Objective: To evaluate the potential utility of a renal function adjusted D-dimer cut-off value for improving the exclusion of pulmonary embolism.
Methods: Retrospective analyses were performed for 1 784 inpatients and outpatients with low and moderate probability of pulmonary embolism at Anzhen Hospital from January 2011 to June 2013. The Well's score was used. The diagnoses of pulmonary embolism were confirmed by computed tomography pulmonary angiogram and ventilation-perfusion scan. Based upon estimated glomerular filtration rate (GFR), they were divided into three subgroups of normal renal function, mild renal impairment and moderate renal impairment. Negative D-dimer was defined as a level of age-standardized D-dimer value<500 µg/L. The proportions of patients with negative D-dimer and the utility of D-dimer for ruling out pulmonary embolism were compared between three subgroups. A new D-dimer cut-off point in patients with renal impairment was developed by receiver operating characteristics (ROC) curves and the effect of diagnostic efficiency of ruling out pulmonary embolism with renal function adjusted D-dimer cut-off was assessed.
Results: The medians of D-dimer of three subgroups with normal renal function, mild renal impairment and moderate renal impairment were 291.5, 995.5 and 1 901.5 µg/L (P<0.001) respectively. The sensitivity of negative D-dimer for ruling out pulmonary embolism was 99%-100% and the specificity 70%, 42% and 23% in three subgroups. The number of patients needed to test (NNT) for one negative test of D-dimer in three subgroups of different renal function was 3.32, 3.58 and 3.95 respectively. The new D-dimer cut-off value increased to 1.2 and 1.75 times of old one in patients with mild and moderate renal impairments and the proportion of patients with a negative D-dimer level rose from 48.7% to 53.0% as compared with old D-dimer cut-off value. The sensitivity and specificity of negative D-dimer for ruling out pulmonary embolism was 98% and 62% with the new cut-off value versus 99% and 57% with the old cut-off value and the NNT of D-dimer for ruling out pulmonary embolism declined from 3.52 to 3.34.
Conclusion: The renal function adjusted D-dimer cut-off point can improve the diagnostic efficiency of D-dimer test for ruling out pulmonary embolism.