Deep vein thrombosis in pregnancy: Epidemiology, pathogenesis, and diagnosis

INTRODUCTION — Pregnancy and the puerperium (postpartum period) are well-established risk factors for venous thromboembolism (VTE), which occurs with an incidence of 1 in 1600 [1-7]. The overlap with symptoms of pregnancy may impair clinical suspicion, making diagnosis of VTE more challenging.

VTE can manifest during pregnancy as an isolated lower extremity deep vein thrombosis (DVT), or clot can break off from the lower extremities and travel to the lung to present as a pulmonary embolus (PE) [8-10]. PE is the seventh leading cause of maternal mortality, responsible for 9 percent of maternal deaths [11-13]. Thus, the detection of DVT during pregnancy is critical to preventing deaths from PE.

The epidemiology, pathogenesis, and diagnosis of DVT during pregnancy and the puerperium will be reviewed here. The epidemiology, pathogenesis, and diagnosis of PE, as well as the prevention and treatment of DVT and PE during pregnancy are discussed separately. (See "Pulmonary embolism in pregnancy: Clinical presentation and diagnosis" and "Venous thromboembolism in pregnancy: Prevention" and "Venous thromboembolism in pregnancy and postpartum: Treatment" and "Use of anticoagulants during pregnancy and postpartum".)

EPIDEMIOLOGY — The overall incidence of venous thromboembolism (VTE) in pregnancy is low. In the United States, VTE is diagnosed during 1 in 500 to 2000 pregnancies (absolute incidence; 0.025 to 0.1 percent) [1-6,14]. In a retrospective case-control study of 395,335 pregnant patients at 24 weeks of gestation, the incidence of VTE was 85 per 100,000 pregnancies [15]. A population-based inception cohort study over a 30 year period detected an overall incidence of VTE of 200 per 100,000 woman-years [5]. DVT was three times more common than PE [5].

Similar rates are observed in Europe [2,5]. In one retrospective study of over 72,000 deliveries, the incidence of DVT was 0.71 per 1000 deliveries (95% CI 0.5-0.9) with 0.5 (95% CI 0.34-0.66) and 0.21 (95% CI 0.11-0.31) occurring antenatally and postnatally, respectively [16]. The incidence of PE was 0.15 per 1000 deliveries (95% CI 0.06-0.24) with 0.07 (95% CI 0.01-0.13) and 0.08 (95% CI 0.02-0.14) occurring antenatally and postnatally, respectively [16].

PE is the seventh leading cause of maternal mortality and accounts for 9 percent of maternal deaths [17-19].

Pregnancy-related mortality has the largest racial disparity among the maternal and child health indicators [11,20]. Black females have a three to four times higher pregnancy-related mortality rate than White females. Although the reasons for this cannot be directly attributed to increased frequency of VTE, compared with White females, deaths from VTE are higher in Black females [11]. Additionally, compared with White females, Black females had a higher risk of VTE (odds ratio 1.5) [20].

Whether rates are falling or rising is unclear since data are conflicting. Two studies suggest that the incidence of VTE appears to be decreasing over the last three decades, largely due to a decrease in the incidence of VTE in the postpartum period [5,21]. In contrast, another study reports a 14 percent increase in VTE-associated pregnancy hospitalizations between 1994 and 2009, with a concomitant increase in comorbid conditions such as obesity and hypertension among those admitted for VTE [22]. In agreement, a cross sectional analysis of the National Inpatient Sample reported a rising rate of VTE among vaginal deliveries (average annual percent change 2.5 percent between 2000 and 2018) but not among cesarean deliveries (average annual percent change 0.3 percent) [23].

RISK FACTORS — Pregnancy, in itself, is a risk factor for the development of venous thromboembolism (VTE), with a reported incidence that is 4 to 50 times higher compared with nonpregnant females (table 1) [1-7]. Increased risk for VTE is highest in the postpartum period, with a higher than usual development of clot in the left lower extremity and pelvis. In addition, risk is augmented further in females with inherited thrombophilias. These risk factors are discussed in the following sections.

