Annals of Vascular Surgery
Volume 25, Issue 1 , Pages 9-14, January 2011

Incidence and Characteristics of Venous Thromboembolic Disease During Pregnancy and the Postnatal Period: A Contemporary Series

Presented at the 20th Annual Winter Meeting of the Peripheral Vascular Surgery Society, Vail, CO, January 29-31, 2010.

Division of Vascular Surgery, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY

published online 27 August 2010.

Article Outline

Background

To evaluate the incidence and characteristics of venous thromboembolic events (VTE) associated with pregnancy in a contemporary patient series.

Methods

We performed a retrospective review of 33,311 deliveries between June 2003 and June 2008. Patients with objective documentation of a VTE during pregnancy or the 3-month postnatal period were identified from hospital discharge International Classification of Disease Codes edition 9 codes. Diagnosis of deep venous thrombosis (DVT) was largely made by a Duplex ultrasound, whereas pulmonary embolism (PE) was diagnosed by a computerized tomographic angiography (CTA).

Results

Of 33,311 deliveries during the study period, 74 patients (0.22%) had a VTE. There were 40 incidents of DVT (0.12%) and 37 of PE (0.11%). DVT involved the iliac veins (6), the femoral or popliteal veins (16), the infrapopliteal veins (17), and the axillary vein (1). Most (57.5%) of the DVTs involved the left lower extremity. Thirty-eight (51.6%) of the VTEs occurred in the postnatal period, and of those 33 (87%) occurred within 1 week of delivery. Most of the postnatal VTEs (68%) were seen in patients who underwent a cesarean section. Among patients with VTE during pregnancy, there were 28% in the first trimester, 25% in the second, and 47% in the third. Events were distributed among maternal age groups as follows: 26% aged 13-24, 50% aged 25-34, and 24% aged 35-54. Of the 35 patients tested for a hypercoagulable disorder, 12 were found to have a positive test result. Five (6.8%) of these 74 patients had a prior history of VTE, with two having a hypercoagulable disorder. In addition, 45 of the 74 patients were on oral contraceptive therapy or received hormonal stimulation therapy before pregnancy. Patients with a VTE during pregnancy were treated with low molecular weight or unfractionated heparin. Most postnatal patients were treated with subcutaneous low molecular weight heparin and coumadin. Six inferior vena cava filters were placed in patients with bleeding complications as a result of anticoagulation. There were no deaths during the study period.

Conclusions

Comparing our results with historic controls (DVT: 0.04-0.14% and PE: 0.003-0.04%), the incidence of DVT in pregnancy has not changed significantly. We note, however, that the incidence of pulmonary embolus in our series is higher than previously reported. CTA has been used for the diagnosis of PE since the past decade. The increase in the rate of PE in the current series may be because of the higher sensitivity of CTA when compared with previous diagnostic modalities.

 

Back to Article Outline

Introduction 

The pregnancy and postnatal periods are associated with an increased risk of thromboembolic disease, particularly deep venous thrombosis (DVT) and pulmonary embolism (PE). The proposed mechanism has not been yet completely established; however, it is believed to arise from increased venous stasis caused by compression from the gravid uterus and hemostatic imbalance secondary to increased plasma levels of pregnancy-related hormones.1 Venous thromboembolic disease is a leading cause of maternal morbidity, and pulmonary embolus is the most common cause of maternal mortality in the developed world.2 Therefore, timely identification of patients with DVT and PE is extremely important. Although the overall risk of a venous thromboembolic event (VTE) is small, pregnant and postpartum patients have a 5 times greater chance of developing an event as compared with nonpregnant women of similar age.3

Numerous studies have examined the incidence and risk factors of venous thromboembolic disease in pregnancy.3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 Commonly accepted risk factors predisposing women to venous thrombosis during the puerperal period include age over 35, obesity (body mass index >30 kg/m2), multiparity (>3 prior deliveries), personal history of DVT or PE, inherited thrombophilia, surgery or cesarean delivery, smoking, and hormonal therapies.15 Other associated risk factors include gestational diabetes, placental abruption, and eclampsia.14

