In pregnancy, DVT involves more frequently the iliofemoral system (72% vs 9%) and the left side (85% vs 55%). On the left side, the iliac artery crosses over the vein, resulting in additional stasis. Upper extremity VTE is exceedingly rare and should raise the suspicion of intravenous drug use (particularly cocaine), a severe hypercoagulable state or, more commonly, a needle or a catheter left in the vein too long.

Symptomatic VTE in pregnant women without previous episodes and no obvious risk factors is 0.5–1.8 per 1000. Older studies reported higher frequencies because objective documentation of thrombosis was not usually required. More recent publications quote 1 in 1000 deliveries. Subclinical, asymptomatic episodes of thrombosis, particularly in the calf and during the puerperium, are probably much more common and incidences as high as 3% are quoted, but no prospective trials involving pregnant or postpartum women have been done. Moreover, these quotes are extrapolated from studies in nonpregnant patients undergoing a variety of elective surgical procedures with preoperative injection of fibrinogenlabeled iodine 131, which for obvious reasons cannot be used during pregnancy.

In our own experience, during a 19-year period there were 165 VTE complicated pregnancies and 268,036 deliveries (1 per 1627 births – 0.06%). There were 127 cases of DVT and 38 cases of PE.

The recurrence rate during pregnancy in women with a previous VTE episode depends on the specific risks associated with the previous episode. The risk was higher when it was associated with a hypercoagulable state (thrombophilia) or if it was labeled as “idiopathic” and much lower if the risk factors were transient, such as trauma or surgery.

The risk of recurrence in thrombophilias, either hereditary or acquired, depends on the specific condition. Heterozygous women who are carriers of the G1691A factor V Leiden mutation have a 3–9-fold higher risk but 49–80-fold if they are homozygous. The G2021A mutation in the prothrombin gene carries a 2–9-fold higher risk for the heterozygotes and 16-fold for the homozygous carriers. If the patient happens to be heterozygote for both the factor V Leiden and prothrombin G2021A gene mutation (“compound mutation”), the risk increases by 150-fold. Antithrombin III (AT-III) deficiency increases the risk 25–50-fold, making it the single most thrombogenic condition. Protein C deficiency causes a 3–15-fold higher risk and protein S deficiency twofold. Hyperhomocysteinemia carries a 2.5–4-fold higher risk and antiphospholipid antibodies increase the risk by 5–6-fold. Women who are relatives of persons with VTE but who themselves have only laboratory evidence of risk factors, and not previous clinical VTE episodes, were found to have a 4% risk of VTE during pregnancy.

At present, 50% or more of pregnancy-related VTE cases occur in women with a hereditary or acquired thrombophilia. Those without discernible risk factors are said to have an “idiopathic episode” but they are at a high risk of recurring VTE during subsequent pregnancies. It is possible that the VTE currently labeled as “idiopathic” are actually cases of still unknown thrombophilias.

The risk of VTE was thought to increase with progression of pregnancy but recent studies indicate that the risk is evenly divided throughout all trimesters and that it is already well established in early pregnancy, by the end of the first trimester. The immediate postpartum remains the highest risk period for PE, with ≥ 80% occurring after operative deliveries. Sixty-one percent of all PE in our series occurred post partum and 82.6% of all postpartum PE occurred after cesarean sections.

The overall risk of VTE at present is still higher after delivery than antepartum. Nevertheless, the risk of postpartum VTE has decreased to a greater extent than antepartum in comparison to older studies and most likely because of changes in obstetric practices such as shorter hospital stays, earlier postpartum ambulation and bedrest not being recommended as frequently. And when bedrest is recommended, measures are generally applied to reduce risks by using elastic stockings, intermittent pneumatic leg compressors or prophylactic anticoagulation in the high-risk groups. Furthermore, estrogens are no longer used to suppress lactation.

Rudolf Virchow is credited with establishing the basic risk factors for VTE in his now classic 1845 lecture, later published in 1856, and known as the “Virchow’s triad”: venous stasis, “alterations of the blood,” and endothelial vascular injury. These risk factors are present during gestation and are believed to be the reason for the so-called “hypercoagulable state of pregnancy.”

• The velocity of the venous return may decrease by as much as 50% towards the end of pregnancy due to venous dilation caused by progesterone (already present in early pregnancy) and mechanical obstruction by the fetus, which increases with the progression of pregnancy.
  
• There is also a 20–200% increase in the amount of clotting factors such as fibrinogen and factors II (prothrombin), VII, VIII, X and XII, with a secondary increase after the delivery of the placenta. There is a concomitant decrease in the natural inhibitors of coagulation, indicated by lower protein S levels, and lower activity of the fibrinolytic system evidenced by higher levels of the plasminogen activator inhibitors 1 and 2 and thrombin activator fibrinolysis inhibitor.
  
• Endothelial vascular injury leading to release of tissue thromboplastin may occur during vaginal delivery and placental separation, but it is more pronounced after cesarean sections and traumatic deliveries with tissue rupture. Additional tissue damage by postpartum infections may further increase the risk.

Other factors that may additionally increase the risk include older age (> 35 years), obesity (pre-pregnancy BMI ≥ 30 kg/m2), bedrest, smoking, hypertension, venous damage from previous thrombosis (postphlebitic syndrome), large varicose veins, nephrotic syndrome, dehydration (e.g. in severe hyperemesis), intravenous drug abuse and long-distance air travel.

