Are the Wells Score and the Revised Geneva Score valuable for the diagnosis of pulmonary embolism in pregnancy?

To evaluate and to compare the predictive accuracy of the Wells score and the revised Geneva scores for the diagnosis of pulmonary embolism in the pregnant and postpartum population.

All pregnant or post-partum patients with a suspected PE and for whom a diagnostic imaging testing was performed (VQ scintigraphy or computed tomography pulmonary angiography) over a 3-year period were included in the study. The Wells and Revised Geneva Scores were calculated on the same cohort of patients and dichotomized into low and intermediate/high probability groups. The sensitivities and specificities were calculated. Overall accuracy was determined using receiver operator characteristic curve analysis.

A total of 103 patients were included. The overall prevalence of PE was 26.2% (27/103). Using the Wells Score, the prevalence of patients with PE in the low, intermediate and high probability categories was 20.5%, 43.5% and 50% respectively. Using the Revised Geneva Score, the prevalence of patients with PE in the low, intermediate and high probability categories was 17%, 36.2 and 33.3% respectively. In low risk groups of the Wells score and the simplified revised Geneva score the prevalence of PE was not statistically significantly different: respectively: 20.5% and 17.5% (p = 0,232). The agreement on clinical assessment using the Wells score and using the revised Geneva score was weak (κ coefficient = 0.154). In total, 26 (25.2%) patients were classified differently using the 2 scores. There was no significant difference in the overall accuracies of the Wells (0.67, 95% CI 0.54-0.79) and Revised Geneva Scores (0.64, 95% CI 0.52-0.76) as determined by the area under the ROC curves (P = 0.628). The sensitivity, specificity, PPV and NPV of the Wells score and the revised Geneva score were respectively: 40.7%, 81.5%, 44%, 79.4% and 62.9%, 59.2%, 35.4%, 81.8%.

The Wells score and the revised Geneva seems not to be valuable in the pregnant and post partum population. A specific risk score of PE for pregnant and postpartum population is needed to reduce the rate of unnecessary imaging studies, especially in this specific population were the use of radiation and contrast agent is problematic.