The goal of our study was to determine whether the incidence of DVT and pulmonary embolus has changed, using current diagnostic modalities in a contemporary series of patients. Prior studies have found a much higher incidence of DVT than pulmonary embolus.3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 These studies, however, were primarily conducted in the 1980s and 1990s using older diagnostic tests to examine for pulmonary embolus. Pulmonary embolus was typically diagnosed using ventilation perfusion scans. The established sensitivity of ventilation perfusion scans is 65% in modern series.16 Over the past 10 years, computerized tomographic angiography (CTA) for the diagnosis of PE has evolved into the test of choice for diagnosis of PE, with a significantly higher sensitivity of 83%.17 CTA has been used routinely at our institution for the diagnosis of pulmonary embolus for the past decade, including those in pregnant patients. Duplex ultrasonography, however, has been well established for the diagnosis of DVT for over 20 years. Its sensitivity and specificity is more than 90% for diagnosing proximal vein thrombosis in pregnant patients.18

Back to Article Outline

Materials and Methods 

This retrospective study was conducted at two affiliated urban tertiary care facilities. Institutional review board approval was obtained from the participating institutions. Women with a diagnosis of DVT or pulmonary embolus during pregnancy or the first 3 months thereafter, were selected through hospital discharge International Classification of Disease Codes, editions 9 and 10 between June 1, 2003 and June 30, 2008. The total number of pregnancies during this period was 33,311. Only patients with objective documentation of DVT or PE were included in this study. Acceptable tests for diagnosis of DVT included duplex ultrasonography, magnetic resonance venography, computer tomographic venography, and conventional venography. Examinations for diagnosis of PE included ventilation perfusion scanning (V/Q scan), CTA, and conventional pulmonary angiography. A V/Q scan was considered positive if an intermediate or high probability result was obtained.

The diagnosis for DVT was largely made by duplex ultrasonography and for PE by CTA. All patients with clinical suspicion for DVT or PE underwent duplex ultrasonography to exclude lower extremity DVT. Ultrasonography was performed on both extremities and examined the inferior vena cava (IVC) down to the level of the calf veins on all patients. Patients suspected of having a pulmonary embolus underwent multidetector CTA as a first-line diagnostic tool whenever possible. Lead shields were placed over the abdomen and pelvis of pregnant patients during CTA, and examination was performed inferiorly as far as the upper abdomen. Those refusing CTA or possessing an allergy to intravenous iodinated contrast underwent V/Q scanning for diagnosis.

Data collection was obtained by chart review, with data recorded on a standardized form. Recorded information included age at diagnosis of event, smoking status, prior oral contraceptive or hormonal stimulation therapy, parity, previous history of DVT or PE, documented history of a thrombophilia, timing during pregnancy or postpartum, side and proximal location of DVT, and initial treatment on diagnosis. Pearson correlation coefficients were calculated to determine any correlation between side of DVT and trimester of presentation and between side of DVT and proximal extent of the DVT. Testing for hypercoagulable genetic traits was performed on 35 of the patients at the primary practitioner's discretion. Chemical thromboprophylaxis during pregnancy was given only to women with a prior history of venous thromboembolic disease or a known thrombophilia. Prophylactic anticoagulation with subcutaneous heparin 5,000 units every 12 hours was given to all patients after delivery. Sequential compressions devices and Thromboembolic deterrent (TED) stockings were routinely placed on all pregnant women admitted to the labor unit and continued until discharge.

