Fewer tests reduce radiation for pregnant patients with thromboembolic disease

Suspicion of venous thromboembolism (VTE) in a pregnant patient will quickly bring a radiologist
to a choice, where the next step holds potentially significant consequences for both the
mother and unborn child. At ESTI 2012, Fergus Gleeson, MD, from the Oxford Radcliffe Hospitals
NHS Trust, will present colleagues a way forward for these cases, which are not frequently encountered in practice, yet represent a leading, preventable cause of maternal mortality.

Fergus Gleeson
Fergus Gleeson

“It is simply not possible to treat pregnant patients suspected for pulmonary embolic disease in the same manner as those who are not pregnant, and the investigation needs to be approached with care,” he said. “The standard D-dimer test that has a high negative  predictive value for non-pregnant patients is of no use here, as the levels are elevated in pregnancy,” he said.

The physiologic changes in pregnancy mimic pulmonary embolic disease with swollen legs,  pelvic pain, and possibly rib pain, he explained, making it difficult for a referring physician to make a clinical assessment. Yet concerned they may miss a potentially fatal case of VTE, and with the threshold for suspecting the diagnosis lowered, there is a potential for over-diagnosis. To exclude a chest infection or pneumothorax, the radiologist can proceed with confidence in ordering a chest x-ray. In a second step, compression ultrasonography of the lower limb is required to exclude deep vein thrombosis. “The diagnostic yield for this examination will be low”, Gleeson said,“but it has the advantage that if a clot is found, it points to appropriate treatment without having exposed either the mother or the fetus to any risk of further radiation.”

With inconclusive results to this point, the radiologist moves onto uncertain ground in considering a computed tomographic pulmonary angiography (CTPA), which is considered the gold standard test for patients who are not pregnant and radionuclide scintigraphy. Both tests expose the mother to a high dose of radiation with a risk of latent carcinogenic effects for radiosensitive breast tissue and uncertain effects for the fetus. Yet, while it is important to avoid ionizing radiation exposure during pregnancy, the risks of an undiagnosed pulmonary embolism (PE) are much greater than the theoretical risk to the fetus from diagnostic imaging. Gleeson and colleagues at the Oxford Radcliffe NHS Trust in Oxford conducted a number of studies testing different algorithms for patients with suspected pulmonary emboli to determine which produces the highest yield with the lowest radiation dose to both the mother and the baby.

First published in Clinical Radiology in 2006, their protocol calls for half-dose perfusion scintigraphy as the first line investigation in pregnant women with suspected PE. Evidence subsequently published in 2010 in The Lancet found this approach results in “excellent diagnostic accuracy in pregnant patients with a low non-diagnostic rate.”

In his presentation at ESTI, Gleeson said he is “going to discuss using the best test in the best order and reducing dose. And I will discuss the limitations of each of the tests, chest x-ray, ventilation perfusion scanning, and CT pulmonary angiography.” “Up to a third of pregnant women risk being exposed to radiation for no measurable benefit if CTPA is used first-line,” he said. Adding that CTPA should be reserved for the small number of pregnant patients with underlying lung disease or a non-diagnostic lung scintigram.

”Where the incidence of non-diagnostic scans with lung scintigraphy is high in nonpregnant patients, mainly as a result of chronic lung disease, “a great advantage we have is that women who are pregnant are young, fit, and often have good cardio-respiratory reserve,” he said. “That really stacks the odds in our favor. It means we can do simple tests that, if normal, will exclude pulmonary emboli.” “All of us in thoracic imaging will come across pulmonary emboli in pregnant women, and it is a difficult topic,” said Gleeson. “This presentation will review the current status of available information, will provide a discussion of the significance of the evidence, and will propose imaging algorithms that are highly useful.”

For your diary
Thromboembolic disease –
imaging algorithms in pregnancy
Session ”ESTI meets the British
Society of Thoracic Imaging –
pulmonary vascular disease”
Fergus Gleeson, Oxford, UK
Friday June 22 at 17:00 – 17:30





Fergus Gleeson was appointed Consultant Radiologist at the Oxford University Hospitals NHS Trust in 1992 and Professor of Radiology, Oxford University, in 2010. He has written over 150 papers and book chapters on Thoracic Imaging. He was editor of the British Journal of Radiology and is currently on the scientific committee boards of the European Society of Thoracic Imaging, European Congress of Radiology, and the American College of Chest Physicians. He is currently the Director of Imaging for the Oxford NIHR Biomedical Research Centre and the Director of Academic Radiology in Oxford.




More on the subject:
Read all latest stories

Related articles


Chest X-ray, CT and more

Imaging the coronavirus disease COVID-19

Chest X-ray is the first imaging method to diagnose COVID-19 coronavirus infection in Spain, but in the light of new evidence this may change soon, according to Milagros Martí de Gracia, Vice…


Emergency care

Keep it simple and straightforward

Emergency medicine requires smooth, patient-oriented and perfectly timed cooperation of several clinical disciplines.


ESTI 2016 cancelled due to security fears

Citing concerns over attendee safety, the European Society of Thoracic Imaging (ESTI) has cancelled its 2016 annual scientific meeting which had been scheduled for May 25 to 28 in Istanbul.

Related products