Article Text

Download PDFPDF
19 Accuracy of biomarkers for venous thromboembolism in pregnancy: the diagnosis of pulmonary embolism in pregnancy (DiPEP) biomarker study
  1. Steve Goodacre1,
  2. Beverley Hunt2,
  3. Kiran Parmar2,
  4. Neil Shephard1,
  5. Kimberley Horspool1
  1. 1University of Sheffield
  2. 2Guy’s and St. Thomas’ NHS Foundation Trust


Objective To estimate the accuracy of biomarkers for venous thromboembolism (VTE) in pregnant and postpartum women with suspected pulmonary embolism (PE).

Design Observational cohort study augmented with additional cases with VTE.

Setting Emergency departments and maternity units at eleven prospectively recruiting sites.

Participants Pregnant or postpartum women presenting with suspected PE or diagnosed DVT.

Interventions Research nurses/midwives collected a blood sample from women with suspected PE or diagnosed DVT and data relating to diagnostic imaging, treatment and adverse outcomes. Blood samples were centrifuged, stored and then transported to Guy’s St Thomas Trust for analysis.

Main outcome measures Women were classified as having or not having VTE on the basis of diagnostic imaging, treatment and subsequent adverse outcomes. Primary analysis was limited to women with conclusive diagnostic imaging. Secondary analyses included women with clinically diagnosed or ruled out PE.

Results Usable blood samples were taken from 18 women with diagnosed DVT and 310 women with suspected PE, of whom 18 had PE confirmed by imaging and 247 had PE ruled out after imaging and were included in the primary analysis. Mean biomarker levels only significantly differed between women with and without PE for Clauss Fibrinogen (p=0.007), ELISA D-Dimer (p=0.001), Innovance D-Dimer (p=0.004), Thrombin Generation Lag Time (p<0.001), Thrombin Generation Time to Peak (p=0.001) and Plasmin Antiplasmin (p=0.004). The AUC (95% CI) for each biomarker was APTT 0.669 (0.570 to 0.768), BNP 0.549 (0.453 to 0.645), C-Reactive Protein 0.542 (0.445 to 0.639), Clauss Fibrinogen 0.589 (0.476 to 0.701), ELISA D-Dimer 0.668 (0.561 to 0.776), Innovance D-Dimer 0.651 (0.545 to 0.758), MRproANP 0.524 (0.418 to 0.630), PF1 +2 00.562 (0.462 to 0.661), Plasmin-antiplasmin 0.639 (0.536 to 0.742), Prothombin Time 0.613 (0.508 to 0.718), Thrombin Generation Lag Time 0.702 (0.598 to 0.806), Thrombin Generation Endogenous Potential 0.559 (0.437 to 0.681), Thrombin Generation Peak 0.596 (0.478 to 0.715), Thrombin Generation Time to Peak 0.655 (0.541 to 0.769), Tissue Factor 0.531 (0.424 to 0.638) and Troponin 0.597 (0.499 to 0.695). ROC analysis showed that only thrombin generation lag time had any potential to rule out PE with sufficient sensitivity while achieving meaningful specificity, with sensitivity of 97% and specificity of 25% at the threshold that optimised sensitivity. Repeat analysis excluding women who had received anticoagulation was limited by the small number (n=4) who had PE.

Conclusion Currently available biomarkers show little potential for aiding the diagnosis of suspected PE in pregnancy and postpartum.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.