A risk model for VTE within the first six weeks postpartum has been developed and externally validated, based upon two large European cohorts [14]. Among the many factors that contribute to the model, those associated with the highest risk were emergency cesarean section, stillbirth, varicose veins, pre-eclampsia/eclampsia, postpartum infection, and comorbidities. Further validation is needed to confirm these results.

Timing during pregnancy — Compared with the nonpregnant population, the risk of VTE is higher in all stages of pregnancy. However, it is greatest in the postpartum period. The identified risk factors associated with the antepartum and postpartum periods are listed below.

●Antepartum – Most studies report equal distribution of VTE across the trimesters of pregnancy [1,2,15,21,24-27]. However, two large conflicting retrospective studies reported a first trimester predominance (50 percent before 15 weeks) and third trimester predominance (60 percent), respectively [28,29]. Factors that increase the risk of VTE antepartum are less well described but include the following:

•Multiple births [15]

•Varicose veins [30]

•Inflammatory bowel disease [30]

•Urinary tract infection [30]

•Diabetes [30]

•Hospitalization for non-delivery reasons (particularly those >3 days) [31]

•Body mass index (BMI) ≥30 kg/m² [31]

•Increased maternal age ≥35 years [31]

●Postpartum – Compared with the antepartum period, VTE is two to five times more common postpartum [5,15,21,24,25,27,29,32-34]. The risk is highest in the first six weeks postpartum and declines to rates that approximate that of the general population by about 13 to 18 weeks [32].

Commonly cited factors that increase the risk of VTE postpartum include the following:

•Cesarean section (CS), especially emergent CS [21,30,32-35]

•Medical comorbidities (eg, varicose veins, cardiac disease, inflammatory bowel disease) [15,30]

•BMI ≥25 kg/m² [15,30,34]

•Young gestational age (preterm delivery <36 weeks) [15,27,29,30,34]

•Obstetric hemorrhage [30,33,34]

•Stillbirth [30]

•Increased maternal age ≥35 years [15,32]

•Hypertension [21]

•Smoking [32,36]

•Eclampsia or preeclampsia [32-34]

•Postpartum infection [33,34]

•Transfusion [23]

•Thrombophilias [23]

Thrombotic risk is highest in the first six weeks postpartum, and although the risk persists until 12 weeks, the absolute risk beyond six weeks appears to be low. Data best illustrating this persistent (though attenuated) postpartum thrombotic risk are derived from a five-year retrospective crossover-cohort analysis of an insurance claims database including almost 1.7 million patients that examined rates of thrombosis during the first year following delivery [32]. VTE accounted for the majority (68 percent) of reported thrombotic events, followed by stroke (28 percent) and myocardial infarction (4 percent). Compared with rates at one year, the risk of thrombosis was highest in the first six weeks postpartum (25 versus 2 per 100,000 deliveries; odds ratio [OR] 11, 95% CI 8-15) and declined to lower but still elevated rates at 7 to 12 weeks (6 versus 3 per 100,000 deliveries; OR 2, 95% CI 1.5-3.1). Rates were close to that of the general population by 13 to 18 weeks (3 versus 2 per 100,000 deliveries; OR 1.4, 95% CI 0.9-2.1). As compared with patients without postpartum thrombosis, those with thrombotic events were more likely to be older (≥35 years) and have risk factors for thrombosis (eg, eclampsia, hypercoagulable state, smoking, cesarean section).

Inherited thrombophilias — The risk of VTE is further magnified in pregnant patients who have inherited thrombophilias (eg, factor V Leiden; antithrombin III, protein S, or protein C deficiency; or antiphospholipid syndrome) [4,16,37-42]. The inherited thrombophilias during pregnancy are discussed in detail separately. (See "Inherited thrombophilias in pregnancy".)

In brief, the variable range in risk of VTE in pregnant patients with the most common inherited thrombophilias is illustrated by the following studies:

●Compared with the general population, the thrombotic risk is three times higher for pregnant patients with factor V Leiden [16]. Further augmenting that risk in pregnant patients with factor V Leiden deficiency and G20210A prothrombin-gene mutation is a history of prior VTE in the patient or an affected first-degree relative (up to 50-fold) [42]. (See "Factor V Leiden and activated protein C resistance" and "Prothrombin G20210A", section on 'Obstetric issues'.)

