Hemostatic and inflammatory biomarkers are associated with functional limitations after venous thromboembolism: Results from a prospective cohort study

Introduction: Functional limitations often persist in patients with venous thromboembolism (VTE) despite effective anticoagulant treatment. The underlying mechanisms are not well understood, and it remains uncertain whether biomarker candidates are associated with functional limitations following VTE. Therefore, we aimed to investigate the association between hemostatic, inflammatory, and cardiovascular biomarker candidates and functional limitations three months after VTE.

Method: We conducted a prospective, observational cohort study, including patients with acute VTE within 21 days of diagnosis. Biomarker candidate levels (D-dimer, fibrinogen, factor VIII (FVIII), von Willebrand Factor antigen (VWF), C-reactive protein (CRP), troponin T, N-terminal pro b-type natriuretic peptide (proBNP)) were measured at study inclusion and after three months. Functional limitations after three months were evaluated with the post-VTE functional status (PVFS) scale (0-4, higher values indicating more limitations). The association of biomarker candidates with functional limitations was assessed using proportional odds logistic regression models, adjusting for confounders. Furthermore, we compared proportions of patients per PVFS scale categories in patients with high (≥50^th percentile) and low (<50^th percentile) biomarker levels. The discriminative performance of biomarker candidates for the presence of PVFS scale>1 was assessed with the area under the receiver operating characteristic curve (AUC-ROC).

Results: We included 290 patients (41.4% women) with a median age of 54.9 years (interquartile range, IQR: 43.1-64.2), comprising 134 patients (46.2%) with pulmonary embolism and 174 (60.0%) with unprovoked VTE. After three months, median PVFS scale was 1 (IQR: 0-2). FVIII, VWF, D-dimer, fibrinogen, and CRP measured at baseline and D-dimer measured at three months were independently associated with functional limitations at three months ([Fig. 2]). When comparing high versus low biomarker candidate levels, VWF and D-dimer showed the best performance in differentiating patients ([Fig. 1]). The highest observed AUC-ROCs (95% confidence interval) for the presence of functional limitations at three months were 0.62 (0.55-0.69) for VWF measured at baseline, 0.62 (0.55-0.69) for D-dimer measured at three months, and 0.63 (0.55-0.69) for fibrinogen measured at three months.

Fig. 2 Association of biomarkers at baseline and three months with functional limitations at three months ^a Adjusted for sex, age, BMI, VTE type (i.e., pulmonary embolism or deep vein thrombosis), history of cardiovascular or pulmonary disease, and post-VTE functional status scale before VTE diagnosis. Age was modelled as continuous variable using restricted cubic splines with four knots at the 5th, 35th, 65th, and 95th percentile. ^b Unit of FVIII and VWF is%. ^c Troponin T and proBNP were only evaluated in patients with pulmonary embolism, therefore the multivariable model for these variables did not include VTE type.

Fig. 1 Proportions of patients per PVFS scale category at three months with high and low biomarker levels Biomarker candidate levels were measured at baseline. Dark green refers to PVFS scale of 0 (no functional limitations), light green to 1 (negligible functional limitations), yellow to 2 (slight functional limitations), orange to 3 (moderate functional limitations), and red to 4 (severe functional limitations).

Conclusion: We identified biomarker candidates which were independently associated with persisting functional limitations after three months. These biomarker candidates might aid in early identification of patients at higher risk of developing persistent functional limitations and could be suggestive of an underlying pathophysiological mechanism.

Publikationsverlauf

Artikel online veröffentlicht:
13. Februar 2025