Back to Article Outline

Results 

During the period from June 1, 2003 to June 30, 2008, there were 33,311 deliveries. The total number of vaginal deliveries was 23,753 (71.3%) and the total number of cesarean deliveries was 9,558 (28.7%). Of these, a total of 74 patients (0.22%) had a VTE, or two in 1,000 pregnancies. There were 40 incidents of DVT (0.12%) and 37 of PE (0.11%). Six patients (0.02%) had simultaneous DVT and PE at the time of diagnosis. More than half of the events (38; 51.6%) occurred in the postpartum period. Of these, 26 (68%) developed after cesarean delivery, and 12 (32%) after vaginal delivery. The number of days post delivery ranged from 0 to 43 days, with the majority (n = 33, 87%) diagnosed in the first postpartum week. Among the 36 events during pregnancy, 10 (28%) were in the first trimester, nine (25%) in the second trimester, and 17 (47%) in the third trimester.

The average age of the affected patients was 33 years at the time of diagnosis (range, 13-41). In patients aged 13-24 years, there were 19 events (26%); in those aged 25-29 years, 19 (26%); 30-34 years, 18 (24%); and those aged >35 had 18 (24%). Of the 74 patients, 5 (6.8%) had a prior history of venous thrombosis during pregnancy. These patients had received thromboprophylaxis in the form of low-molecular weight heparin after their pregnancies had been confirmed in the first trimester. Twelve (16.2%) patients tested positive for a hypercoagulable disorder. The number of patients with a specific risk factor is shown in Table I and the number of risk factors per patient is depicted graphically in Figure 1. In addition, in patients with a VTE, no patient had gestational diabetes, two had placental abruption, and one had eclampsia.

Table I. Risk factors in patients with an event
Risk factorNumber of patients (% of total, n = 74)
Thrombophilia12 (16)
Prior VTE5 (7)
Age > 35 years18 (24)
Prior hormone therapy45 (61)
Obesity (BMI > 30 kg/m2)4 (5)
Smoking6 (8)
Multiparity (>3)19 (26)
Cesarean delivery26 (35)

VTE, venous thromboembolic event; BMI, body mass index.

Of the 40 DVT events, 23 (57.5%) were on the left, 14 (35%) on the right, and three (7.5%) were bilateral; 16 occurred during pregnancy, 24 postpartum. The proximal extent of the DVTs is illustrated in Table II. The majority extended above the knee (55%). There was one axillary vein thrombosis not related to any catheterizations. Statistical analysis was performed to determine whether there was any correlation between trimester and side of presentation of DVT in pregnant patients and whether there was any association between side of presentation and proximal extension of the DVT above the popliteal vein. There was no overall correlation between trimester and side of presentation (p = .212) during pregnancy and no correlation between side and extension above the popliteal vein (p = .338). There were three PEs diagnosed by V/Q scan alone and 34 by CTA. A total of 21 PEs were diagnosed during pregnancy and 16 in the postpartum period.

Table II. Location of deep venous thrombosis
Proximal extent of DVTNumber of patients (% of total)
Iliac6 (15)
Femoral12 (30)
Popliteal4 (10)
Infrapopliteal17 (43)
Axillary1 (2)
Left side23 (57)
Right side14 (35)
Bilateral3 (8)

DVT, deep venous thrombosis.

Pregnant patients were treated with subcutaneous unfractionated heparin (11 patients) titrated to activated partial thromboplastin times between 1.5 to 2 times control, or were given low molecular weight heparin (23 patients). Two pregnant patients required IVC filter placement because of bleeding complications. Postpartum patients were treated with intravenous heparin (five patients) or low molecular weight heparin (25 patients), along with adjunctive warfarin therapy. Four patients after delivery required IVC filter placement also for bleeding complications secondary to anticoagulation. None of the filters placed were retrievable because the patients did not opt to undergo a second procedure for removal of a retrievable filter.

Follow-up on the IVC filter patients averaged 1 year (range, 1 month-3.5 years), there were no recurrent PEs or insertion-related DVTs. Each IVC filter patient received a surveillance duplex in office follow-up, ranging from 1 month to 1 year after insertion.

Two patients with calf DVTs received no treatment and had follow-up duplex examinations 2 weeks later that showed no progression of the thrombosis with partial resolution. At 6 weeks-2 months, repeat duplex examinations showed complete resolution. There were no deaths during the study period.

