Accuracy of D-dimer:fibrinogen ratio to diagnose pulmonary thromboembolism in patients admitted to intensive care units

Introduction

Pulmonary thromboembolism (PTE) may increase D-dimer and decrease fibrinogen levels. However, in settings such as intensive care units (ICU) and in long-term hospitalised patients, several factors may influence D-dimer and fibrinogen concentrations and make them unreliable indicators for the diagnosis of PTE. The aim of this study was to evaluate the accuracy of D-dimer:fibrinogen ratio (DDFR) for the diagnosis of PTE in ICU patients.

Methods

ICU patients who were suspected of having a first PTE and had no history of using anti-coagulants and contraceptives were included in the study. Levels of D-dimer and fibrinogen were measured for each patient prior to any intervention. Angiography or CT angiography was done in order to establish a definite diagnosis for each patient. Suitable analytical tests were performed to compare means.

Results

Eighty-one patients were included in the study, of whom 41 had PTE and 40 did not. Mean values of D-dimer and fibrinogen were 3.97 ± 3.22 μg/ml and 560.6 ± 197.3 mg/dl, respectively. Significantly higher levels of D-dimer (4.65 ± 3.46 vs 2.25 ± 2.55 μg/ml, p = 0.006) and DDFR (0.913 ± 0.716 vs 483 ± 0.440 × 10^-3, p = 0.003) were seen in PTE patients than in those without PTE. Receiver operating characteristic (ROC) analysis showed a 70.3% sensitivity and 70.1% specificity with a D-dimer value of 2.43 μg/ml (AUC = 0.714, p = 0.002) as the best cut-off point; and a 70.3% sensitivity and 61.6% specificity with a DDFR value of 0.417 × 10^-3 (AUC = 0.710, p = 0.004) as the best cut-off point. In backward stepwise regression analysis, DDRF (OR = 0.72, p = 0.025), gender (OR = 0.76, p = 0.049) and white blood cell count (OR = 1.11, p = 0.373) were modelled (p = 0.029, R² = 0.577).

Conclusion

For diagnosis of PTE, DDFR can be considered to have almost the same importance as D-dimer level. Moreover, it was possible to rule out PTE with only a D-dimer cut-off value < 0.43 mg/dl, without the use of DDFR. However, these values cannot be used as a replacement for angiography or CT angiography