

The approach to the management of a massive pulmonary embolism should be individualized and adapted to changing circumstances; it could include cardiopulmonary bypass with surgical embolectomy followed by cesarean section, or percutaneous mechanical clot fragmentation and placement of an inferior vena cava filter. Although thrombolytic therapy is considered to be contraindicated, successful outcomes with the use of thrombolytic therapy during labor have been reported.

Postpartum thromboprophylaxis

Pregnant women at risk for VTE are candidates for postpartum thromboprophylaxis. As a general rule, the doses of postpartum UFH or LMWH should be equal or greater than the antepartum doses. UFH or LMWH may be re-started 4-6 hours after vaginal birth or 6-12 hours after cesarean if the patient is clinically stable. For women who will not be receiving long-term thromboprophylaxis with warfarin, continuing subcutaneous heparin during the period of postpartum thromboprophylaxis is acceptable.

For those patients who are bridging to warfarin thromboprophylaxis, warfarin may be initiated the evening of delivery at 5mg per day and adjusted thereafter according to daily INR determinations. UFH or LMWH should be continued until the INR is 2.0-3.0 for 2 consecutive days.

As a general rule, those women designated as being at intermediate or higher risk for pregnancy-related VTE should receive postpartum thromboprophylaxis a full 6 weeks postpartum.

Several recent publications have shown that the window of risk for postpartum VTE extends beyond the 6-week interval traditionally used for prophylaxis. A retrospective analysis of discharge data found the risk of a venous thrombotic event to be 16.5 per 100,000 deliveries in the first 6 weeks postpartum and 4.3 per 100,000 deliveries in weeks 7-12 postpartum, with the latter remaining significantly elevated compared to the control period. As yet, however, these findings have not led to alteration of national guidelines, and changes to practice are not yet warranted.

4. Complications

Complications of the disease process

Up to one-third of women will develop post thrombotic syndrome (PTS) after DVT, and 5-10% will develop severe manifestations. The features of PTS include chronic aching and pain, swelling, heaviness, edema and skin changes in the affected limb. These symptoms are usually aggravated by standing or walking and improve with rest and elevation of the affected leg. Risk factors for PTS include persistent leg symptoms one month after acute DVT, anatomically extensive DVT, recurrent ipsilateral DVT, obesity and older age.

Complications of treatment

Aside from the obvious risk of hemorrhage with UFH use, UFH is also associated with two other complications: heparin-induced thrombocytopenia (HIT) and osteoporosis. HIT occurs in up to 5% of pregnant patients treated with UFH, but is far less frequent in those treated with LMWH. It is an idiosyncratic immune reaction. The result is thrombin generation leading to a prothrombotic state in spite of platelet consumption and thrombocytopenia. HIT is to be distinguished from heparin-associated thrombocytopenia, a non-immune and far less dangerous condition seen in patients on UFH and LMWH.

The diagnosis of HIT is made by:

1. One or more of the following clinical criteria:

  • Otherwise unexplained thrombocytopenia: defined as a decrease in the platelet count by 50%, even if the nadir is greater than 150 X 109/L. The comparison should be to the highest platelet count during the 2-week period following initiation of heparin and immediately preceding the decline. The typical platelet count in HIT is 50-60 X 109 platelets/L, and counts less than 20 X 109 platelets/L are distinctly unusual in HIT. The typical time of onset for a heparin-naive individual is 5-10 days after starting heparin.

  • Venous or arterial thrombosis.

  • Skin lesions at heparin injection sites.

  • Acute systemic (anaphylactoid) reactions.


2. Presence of HIT antibodies detected by in vitro assays: The diagnosis of HIT is confirmed or refuted by appropriate laboratory tests. An immunoassay for anti-heparin/PF4 antibody levels is offered by most laboratories and has a rapid turnaround time and a high sensitivity for HIT. This test, however, has no better than modest specificity due to the fact that anti-heparin/PF4 antibodies may be found in individuals without HIT. Other more specific tests are available, the most popular of which are platelet serotonin release assays and function assays to detect heparin-dependent antibodies that bind to Fc receptors on platelets and induce activation.

In terms of monitoring, obstetric patients being treated with UFH for more than 3 days in the post-operative period, prophylactic or therapeutic, should be monitored for HIT. To do so, the platelet count should be determined on the 4th day after starting heparin and then every 2-3 days for 14 days. In addition, any patient being treated with UFH who has been exposed to UFH in the previous 100 days also should have monitoring for HIT by evaluating the platelet count before initiating therapy and 24 hours after initiation. Patients receiving LMWH or UFH not related to a surgical procedure and who have not been exposed to UFH within the previous 100 days do not require platelet count monitoring because their risk of HIT is less than 1%.

When the diagnosis of HIT is entertained in a pregnant patient, e.g. because of thrombocytopenia, all heparin products should be discontinued, including heparin flushes, and an alternative anticoagulant should be started. Though direct thrombin inhibitors are usually the agents of first choice in non-pregnant patients, danaparoid or fondaparinux sodium is used in pregnancy since neither crosses the placenta to a clinically-important degree. Danaparoid is not available in the U.S.

As with most drug-induced thrombocytopenias, patients with a prior history of HIT should not be re-exposed to heparin. Note that anti-heparin/PF4 antibodies appear not to persist in the circulation and are frequently not detectable by 3 months after the diagnosis of HIT.

5. What is the evidence for specific management and treatment recommendations

These four references are the most recently published guidelines from the major medical organizations that are charged with the care of pregnant women experiencing VTE or at risk for the condition.

“Bates, Shannon et al. VTE, Thrombophilia, Antithrombotic Therapy, and Pregnancy: Antithrombotic Therapy and Prevention of Thrombosis, 9 ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines”. Chest. vol. 141. 2012. pp. e691S-e736S. (These are the most comprehensive guidelines regarding VTE in pregnancy, with the newest revision published in 2012.)

“American Congress of Obstetricians and Gynecologists (ACOG). Thromboembolism in Pregnancy”. 2011 September. (The recommendations from the governing body of obstetrics and gynecology in the United States as of 2011.)

(These are the guidelines from the group in the United Kingdom that governs obstetrics and gynecology as of 2009.)

Horlocker, TT. “Regional Anesthesia in the Patient Receiving Antithrombotic or Thrombolytic Therapy: American Society of Regional Anesthesia and Pain Medicine Evidence-Based Guidelines (Third Edition)”. Reg Anesth Pain Med. vol. 35. 2010. pp. 64-101. (This is the guideline published by the governing body for anesthesiologists in the United States as of 2010.)