●When compared with pregnant patients without a known thrombophilia, pregnant patients with an inheritable deficiency of antithrombin III, protein S, or protein C had an eight-fold increased risk of venous thrombosis in the antepartum and postpartum periods combined [37]. (See "Antithrombin deficiency" and "Protein S deficiency" and "Protein C deficiency".)

●Among patients with known antiphospholipid syndrome, one prospective study described a 5 percent risk of thrombosis during pregnancy [43]. (See "Clinical manifestations of antiphospholipid syndrome".)

PATHOGENESIS — Pregnancy and the postpartum period are marked by the presence of all three components of Virchow's triad: venous stasis, endothelial injury, and a hypercoagulable state [2]. All features likely contribute to the increased risk of venous thromboembolism (VTE) in pregnancy.

●Stasis – Venous stasis of the lower extremities occurs during pregnancy because of two factors: pregnancy-associated changes in venous capacitance and compression of large veins by the gravid uterus. The lower extremity veins of pregnant patients appear to be subject to increased stasis even before the uterus has enlarged substantially. Although blood volume and total venous return are supranormal in pregnancy, the linear flow velocity in the lower extremity veins is decreased due to hormonally induced dilation of capacitance veins, leading to venous pooling and valvular incompetence [44]. (See "Maternal adaptations to pregnancy: Cardiovascular and hemodynamic changes".)

These early changes are amplified by inferior vena caval and iliac vein compression by the gravid uterus [44,45]. One study that assessed 24 pregnant females with monthly compressive ultrasound examinations found progressive dilation of the deep veins of the legs during gestation [46]. This corresponded to a decreased flow velocity in the left common femoral vein and inferior vena cava that was most severe in the supine position. Assuming the left lateral decubitus position significantly increased the velocity in both lower extremities. In addition, as noted above, compression of the left iliac vein by the right iliac artery is thought to contribute to the predilection for left-sided DVT during pregnancy [1,8].

●Endothelial injury – Delivery is associated with vascular injury and changes at the uteroplacental surface, which probably contribute to the increased risk of VTE in the immediate postpartum period. Forceps, vacuum extraction, or surgical delivery can exaggerate vascular intimal injury and amplify this phenomenon [1].

●Hypercoagulability – Pregnancy is a hypercoagulable state associated with progressive increases in several coagulation factors, including factors I, II, VII, VIII, IX, and X, along with a decrease in protein S [1,16,47]. A progressive increase in resistance to activated protein C is normally observed in the second and third trimesters [48], and high resistance to activated protein C was shown in one study to be associated with an increased risk for pregnancy-related venous thrombosis [49]. Activity of the fibrinolytic inhibitors PAI-1 and PAI-2 is increased during pregnancy, although total fibrinolytic activity may not be impaired [50,51].

CLINICAL PRESENTATION

Signs and symptoms — Other than the higher propensity to develop left-sided DVT and iliac vein thrombosis, the clinical presentation of DVT during pregnancy is identical to that in nonpregnant females. Signs and symptoms suggestive of proximal vein thrombosis are diffuse pain and swelling that may or may not be associated with erythema, warmth and tenderness of the lower extremity. Symptoms of iliac vein thrombosis include swelling of the entire leg with or without flank, lower abdomen, buttock, or back pain [52,53]. The clinical signs and symptoms of DVT in nonpregnant patients are discussed separately. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity" and 'Anatomic location of DVT' below.)

The clinical features of deep vein thrombosis (DVT) in pregnancy and the puerperium overlap with many of the features of normal pregnancy. Thus, distinguishing the features associated with the hemodynamic changes of pregnancy from clinically important DVT can be difficult. (See "Clinical manifestations and diagnosis of early pregnancy", section on 'Clinical manifestations of early pregnancy' and "Maternal adaptations to pregnancy: Cardiovascular and hemodynamic changes".)