Back to Article Outline

Discussion 

Pregnancy and the postpartum period place women of reproductive age at risk for development of DVT and PE. The two underlying mechanisms hypothesized to play a role during pregnancy are hormonal induced hypercoagulability and mechanical compression from the gravid uterus.19 Normal pregnancy results in an increased level of clotting factors VII, VIII, X, von Willebrand Factor, and higher concentrations of fibrinogen.20 In addition, anticoagulant proteins decrease. Free levels of protein S decrease because of increased binding with complement C4b, and antithrombin III is partially lost in the urine.20

Decreased venous capacitance and outflow occur from hormonal effects on the venous system.19 The gravid uterus also exerts compressive force on the iliac veins. In pregnant patients, DVT is more likely to occur in the left leg. This has been confirmed in previous studies9, 10, 11 and our study also demonstrated that 57% of DVT occur in the left leg. According to a previous study, relative stenosis of the left common iliac vein between the lumbar vertebral body and right common iliac artery occurs.21 This mechanism may help explain why all six of the iliac vein DVTs in our study occurred on the left side. In addition, all six iliac DVTs occurred after the uterus had significant time to enlarge, with one diagnosed in the second trimester, two in the third, and three during the postpartum period. However, we found no overall correlation between trimester and side of presentation during pregnancy.

Estimating the likelihood of developing a VTE and identifying high-risk patients are important for patient management during the puerperal period. Our risk factor analysis reveals that 85% of patients have at least one of the following risk factors: inherited thrombophilia, personal history of a VTE, age >35, previous hormonal stimulation or oral contraceptive use, obesity, history of smoking, multiparity (>3), and cesarean delivery. Previous studies have identified a personal history of venous thrombosis and an inherited thrombophilia as the most significant risk factors for development of a VTE event.9, 10 A high index of suspicion should especially be maintained in patients with one or more of these risk factors, complaining of symptoms relevant to venous thromboembolic disease, and antepartum thromboprophylaxis may be considered.

The diagnostic modalities used to diagnose venous thromboembolic disease may play a role in determining the overall incidence in a pregnant population. In our study, we found that 0.12% of our patients had a DVT. Previous studies have estimated the incidence of DVT in the pregnant and postpartum period to be between 0.04 and 0.14%.3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 We hypothesize that our rate of DVT is similar to historic controls because duplex ultrasonography was the primary mode of diagnosis in all studies. Of interest is the much higher incidence of PE in our series as compared with previous reports. We found an overall rate of PE of 0.11%, whereas historically PE has only occurred in 0.003-0.04% of patients.3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 Our study used multidetector computed tomography for diagnosis of pulmonary embolus whenever possible, which has an established positive predictive value of 96% when clinically concordant.22 The established high sensitivity may help explain this much higher incidence.

Our study is limited by its retrospective design and lack of a control group. The demographic data of the 33,311 pregnancies during this period could not be determined. It is unknown whether the risk factors present in our population may have affected the overall incidence of venous thromboembolic disease compared with previous studies. The retrospective study design and the use of International Classification of Disease codes to identify patients with an event are also limitations of this study because not all patients may have been detected.

Pregnant patients with venous thromboembolic disease have traditionally been treated with unfractionated heparin through intravenous or subcutaneous routes. Warfarin is avoided because of the development of congenital defects with its use during pregnancy. Recent studies comparing administration of subcutaneous low molecular weight heparin with unfractionated heparin during pregnancy demonstrate similar safety profiles and efficacy.23 Low molecular weight heparins have the advantage of not needing to monitor activated partial thromboplastin times and a lower chance of developing osteopenia compared with unfractionated heparin.23 In our series, pregnant patients with VTE were treated with low molecular weight heparin during the first, second, and early third trimesters. Twenty-three patients met this criterion and were treated successfully during their pregnancy. Two weeks before anticipated delivery or during admission to the labor department, low molecular weight heparin was changed to therapeutic doses of subcutaneous heparin because of its shorter half-life. The 11 patients diagnosed late in the third trimester were initially treated with subcutaneous heparin without labor complications. All patients diagnosed during pregnancy were anticoagulated throughout their pregnancy and for at least 6 weeks postpartum. The total minimum duration of therapy was 6 months, as recommended by the American College of Chest Physicians.24

