D-dimer level significance for deep vein thrombosis screening in the third trimester: a retrospective study

BACKGROUND: Venous thromboembolism often develops after surgery and childbirth, resulting in death in some cases. Although early deep vein thrombosis (DVT) detection can predict pulmonary thromboembolism, there is no early screening method for DVT in pregnant women. Lack of consensus regarding significance or setting and cut-off value interpretation of D-dimer levels further impedes venous thromboembolism screening in pregnant women. This study aimed to examine the utility of third-trimester serum D-dimer levels as a screening test for DVT during pregnancy and to determine the frequency of asymptomatic DVT using lower-limb compression ultrasonography. METHODS: This single-center retrospective study included 497 pregnant women who underwent elective cesarean section at term in our hospital between January 2013 and December 2019. Serum D-dimer levels were preoperatively measured at 32–37 weeks’ gestation. The presence or absence of DVT in patients with serum D-dimer levels ≥ 3.0 µg/ml, the cut-off value, was examined using compression ultrasonography. In all patients, the presence or absence of clinical venous thrombosis (symptoms such as lower-limb pain, swelling, and heat sensation) was examined within 4 postoperative weeks. The Royal College of Obstetricians and Gynecologists Guideline 2015 was referred to determine risk factors for the onset of venous thrombosis during pregnancy. Among those, we examined the risk factors for DVT that result in high D-dimer levels during pregnancy. RESULTS: The median age and body mass index were 35 (20–47) years and 21.2 (16.4–41.1) kg/m(2), respectively. Further, the median gestational age and D-dimer levels were 37 weeks and 2.1 (0.2–16.0) µg/ml, respectively. Compression ultrasonography was performed on 135 (26.5%) patients with a D-dimer level ≥ 3.0 µg/ml, with none of the patients showing DVT. All patients were followed up for 4 postoperative weeks, with none presenting with venous thromboembolism. Multivariate analysis showed that hypertensive disorders of pregnancy are an independent risk factor for venous thromboembolism that causes high D-dimer levels (odds ratio: 2.48, 95% confidence interval: 1.05–6.50, P = 0.04). CONCLUSION: There may be low utility in screening for DVT using D-dimer levels in the third trimester. Further, prepartum asymptomatic DVT has a low frequency, indicating the low utility of compression ultrasonography. TRIAL REGISTRATION: Institutional Review Board of Tottori University Hospital (IRB no. 20A149).