Anatomic location of DVT — The majority of lower extremity DVTs during pregnancy are left-sided. In addition, the incidence of pelvic vein thrombosis is significantly higher during pregnancy and the puerperium, although DVT remains most often found in the proximal veins (eg, femoral vein). No studies describe an increased incidence of upper extremity DVT during pregnancy or the puerperium.

●Left lower extremity DVT – DVT is predominantly left-sided in pregnancy (70 to 90 percent). In one study of 60 pregnant patients with a first episode of VTE, there were 58 isolated left lower extremity DVTs, two bilateral DVTs, and no isolated right lower extremity DVTs [24]. In another retrospective study of 124 pregnant patients with a diagnosis of DVT, involvement of the left leg was reported in 88 percent of the 96 patients for whom the affected side was known [54]. This striking distribution has been attributed to increased venous stasis in the left leg related to compression of the left iliac vein by the right iliac artery, coupled with compression of the inferior vena cava by the gravid uterus [24,55,56]. (See "Overview of the causes of venous thrombosis", section on 'May-Thurner syndrome'.)

●Pelvic vein DVT – Pelvic vein DVT is more commonly diagnosed in pregnancy than in the general population. However, its true incidence during pregnancy is unknown. This may be due to the poor sensitivity of compressive proximal vein ultrasound for the diagnosis of thrombosis in the pelvic veins [57]. In an analysis of the DVT-Free Registry enrolling 5451 consecutive patients with ultrasound-confirmed DVT, the rates of isolated pelvic vein DVT were higher in antepartum and postpartum females than in nonpregnant females (12 and 11 percent versus 1 percent, respectively) [7]. Another retrospective study of 124 pregnant patients showed 64 percent of proximal DVTs were restricted to the iliac and/or femoral vein [54].

LABORATORY TESTING — Routine laboratory tests (eg, complete blood count, chemistries, liver function tests, coagulation studies) are not useful diagnostically, but may provide clues as to the underlying cause and may influence treatment decisions if DVT is confirmed. D-dimer is useful if negative, but not useful if positive. Arterial blood gases are not indicated routinely in the diagnosis of DVT. Laboratory testing for suspected DVT in the nonpregnant patient is discussed separately. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)

D-dimer — Assays (enzyme-linked, turbidimetric, hemagglutination) for serum D-dimer, a breakdown product of cross-linked fibrin, have been extensively studied for use in the diagnosis of DVT. In nonpregnant patients, the negative predictive value of D-dimer in ruling out DVT is high, particularly when combined with clinical probability models or with a negative compressive ultrasound. However, serum D-dimer increases during the course of pregnancy with a slow decline postpartum. Normal reference ranges for the antepartum and postpartum periods have not been established so elevated levels during pregnancy and postpartum are not useful in diagnosis of DVT. On the other hand, the negative predictive value of D-dimer in ruling out DVT is high, particularly when combined with clinical probability models or with a negative compressive ultrasound. The assays used and the diagnostic utility of D-dimer in nonpregnant patients are discussed in detail separately. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'D-dimer'.)

The following studies have confirmed the limited value of D-dimer during pregnancy:

●In a study of 50 pregnant patients without venous thrombosis, D-dimer levels increased with each trimester such that 22 percent of patients in the second trimester and none in the third trimester had a normal D-dimer concentration (ie, <500 ng/mL) [58].

●In another study of 149 consecutive pregnant patients presenting with suspected DVT, in whom this diagnosis was ruled out by compression ultrasonography, D-dimer was negative in 100, 76, and 49 percent of those in the first, second, and third trimester, respectively [59].

●Two small studies confirmed that pregnant patients with DVT had higher D-dimer levels than pregnant patients without DVT [60,61]. Using higher cut-off values resulted in maintenance of high sensitivity (80 to 100 percent) and improved specificity (62 to 79 percent) of D-dimer for DVT diagnosis ("moderate to high sensitivity" D-dimer) [60]. Validation in prospective studies is needed before higher cut-off values of D-dimer are used as a routine pretest probability tool during pregnancy.

IMAGING — The diagnosis of DVT during pregnancy is most often made by demonstrating poor compressibility of the proximal veins on compressive ultrasonography (CUS); contrast and magnetic resonance venography are rarely used.

