Venous thromboembolism (VTE) constitutes the leading cause of direct maternal mortality in the developed world. These rare but potentially devastating events are responsible for 2.5-4.6 maternal deaths per 100,000 live births. In recent studies in which patients underwent objective testing, the incidence of VTE in pregnancy and postpartum ranged from 0.5–3 per 1000 pregnancies, representing a 5- to 10-fold increase in risk as compared to age-matched nonpregnant controls. Rodger and colleagues have previously objectively demonstrated a surprising incidence of pelvic deep vein thrombosis (DVT) using magnetic resonance (MR) venography (MRV) in moderate- to high-risk patients following cesarean section (46%, 95% confidence interval [CI], 21–73%). The clinical significance of these findings remains unknown, as there are no studies establishing the incidence of pelvic DVT using MRV in asymptomatic low-risk patients. Current guidelines recommend the administration of thromboprophylaxis in women considered at increased risk of VTE. These guidelines are based on expert clinical opinion as no large experimental trials have been conducted in this setting.

In a recent pilot study we randomized women at moderate to high thrombosis risk to receive either prophylactic low-molecular-weight heparin, or no intervention after cesarean section. The study showed a 46% (95% CI, 21–73%) incidence of pelvic vein thrombosis diagnosed by MRV. In the intervention group, 37.5% of women were diagnosed as having definite intraluminal filling defects. The clinical significance of these findings is uncertain. Intraluminal filling defects may represent thrombosis, imaging artifact, or perhaps physiologic postpartum changes in pelvic vasculature. The implications of our prior study included calling into question current diagnostic approaches to pelvic thrombosis and septic pelvic thrombophlebitis (SPT). A diagnosis of SPT is confirmed if fever of unknown origin and MRV thrombosis are documented in the postpartum patient.

To date, there are no studies using MRV to delineate the incidence and natural history of intraluminal filling defects in the postpartum period in patients at low thrombosis risk. Such a study could shed further light on our prior findings. We present the results of our study assessing the incidence of pelvic vein thrombosis after vaginal delivery in the low-risk obstetrical population as determined by MRV.