D-dimer value in the diagnosis of pulmonary embolism-may it exclude only?

BACKGROUND

Pulmonary embolism (PE) is the third most common cause of death for cardiovascular diseases in Europe. Quick PE diagnosis is therefore crucial for prognosis improvement. It is critical to have suitable screening tests both to exclude PE as well to select patient with highest likelihood of PE occurrence. Currently D-dimer test is accepted as important tool useful to exclude PE in low risk patients. Our goal was to assess the D-dimer test positive prognostic value.

METHODS

A retrospective study based on medical record analysis of consecutively admitted patients to 9 wards of The University Clinical Center in Katowice who were hospitalized during four consecutive years was performed. Three hundred and seventy patients met the inclusion criteria for the study, which involved the D-dimer tests and computed tomographic pulmonary angiography (CTPA) performed during hospitalization. Assessed patients were divided into two groups: PE confirmed and PE excluded by CTPA.

RESULTS

We have found that patients with D-dimer levels higher than 2,152 ng/mL had significantly increased risk of PE [area under curve (AUC) of 0.69; 95% CI, 0.64-0.75; P<0.05]. Positive predictive value (PPV) reached the level of 53%, whereas negative predictive value (NPV) reached 82%. We also found that patients with the history of neoplasm and at >65 years of age had D-dimer cut-off point moved to the level of 2,652 ng/mL (AUC of 0.67; 95% CI, 0.52-0.81; P<0.05).

CONCLUSIONS

Whereas the NPV of the D-dimer test is generally accepted our results suggest that, in selected cases, an increased plasmatic D-dimer levels may have PPV in PE diagnosis. Patients with the history of neoplasm have higher cut-off D-dimer points above which we should consider increased PE likelihood. CTPA should be considered even for patients with low probability of PE when D-dimer values exceed four times the normal level.