Much of the data to support imaging studies for the diagnosis of DVT in pregnancy is extrapolated from large studies in the nonpregnant population with smaller studies suggesting similar efficacy in pregnancy. The diagnosis of DVT during pregnancy is most often made by demonstrating poor compressibility of the proximal veins on compressive ultrasonography (CUS). The diagnosis is rarely made by the demonstration of a filling defect on contrast or magnetic resonance venography.

Proximal vein CUS is a highly sensitive and specific tool for the diagnosis of DVT in both pregnant and nonpregnant patients. However, it is less sensitive for pelvic vein thrombosis (more common in pregnancy) and for calf vein thrombosis (less common in pregnancy) [57]. In cases where CUS is negative, poor Doppler flow in the iliac vein has reasonable accuracy for the diagnosis of suspected pelvic vein DVT; serial compressive ultrasound is a sensitive strategy that can be used to follow suspected calf vein DVT in the rare circumstance that it propagates proximally. The approach to diagnosing suspected DVT in pregnancy is discussed in detail separately. (See 'Diagnostic algorithm' below.)

Compression ultrasonography — In pregnant patients, poor compressibility of a thigh vein with the ultrasound probe is highly sensitive (95 percent) and specific (>95 percent) for the diagnosis of symptomatic proximal vein thrombosis [62]. The addition of Doppler analysis of flow variation with respiration with the patient in the left lateral decubitus position assists in diagnosing isolated iliac vein thrombosis during pregnancy [46]. When positive, the diagnosis of DVT by CUS in a pregnant patient should prompt immediate anticoagulation. (See "Venous thromboembolism in pregnancy and postpartum: Treatment" and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Diagnostic ultrasonography suspected first DVT'.)

CUS is less sensitive for pelvic vein thrombosis and for calf vein thrombosis [57]. In cases where CUS is negative, pelvic vein thrombosis may be suspected when the visualized vein is compressible but there is absence of normal changes in flow during respiration or with the Valsalva maneuver. Serial CUS can be used to detect suspected calf vein thrombosis that propagates proximally as the pregnancy progresses.

Serial compression ultrasonography — Proximal vein CUS is less sensitive for the diagnosis of calf vein thrombosis than for proximal vein thrombosis [57]. In nonpregnant patients, calf vein thromboses can propagate proximally in approximately 20 percent of cases [63]. Serial CUS (performed on days 3 and 7) has been well validated in nonpregnant patients with suspected DVT for the detection of calf DVT in the setting of an initially negative ultrasound [63-65]. Using this modality to follow nonpregnant patients with suspected DVT with initial negative CUS, only 2 percent are subsequently diagnosed with DVT [64,65]. A number of small prospective studies have replicated the same findings in pregnant patients.

●A multicenter, prospective study of 149 pregnant females with suspected DVT and negative initial CUS, in whom anticoagulation was withheld for three months, showed that serial proximal CUS (performed on day 3 and 7 post initial testing), had a false negative rate of 0.7 percent for the exclusion of DVT [59].

●A single center, prospective study of 221 patients with suspected DVT and initial negative ultrasound showed that serial proximal CUS (days 3 and 7) excluded DVT with a sensitivity and negative predictive value of 94.1 and 99.5 percent, respectively, at three months [66].

●Two additional studies (one retrospective, one observational) in a similar population of pregnant patients imaged with serial whole leg CUS had very few DVTs detected in the follow-up period after initial negative testing [67,68].

These studies suggest that, similar to what is observed in the general population, serial CUS is valuable as a sensitive imaging modality to diagnose the rare case of calf DVT that propagates proximally during pregnancy. (See 'Diagnostic algorithm' below.)

Magnetic resonance venography — Magnetic resonance venography is a modality that can detect both thigh and pelvic vein DVT with a sensitivity that approaches 100 percent in the nonpregnant population [69]. Data are limited in pregnancy. However, small case series of pregnant patients suggested that this modality was useful for the diagnosis of pelvic and femoral vein thrombosis in situations where other noninvasive examinations were equivocal [70,71]. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Alternative imaging'.)

