D-Dimer and thrombus burden in acute pulmonary embolism - 31/01/18

Doi : 10.1016/j.ajem.2018.01.048

Karsten Keller, MD ^a,^b,^{^⁎} , Johannes Beule, MD ^c, Jörn Oliver Balzer, MD ^d,^e,^{^{1
Both authors were co-shared last authors.}}, Wolfgang Dippold, MD ^c,^{^{1
Both authors were co-shared last authors.}}
^a Center for thrombosis and Hemostasis, University Medical Center Mainz, Mainz, Johannes Gutenberg-University Mainz, Germany
^b Cardiology I, Center of Cardiology, University Medical Center Mainz, Mainz, Johannes Gutenberg-University Mainz, Germany
^c Department of Internal Medicine, St. Vincenz and Elisabeth Hospital Mainz (KKM), Mainz, Germany
^d Department of Radiology and Nuclear Medicine, Catholic Clinic Mainz (KKM), Mainz, Germany
^e Department of Diagnostic and Interventional Radiology, University Clinic, Johann Wolfgang Goethe-University Frankfurt (Main), Frankfurt, Germany

^⁎Corresponding author at: Center for thrombosis and Hemostasis, University Medical Center Mainz, Johannes Gutenberg-University Mainz, Langenbeckstr. 1, 55131 Mainz, Germany.Center for thrombosis and HemostasisUniversity Medical Center MainzJohannes Gutenberg-University MainzLangenbeckstr. 1Mainz55131Germany

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Wednesday 31 January 2018
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Abstract

Background

Thrombus burden in pulmonary embolism (PE) is associated with higher D-Dimer-levels and poorer prognosis. We aimed to investigate i) the influence of right ventricular dysfunction (RVD), deep venous thrombosis (DVT), and high-risk PE-status on D-Dimer-levels and ii) effectiveness of D-Dimer to predict RVD in normotensive PE patients.

Methods

Overall, 161 PE patients were analyzed retrospectively, classified in 5 subgroups of thrombus burden according to clinical indications and compared regarding D-Dimer-levels. Linear regression models were computed to investigate the association between D-Dimer and the groups. In hemodynamically stable PE patients, a ROC curve was calculated to assess the effectiveness of D-Dimer for predicting RVD.

Results

Overall, 161 patients (60.9% females, 54.0% aged >70 years) were included in this analysis. The D-Dimer-level was associated with group-category in a univariate linear regression model (β 0.050 (95%CI 0.002–0.099), P = .043). After adjustment for age, sex, cancer, and pneumonia in a multivariate model we observed an association between D-Dimer and group-category with borderline significance (β 0.047 (95%CI 0.002–0.096), P = .058). The Kruskal-Wallis test demonstrated that D-Dimer increased significantly with higher group-category.

In 129 normotensive patients, patients with RVD had significantly higher D-Dimer values compared to those without (1.73 (1.11/3.48) vs 1.17 (0.65/2.90) mg/l, P = .049). A ROC curve showed an AUC of 0.61, gender non-specific, with calculated optimal cut-off of 1.18 mg/l. Multi-variate logistic regression model confirmed an association between D-Dimer >1.18 mg/l and RVD (OR2.721 (95%CI 1.196–6.190), P = .017).

Conclusions

Thrombus burden in PE is related to elevated D-Dimer levels, and D-Dimer values >1.18 mg/l were predictive for RVD in normotensive patients. D-Dimer levels were influenced by DVT, but not by cancer, pneumonia, age, or renal impairment.

Le texte complet de cet article est disponible en PDF.

Keywords : D-Dimer, Risk stratification, Pulmonary embolism, Deep venous thrombosis, Thrombus burden