Christoph Weber MD, Presenter: Nothing to Disclose

Henning Sturm MD, Abstract Co-Author: Nothing to Disclose

Gerhard B. Adam MD, Abstract Co-Author: Nothing to Disclose

Martin Lorenzen, Abstract Co-Author: Nothing to Disclose

The detection of pulmonary embolism (PE) with respect to the correlation between d-dimers and multislice computer tomography (MSCT) of the chest.

77 patients, average age of 65 y (25-76 y) with suspected pulmonary embolism and increased d-dimer levels (>190 μg/l) received contrast enhanced MSCT. The data was analyzed by three radiologists using the ROC method. Their findings were geared to the following criteria: existence of PE, localization of the embolism as well as the contrasting and diameter of the vessel lumina of the truncus pulmonalis and the right and left pulmonary artery. The patients were divided into two groups (Group 1: MSCT: existent PE; group 2: MSCT: non existent PE). The correlation between MSCT and d-dimers was analyzed through student t-test. The observer variability was captured by kappa coefficient.

34 of 77 (44%) patients had no MSCT based evidence of PE (Group 2). The average d-dimer value in this group was at 675+/-666. 43 of 77 (56%) patients showed signs of a PE in MSCT (Group 1). In this group the average d-dimer value was 943+/-1175. The correlation of the d-dimer value between the two groups showed no statistically significant difference with regard to the evidence of PE based on MSCT (p=0.377). The ROC Analysis results regarding the existence or non existence of PE showed a correlation coefficient of 0.89 (kappa coefficient (kc): 0.78), regarding an embolism in the truncus pulmonalis of 0.97 (kc: -0.18), in the pulmonary arteries of 0.86 to 0.94 (kc: 0.65 – 0.88), in the segmental arteries of 0.89 to 0.90 (kc: 0,78 – 0.8) and regarding the subsegmental arteries of 0.77 to 0.79 (kc: 0,5 -0,53). The average density of the contrasted vessels was 283+/-91 HE (Group 1) and 242+/-65 HE (Group 2). In group 1, 79% of patients had a PE in the subsegmental, 89% in the segmental, 59% in the pulmonary arteries and 2% in the truncus pulmonalis.

The level of d-dimers does not correlate reliably with the evidence of PE gathered by MSCT, therefore the increased level of d-dimers does not solely verify the need for MSCT to exclude PE.

MSCT to rule out PE can be rejected in the context of low clinical suspicion and negative d-dimers. The d-dimer value needs implementation in an algorithm based on clinical assessment to reduce MSCT.

Weber, C, Sturm, H, Adam, G, Lorenzen, M, Do Increased d-dimer Levels Justify the Use of Multislice Computed Tomography to Rule out Pulmonary Embolism?.  Radiological Society of North America 2008 Scientific Assembly and Annual Meeting, February 18 - February 20, 2008 ,Chicago IL. http://archive.rsna.org/2008/6016918.html