Ascending contrast venography — Visualization of a filling defect by ascending contrast venography is considered the gold standard for the diagnosis of lower extremity DVT in the nonpregnant population [72,73]. Studies that measure the accuracy, sensitivity, and specificity of venography in pregnancy are lacking. Venography is rarely performed in pregnancy due to concerns regarding exposure of the fetus to ionizing radiation, technical difficulties of femoral vein cannulation, and decreased sensitivity for isolated iliofemoral thrombosis due to abdominal-pelvic shielding [1,53,62]. Additionally, the alternative imaging test, CUS, approaches venography in diagnostic sensitivity and specificity without these risks, rendering contrast venography less useful than in the past for the diagnosis of DVT [1,53,62]. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Alternative imaging'.)

DIAGNOSTIC ALGORITHM — The successful diagnosis of suspected deep vein thrombosis (DVT) in pregnancy and the puerperium requires that clinicians have a high index of clinical suspicion and low threshold for use of objective confirmatory testing (algorithm 1). Our approach to the diagnosis of DVT during pregnancy is consistent with evidence-based practice guidelines published by the American College of Chest Physicians (ACCP) in 2012, the American College of Obstetricians and Gynecologists in 2018, and the American Society of Hematology in 2018 [74-76]. These guidelines are useful resources for the clinician to make decisions regarding testing and implementation of anticoagulation based upon the individual assessment of a pregnant patient with suspected DVT.

Pretest probability — Common predictive scoring systems (eg, Wells score), the LEFt clinical prediction rule, and D-dimer levels have all been described as potential clinical probability assessment tools for the diagnosis of suspected DVT. However, these tools have not been validated in large prospective trials and, in general, they are less useful in pregnant patients than in the general population.

Wells score — The Wells and modified Wells scoring systems are the most commonly used scoring systems for nonpregnant patients with suspected DVT. However, they are not validated for use in pregnancy and should be interpreted with caution in this population. Of importance, some of the listed features (eg, active cancer, recent surgery) are not likely to be present in young healthy pregnant females while other features such as pitting edema and lower extremity tenderness are common symptoms of pregnancy in the absence of DVT. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)

LEFt clinical prediction rule — The LEFt clinical prediction rule was developed on the rationale that DVT in pregnancy is predominantly left-sided [77,78]. A cross-sectional study evaluated clinical assessment in predicting the presence of a suspected first-time DVT in 194 pregnant females [77]. Three objective variables were highly predictive of DVT:

●Symptoms in the left leg (L for left)

●Calf circumference difference ≥2 cm (E for edema)

●First trimester presentation (Ft for first trimester presentation)

At least one of these variables was present in all 17 patients (8.8 percent) with DVT. Among patients presenting with none, one, or two to three of these variables, DVT was diagnosed in zero, 16, and 58 percent of patients, respectively. The LEFt clinical prediction rule was further validated in a separate cohort of 157 pregnant patients with suspected DVT [78]. In this population, a negative LEFt rule accurately identified all patients (29 percent) who did not have a DVT. The rule should not be used as a stand-alone test for excluding DVT and needs further prospective validation in a larger population before it can be routinely applied in clinical practice.

D-dimer — Due to the natural rise in levels with each trimester, D-dimer using a cut-off value of >500 ng/mL (by enzyme-linked immunosorbent assay [ELISA] or RBC agglutination [SimpliRED]) has limited value as a pretest probability tool for the diagnosis of DVT in pregnancy [58-61]. However, a negative test (<500 ng/mL), when present, can significantly lower the clinical suspicion for DVT and aid the physician in the decision to avoid further testing (algorithm 1) [59,74]. Moderate to high sensitivity D-dimer testing (ie, a higher cut-off value) is not validated in pregnancy and cannot be routinely used as a pretest probability tool for the diagnosis of suspected DVT in pregnancy. Studies describing the use of D-dimer in pregnancy are discussed in detail separately. (See 'D-dimer' above.)

