Here is our April, 2019 Quick Question, Adapted from Walenga JM. Chapter 35–Normal Hemostasis. In Keohane EM, Otto CN, Walenga JM. Rodak’s Hematology, Clinical Principles and Applications, 6th Edition, 2020. ISBN 978-0-323-53045-3.
A 33-year-old pregnant woman developed deep vein thrombosis in her left leg. She reported a family history—her mother had an episode of superficial thrombophlebitis and her brother had experienced a DVT during a 12-hour flight. Do you consider her DVT to be acquired, or is it likely to be associated with a congenital thrombosis risk factor?
A. Acquired thrombosis, no laboratory follow-up: 5 (20%)
B. Likely congenital risk factor, perform thrombosis risk profile 14 (56%)
C. Could be either 6 (24%)
The case represents the dilemma clinicians face when choosing a thrombosis risk profile. Here is the report of a 2009 study conducted by the Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX: Shen YM, Tsai J, Taiwo E, et al. Analysis of thrombophilia test ordering practices at an academic center: a proposal for appropriate testing to reduce harm and cost. Plos ONE 11: e0155326. doi:10.1371/journal.pone.0155326. The study generated these criteria for ordering a thrombophilia profile:
- Pregnancy morbidity: pre-eclampsia, intrauterine growth retardation
- Recurrent pregnancy loss, three or more instances
- Unprovoked arterial thrombosis
- Unprovoked venous thrombosis
The study also listed criteria for not ordering a profile:
- One or two pregnancy losses
- Provoked venous thrombosis: immobilization, surgery, trauma, and malignancy prior to or at the time of the event
- Provoked arterial thrombosis: hypertension, dyslipidemia, diabetes mellitus, atherosclerotic disease
- Testing without a prior thrombotic event or adverse pregnancy outcome
Results of this study generated an October 19, 2017 ASCP Choosing Wisely recommendation that limits orders for antithrombin, protein C, protein S. ASH posted a similar CW recommendation on December 4, 2013 that limits thrombophilia orders in an acute setting. The Dallas UTSMC lab developed a laboratory advisory panel that reduced thrombophilia orders by 90%.
Back to our case, we can conclude that a thrombophilia profile has little value if performed during the acute clotting episode or during therapy. In fact, pregnancy introduces variables that may invalidate the profile. For instance, protein S results are often below the reference interval during pregnancy. There may be some value in ordering profiles on the mother, brother, and on the patient some time after the birth, however both the patient’s and her brother’s events seem to be provoked, so in this case, clinicians may wish to collect more information on the patient’s family before ordering profiles.