Accuracy of the YEARS Algorithm Compared to Thoracic Imaging for the Diagnosis of Pulmonary Embolism in Pregnant and Postpartum Patients

Abstract

The number of morbidity and mortality caused by Pulmonary Embolism (PE) has increased as of recent. Diagnosing PE during pregnancy and postpartum period is particularly challenging given the overlapping symptoms with physiological changes of pregnancy. Recent evidence suggests that the pregnancy-adapted YEARS algorithm, combining clinical probability assessment with D-dimer measurement, may exclude PE without the need for thoracic imaging, thereby reducing unnecessary radiation exposure.
This systematic review and meta-analysis intends to evaluate the diagnostic accuracy of the pregnancy-adapted YEARS algorithm in comparison with thoracic imaging among pregnant and postpartum individuals that are suspected of PE.
Three databases were searched systematically, including PubMed, Scopus, and Cochrane. Eligible studies included pregnant or postpartum women suspected of having PE who were assessed with the YEARS algorithm, using CT pulmonary angiography (CTPA) or ventilation–perfusion (V/Q) scan as the reference standard. Risk of bias was assessed meticulously using the QUADAS-2 tool. Pooled sensitivity, specificity, as well as the area under the curve (AUC) were calculated by Meta-Disc utilizing a random-effects model.
Five studies comprising 1,036 patients, ultimately with low risk of bias were included. The pregnancy-adapted YEARS algorithm showed a pooled sensitivity of 1.00 (95% CI: 0.94–1.00), pooled specificity of 0.12 (95% CI: 0.10–0.14), and an AUC of 0.72, which indicated adequate rule-out ability but limited utility for ruling in PE.
The YEARS algorithm that has been adapted for pregnancy may be safely used as a screening tool in excluding PE in pregnant and postpartum women, thereby reducing unnecessary maternal and fetal exposure to radiation. However, confirmatory thoracic imaging remains essential for positive cases.