Placental abruption contraindicated anticoagulation in two pregnant patients, requiring IVC filter placement. These filters were placed in the infrarenal position. There were no peripartum complications from filter placement in these patients; however, it is too early to determine the long-term effects. IVC filter use during pregnancy has previously been found safe, with both infrarenal and suprarenal placement.25, 26, 27 Suprarenal placement has been advocated by Greenfield to avoid potential contact between the gravid uterus and the filter and to provide protection against thromboembolism from pelvic or ovarian veins.25 Long-term results with suprarenal Greenfield vena cava filter placement (Boston Scientific, Natick, MA) during pregnancy has been found to be safe.25 We chose infrarenal placement because only the Trapease filter (Cordis, Bridgewater, NJ) was available and suprarenal placement is not well established with this device.

A total of 30 postpartum patients with a VTE were successfully treated with intravenous heparin or low molecular weight heparin, depending on practitioner preference. Subesequently, these patients were bridged to Warfarin therapy. These patients were prescribed anticoagulant therapy for a total of 6 months. The four patients who developed bleeding complications after anticoagulation were successfully treated with infrarenal IVC filters.

Thromboprophylaxis during pregnancy is controversial and there are no randomized controlled studies examining its benefit.28 Current practice guidelines from the American College of Chest Physicians recommends postpartum pharmacologic thromboprophylaxis in patients with a known thrombophilia or a personal history of venous thrombosis.24 The addition of antepartum pharmacologic prophylaxis is recommended for all patients with antithrombin III deficiency and should be considered in those with other inherited thrombophilias or a prior VTE event.24 Further studies are needed to determine which patients may benefit from prophylactic pharmacologic therapy during pregnancy or the postpartum period.

Although the overall risk of venous thromboembolic disease is small, there is a significant group of patients who will develop this condition with the potential for serious morbidity and mortality. Patients are at risk during all three trimesters of the pregnancy period and the immediate postpartum period. Practitioners should especially educate patients of the symptoms of VTE in those with one or more risk factors. Further studies are needed to determine which patients might benefit from pharmacologic prophylaxis in the antepartum and postpartum periods.