It is widely believed that there are still unknown “thrombopreventive and vasoprotective” mechanisms that seem to over-ride the thrombogenic diathesis of pregnancy in most women. It is also likely that future research will disclose more instances of coagulation abnormalities in many of the cases now labeled as “idiopathic.” The reason for the overwhelming occurrence of thrombosis in veins, rather than in arteries, during pregnancy seems to be the increased venous stasis plus the fact that pregnant women are generally young and free of arterial vascular disease (e.g. plaque) predisposing to arterial thrombosis. Nevertheless, some cases of arterial thrombosis (cerebral, cardiac, peripheral) have been reported in pregnant women although almost exclusively in those with the most severe forms of the antiphospholipid syndrome.

An increasing number of pregnant women with VTE are being found to have an underlying hereditary or acquired thrombophilia. A mutation in the factor V gene is associated with resistance to activated protein C (APC-R). This autosomal dominant-inherited defect causes a point mutation in the gene encoding for factor V, also known as Leiden, whereby glutamine is substituted for arginine at the amino acid site 506 which renders factor V resistant to proteolytic downregulation by activated protein C, leading to thrombin generation. The prevalence of APC-R (heterozygous state) in the general population is 2–7% with variations according to ethnicity: 3–7% in Caucasians, 2.21% in Hispanic Americans, 1.25% in Native Americans, 1.23% in African Americans, and 0.45% in Asian Americans. These patients also have an increased risk of adverse pregnancy outcomes, as stated in the introduction.

A mutation in the prothrombin G20210A gene, which results in increased levels of prothrombin, also increases the risk of thrombosis and adverse pregnancy outcomes. It occurs in 2% of the general population. Patients with both factor V and prothrombin gene compound mutation have a markedly increased thrombotic risk.

Antithrombin III deficiency is reported in 0.02–0.17% of the general population. It is considered to be a very thrombogenic condition.

The prevalence of protein S or protein C deficiencies is found in 0.14–0.5% of the general population, and the frequency during pregnancy in women with a previous VTE episode is 0–6% for protein S deficiency and 3–10% for protein C, but higher, 7–22%, among women who develop VTE post partum.

Hyperhomocysteinemia, caused by a mutation in the methylene tetrahydrofolate reductase C677T and A1298C genes, also increases the risk VTE during pregnancy and the risk of recurrent abortion by twofold. Other risks reported include neural tube defects.

Frequently, the precipitating event in persons with an inherited coagulation defect is the addition of an acquired risk for thrombosis such as the use of oral contraceptives, pregnancy, labor and delivery, immobilization, trauma or surgery.

Clinical diagnosis

The clinical manifestations of DVT may include pain, tenderness, swelling, warmth, redness, discoloration, cyanosis, a palpable cord, superficial venous dilation and, in severe cases, massive swelling and phlegmasia caerulea dolens. The diagnosis should never be made on clinical grounds alone because many other conditions may cause similar symptoms. It has been repeatedly shown that the diagnosis of DVT by clinical signs alone is accurate in less than 50% of cases, and in pregnancy, even less than 10% of suspected DVT is confirmed by ancillary testing. A careful personal and family history with emphasis on thromboembolic events should be taken.

Laboratory testing

General laboratory tests include a complete blood count (CBC), platelet count, partial thromboplastin time (PTT), thrombin clotting time and perhaps a bleeding time. A high-sensitivity D-dimer level, if not elevated, is considered strong evidence against the presence of VTE in nonpregnant patients. In pregnancy, a normal D-dimer is considered helpful if the compression ultrasound is also normal; if elevated, even if the ultrasound is negative, this indicates the need for further testing. D-dimer levels increase with the progression of pregnancy and spike if there are complications such as pre-eclampsia and placental abruption. One particular D-dimer assay (SimpliRED) might have a higher specificity, a superior negative predictive value and a lower rate of false-positive results in the first and second trimesters but it needs further validation.

Indications for thrombophilia testing include recurrent VTE, positive family history, and previous adverse pregnancy outcomes such as recurrent miscarriage (3 or more consecutive losses) before the 10th week, one or more fetal losses at or after 10 weeks, premature delivery (before 34 weeks) because of severe pre-eclampsia, eclampsia, HELLP syndrome, placental abruption, IUGR (≤ 5th percentile), or a combination of these factors (see Chapter 28). Whether or not screening for thrombophilia is warranted after the first VTE episode is still a matter of debate.

Basic thrombophilia screening tests include functional AT-III assay, functional proteins C and S assays, APC resistance factor V polymerase chain reaction and/or the more specific DNA-based testing for mutation of the factor V gene, and prothrombin G20210A polymerase chain reaction. Tests for acquired thrombophilias include the lupus anticoagulant, anticardiolipin and anti-β2 glycoprotein antibodies (IgG and IgM). Other possible factors to look for include a homocysteine level (testing 6 hours after l-methionine loading – 0.1g/kg bodyweight – might be more accurate than just a fasting level), thrombomodulin gene variants, protein Z levels, fibrinogen, plasminogen, plasminogen activator inhibitors and fibrinolysis inhibitors.

At present, the cost of thrombophilia screening is very high for routine use. The recommendation from the most recent consensus of the American College of Obstetricians and Gynecologists is to screen patients with a history of thrombosis, unexplained fetal loss at or after 20 weeks’ gestation, severe pre-eclampsia/HELLP occurring at less than 34 weeks’ gestation, severe IUGR or a family history of thrombosis. The basic screening tests should include factor V Leiden mutation, prothrombin G20210A mutation, functional protein C and S deficiencies, AT-III deficiency, lupus anticoagulant, anticardiolipin, anti-β2 glycoprotein antibodies and a homocysteine level.

Imaging

Contrast venography was, for years, the diagnostic gold standard for DVT and the sensitivity and specificity of all others tests were measured against it. It is rarely performed at present since it cannot be repeated serially because of fetal radiation and intravenous X-ray contrast exposure. It is an invasive procedure and carries risk of DVT itself.