Compressive ultrasound and subsequent testing — All pregnant patients suspected of having a DVT should undergo initial evaluation with compressive ultrasound (CUS) (algorithm 1). Proximal CUS is preferred as the first-line test for the diagnosis of suspected DVT in pregnancy over venography or magnetic resonance imaging [74]. In advanced pregnancy, CUS should be performed with the patient in the left lateral decubitus position.

The need for additional testing (eg, Doppler ultrasound of the iliac vein, magnetic resonance, or contrast venography) is dependent upon the CUS results and clinical suspicion (algorithm 1).

Positive CUS — A positive CUS is diagnostic of DVT. Anticoagulation should be initiated. The benefits of anticoagulation for DVT outweigh the bleeding risks during pregnancy or the puerperium. (See 'Imaging' above and "Venous thromboembolism in pregnancy and postpartum: Treatment".)

Negative CUS — A negative CUS does not rule out DVT in the pregnant patient. Further testing depends upon the degree of clinical suspicion. When the initial CUS is negative, but clinical suspicion for DVT remains, there are two reasonable options: further testing concurrent with empiric anticoagulation, or further testing with anticoagulation reserved for confirmed cases on follow-up testing. There is a paucity of data to guide the clinician with this decision. The clinician is reliant upon judgment and should weigh the risks of untreated venous thromboembolism (VTE) and bleeding in a pregnant patient in the context of clinical suspicion (algorithm 1). In general, the following apply:

●When the clinical suspicion is high (eg, suspected iliac vein thrombosis with swelling of the entire leg and buttock), we prefer further evaluation with Doppler ultrasound directed at the iliac vein, followed by magnetic resonance or contrast venography, as needed. For these patients, we suggest empiric anticoagulation. This preference places a high value on diagnostic certainty and risk of maternal death for an untreated thrombosis in pregnancy, and places less value on risk of radiation or contrast exposure. (See 'Imaging' above.)

●For patients in whom the clinical suspicion is not high, we prefer serial CUS (performed on day 3 and 7) without empiric anticoagulation but with clinical follow-up throughout pregnancy. Data from nonpregnant [63,65,74] and pregnant patients [59,66-68,74] suggest that the safety of withholding anticoagulation in this setting is assured. This preference places a high value on avoiding unnecessary anticoagulation when the risk of DVT is low. If serial CUS is not feasible, we suggest measuring D-dimer levels as an alternative. When D-dimer is <500 ng/mL (by ELISA or SimpliRED), DVT is unlikely [59]. (See 'Serial compression ultrasonography' above.)

There is no measurable radiation exposure associated with ultrasound or magnetic resonance imaging. However there are risks to the fetus and mother associated with radiation exposure for contrast venography, and with iodinated and gadolinium contrast for venographic studies. The risks of radiation exposure and contrast administration during pregnancy are discussed in detail separately. (See "Pulmonary embolism in pregnancy: Clinical presentation and diagnosis", section on 'Radiation and contrast risk'.)

DIAGNOSIS — The diagnosis of deep vein thrombosis (DVT) in pregnancy is made by demonstrating a lack of compressibility of the proximal veins on compressive ultrasound (femoral vein thrombosis) or poor flow on Doppler imaging of the femoral-iliac vein (iliac vein thrombosis). The diagnosis is rarely made by the demonstration of a filling defect on contrast or magnetic resonance venography. D-dimer levels and clinical exam cannot be used alone to diagnose DVT. A diagnostic approach for suspected DVT in pregnancy is discussed in detail above (algorithm 1). (See 'Diagnostic algorithm' above.)

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of DVT in pregnancy is similar to that in non-pregnant patients. It includes other entities that present with erythema, warmth, swelling and tenderness of the lower extremity and/or flank, lower abdomen, buttock, or back. However, the clinical signs and symptoms of DVT in pregnancy are masked by many of the physiologic changes of normal pregnancy (eg, lower extremity swelling and cramping).

The clinical suspicion for DVT should be high in the setting of pregnancy. Although not always present, features that are highly suggestive of the diagnosis include unilateral signs and symptoms and the classic symptoms of iliac vein thrombosis (flank/pelvic/buttock pain and swelling). Such symptoms should prompt immediate investigation for DVT with compressive or Doppler ultrasound. It is important to note that DVT can coexist with other conditions. Thus, although the finding of an alternative diagnosis (eg, cellulitis) will lower the clinical suspicion for DVT, it does not always obviate the need for diagnostic imaging.