Back to Article Outline

References 

  1. Kujovich JL. Hormones and pregnancy: thromboembolic risks for women. Br J Haematol. 2004;126:443–454
  2. Chang J, Elam-Evans LD, Berg CJ, et al. Pregnancy-related mortality surveillance—United States, 1991-1999. MMWR Surveill Summ. 2003;52:1–8
  3. Jacobsen AF, Skjeldstad PM, Sandset PM. Incidence and risk patterns of venous thromboembolism in pregnancy and puerperium—a register-based case-control study. Am J Obstet Gynecol. 2008;198:233.e1–233.e7
  4. Heit JA, Kobbervig CE, James AH, Petterson TM, Bailey KR, Melton LJ. Trends in the incidence of venous thromboembolism during pregnancy or postpartum: a 30-year population-based study. Ann Intern Med. 2005;143:697–706
  5. James KV, Lohr JM, Deshmukh RM, Cranley JJ. Venous thrombotic complications of pregnancy. Cardiovasc Surg. 1996;4:777–782
  6. Larsen TB, Sorensen HT, Gislum M, Johnsen SP. Maternal smoking, obesity, and risk of venous thromboembolism during pregnancy and the puerperium: a population-based nested case-control study. Thromb Res. 2007;120:505–509
  7. Soomro RM, Bucur IJ, Noorani S. Cumulative incidence of venous thromboembolism during pregnancy and puerperium: a hospital-based study. Angiology. 2002;53:429–434
  8. Stein PD, Hull RD, Kayali F, et al. Venous thromboembolism in pregnancy: 21-year trends. Am J Med. 2004;117:121–125
  9. Ghereman RB, Goodwin TM, Leung B, Byrne JD, Hethumumi R, Montoro M. Incidence, clinical characteristics, and timing of objectively diagnosed venous thromboembolism during pregnancy. Obstet Gynecol. 1999;94:730–734
  10. Witlin AG, Mattar FM, Saade GR, Van Hook JW, Sibai BM. Presentation of venous thromboembolism during pregnancy. Am J Obstet Gynecol. 1999;181:1118–1121
  11. Ginsberg JS, Brill-Edwards P, Burrows RF, et al. Venous thrombosis during pregnancy: leg and trimester of presentation. Thromb Haemost. 1992;67:519–520
  12. Andersen BS, Steffensen FH, Sorensen HT, et al. The cumulative incidence of venous thromboembolism during pregnancy and puerperium. Acta Obstet Gynecol Scand. 1998;77:170–175
  13. James AH, Tapson VF, Goldhaber SZ. Thrombosis during pregnancy and the postpartum period. Am J Obstet Gynecol. 2005;193:216–219
  14. Simpson EL, Lawrenson RA, Nightingale AL, Farmer RDT. Venous thromboembolism in pregnancy and the puerperium: incidence and additional risk factors from a London perinatal database. Br J Obstet Gynaecol. 2001;108:56–60
  15. Ageno W, Squizzato A, Garcia D, Imberti D. Epidemiology and risk factors of venous thromboembolism. Semin Thromb Hemost. 2006;32:651–658
  16. Mayo JR, Remy-Jardin M, Muller NL, et al. Pulmonary embolism: prospective comparison of spiral CT with ventilation-perfusion scintigraphy. Radiology. 1997;205:447–452
  17. Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med. 2006;354:2317–2327
  18. Rutherford SE, Phelan JP. Deep venous thrombosis and pulmonary embolism in pregnancy. Obstet Gynecol Clin North Am. 1991;18:345–369
  19. James AH. Venous thromboembolism in pregnancy. Arterioscler Thromb Vasc Biol. 2009;29:326–331
  20. Bremme KA. Haemostatic changes in pregnancy. Best Pract Res Clin Haematol. 2003;16:153–168
  21. Cockett FB, Thomas ML. The iliac compression syndrome. Br J Surg. 1965;52:816–821
  22. Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed tomography for acute pulmonary embolism. N Eng J Med. 2006;354:2317–2327
  23. Nelson-Piercy C. Low molecular weight heparin for obstetric thromboprophylaxis. BJOG. 1994;101:6–8
  24. Bates SM, Greer IA, Pabinger I, Sofaer S, Hirsh J. Venous thromboembolism, thrombophilia, antithrombotic therapy, and pregnancy: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(Suppl 6):844S–886S
  25. Greenfield LJ, Cho KJ, Proctor MC, Sobel M, Shah S, Wingo J. Late results of suprarenal Greenfield vena cava filter placement. Arch Surg. 1992;127:969–973
  26. AbuRahma AF, Mullins DA. Endovascular caval interruption in pregnant patients with deep vein thrombosis of the lower extremity. J Vasc Surg. 2001;33:375–378
  27. Gupta S, Ettles DF, Lindow SW. Inferior vena cava filter use in pregnancy: preliminary experience. BJOG. 2008;115:785–788
  28. Dresang LT, Fontaine P, Leeman L, King VJ. Venous thromboembolism during pregnancy. Am Fam Physician. 2008;77:1709–1716

PII: S0890-5096(10)00221-9

doi:10.1016/j.avsg.2010.04.003

Annals of Vascular Surgery
Volume 25, Issue 1 , Pages 9-14, January 2011

Article Tools

* Image Usage & Resolution