The differential diagnosis of DVT in nonpregnant patients and the physiologic changes of pregnancy that can mimic DVT are discussed separately. (See "Maternal adaptations to pregnancy: Cardiovascular and hemodynamic changes" and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Differential diagnosis'.)

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: Superficial vein thrombosis, deep vein thrombosis, and pulmonary embolism" and "Society guideline links: Anticoagulation in pregnancy".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Deep vein thrombosis (blood clot in the leg) (The Basics)")

●Beyond the Basics topics (see "Patient education: Deep vein thrombosis (DVT) (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Epidemiology – Venous thromboembolism (VTE) can manifest during pregnancy as an isolated lower extremity deep venous thrombosis (DVT) or pulmonary embolus (PE). PE is the seventh leading cause of maternal mortality, responsible for 9 percent of maternal deaths. (See 'Epidemiology' above.)

●Risk factors – Pregnancy is a risk factor for the development of VTE with a reported incidence that is 4 to 50 times higher in pregnant patients compared with nonpregnant females (table 1). Increased risk for lower extremity DVT is highest in the first six weeks of the postpartum period, with a higher than usual incidence of left-sided DVT and pelvic vein clot. (See 'Risk factors' above.)

●Clinical presentation – Clinical presentation is as follows:

•Signs and symptoms suggestive of proximal vein thrombosis are diffuse pain and swelling that may or may not be associated with erythema, warmth and tenderness of the lower extremity. Symptoms of iliac vein thrombosis include swelling of the entire leg with or without flank, lower abdomen, buttock, or back pain. (See 'Signs and symptoms' above.)

•The clinical features of DVT in pregnancy overlap with many of the features of normal pregnancy. A high index of clinical suspicion and low threshold for the use of objective confirmatory testing are required to accurately diagnose DVT during pregnancy. (See 'Clinical presentation' above and 'Diagnostic algorithm' above.)

●Laboratory testing – D-dimer and clinical prediction rules are of limited value as pretest probability tools for the diagnosis of DVT during pregnancy and the puerperium. Moderate or highly sensitive D-dimer (with higher than usual cut-off values) have not been adequately validated for routine use in pregnancy. However, a negative D-dimer is associated with a high negative predictive value in any trimester. (See 'Laboratory testing' above and 'Pretest probability' above.)

●Diagnostic evaluation – The diagnosis of DVT in pregnancy is made by demonstrating a lack of compressibility of the proximal veins on compressive ultrasound (femoral vein thrombosis) or poor flow on Doppler imaging of the femoral-iliac vein (iliac vein thrombosis). The diagnosis is rarely made by the demonstration of a filling defect on contrast or magnetic resonance venography. D-dimer levels and clinical exam cannot be used alone to diagnose DVT. (See 'Diagnosis' above.)

•Our approach to the evaluation of a patient with suspected DVT in pregnancy depends at least partially upon the degree of clinical suspicion (algorithm 1).

•For all pregnant patients suspected of having a lower extremity DVT, we recommend proximal vein compression ultrasound (CUS) with the patient in the left lateral decubitus position as the first-line diagnostic test, over venography or magnetic resonance imaging. (See 'Compressive ultrasound and subsequent testing' above.)

-For pregnant patients with a negative CUS in whom the clinical suspicion is not high, we suggest serial CUS performed on day 3 and day 7 with additional clinical observation for the duration of pregnancy. (See 'Compressive ultrasound and subsequent testing' above.)

-For pregnant patients with a negative CUS in whom the clinical suspicion is high, particularly those with signs and symptoms of iliac vein thrombosis (pain and swelling of the entire leg and buttock), we suggest Doppler ultrasound directed at the iliac vein rather than magnetic resonance or contrast venography. (See 'Compressive ultrasound and subsequent testing' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges David R Schwartz, MD, who contributed to earlier versions of this topic review.