D-Dimer for the Diagnosis of DIC
D-Dimer for the Diagnosis of DIC
To evaluate the diagnostic performance of a quantitative, immunoturbidimetric D-dimer assay and compare it with other components of the proposed International Society on Thrombosis and Haemostasis disseminated intravascular coagulation (DIC) diagnostic algorithm, we retrospectively analyzed the D-dimer, platelet count, prothrombin time, and fibrinogen results for all eligible hospitalized patients (n = 241) who had a D-dimer assay ordered during a 12-month period. A receiver operating characteristic (ROC) curve constructed from the maximum D-dimer measurement for all patients was significant (P < .001) with an area under the curve (AUC) of 0.94. The ROC curves of the other tests were each significant (P < .001), but the AUCs of the prothrombin time (0.74), fibrinogen level (0.70), and platelet count (0.67) did not approach that of the D-dimer. A D-dimer cutoff of 8.2 µg/mL (8,200 µg/L) optimized sensitivity and negative predictive value for the total population and patients with a predisposing condition. Validation against 286 additional patients in a separate analysis verified the diagnostic performance of the aforementioned cutoff. A sensitive, immunoturbidimetric D-dimer assay, by itself, provides excellent sensitivity and negative predictive value for the diagnosis of DIC.
Disseminated intravascular coagulation (DIC) remains a clinical diagnosis supported by laboratory data but with no universally accepted diagnostic algorithm. The Japanese Ministry of Health and Welfare (JMHW) proposed criteria for the diagnosis of DIC 2 decades ago. The JMHW criteria include semiquantitation of fibrin degradation products as 1 component of the scoring system. The complexity of the algorithm and the current use of D-dimer assays limits the applicability of this scoring system. The International Society on Thrombosis and Haemostasis (ISTH) recently proposed a DIC scoring system based on 4 laboratory parameters and the presence of a predisposing condition. Elevation of a fibrin-related marker, such as D-dimer, represents a key element of the ISTH algorithm, which also scores elevations in the prothrombin time (PT) and decreases in the platelet count and fibrinogen concentration.
Quantitative, rapid D-dimer assays with clinical performance characteristics comparable to conventional enzyme-linked immunosorbent assays have become widely available during the last several years. The immunoturbidimetric D-dimer assays represent a relatively new class of automated D-dimer tests that are based on photo-optical detection of microlatex particle agglutination.
Little is known about the performance of these sensitive D-dimer assays in the context of patient evaluation for suspected DIC. Therefore, we evaluated the analytic and clinical performance of the STA LIATEST (Diagnostica Stago, Parsippany, NJ) immunoturbidimetric D-dimer assay in healthy people, in hospitalized patients not suspected of having DIC, and in patients who have had D-dimer assays ordered for suspected DIC. Because the measurement of D-dimer has not been harmonized among marketed assays, cutoff values for scoring D-dimer elevations in the ISTH algorithm need to be assay-specific. By using receiver operating characteristic (ROC) curve analysis, we identified a prospective cutoff that maximizes sensitivity and specificity of the immunoturbidimetric D-dimer assay. By using this cutoff, we compared the diagnostic performance of the immunoturbidimetric D-dimer assay with the ISTH scoring system.
Abstract and Introduction
Abstract
To evaluate the diagnostic performance of a quantitative, immunoturbidimetric D-dimer assay and compare it with other components of the proposed International Society on Thrombosis and Haemostasis disseminated intravascular coagulation (DIC) diagnostic algorithm, we retrospectively analyzed the D-dimer, platelet count, prothrombin time, and fibrinogen results for all eligible hospitalized patients (n = 241) who had a D-dimer assay ordered during a 12-month period. A receiver operating characteristic (ROC) curve constructed from the maximum D-dimer measurement for all patients was significant (P < .001) with an area under the curve (AUC) of 0.94. The ROC curves of the other tests were each significant (P < .001), but the AUCs of the prothrombin time (0.74), fibrinogen level (0.70), and platelet count (0.67) did not approach that of the D-dimer. A D-dimer cutoff of 8.2 µg/mL (8,200 µg/L) optimized sensitivity and negative predictive value for the total population and patients with a predisposing condition. Validation against 286 additional patients in a separate analysis verified the diagnostic performance of the aforementioned cutoff. A sensitive, immunoturbidimetric D-dimer assay, by itself, provides excellent sensitivity and negative predictive value for the diagnosis of DIC.
Introduction
Disseminated intravascular coagulation (DIC) remains a clinical diagnosis supported by laboratory data but with no universally accepted diagnostic algorithm. The Japanese Ministry of Health and Welfare (JMHW) proposed criteria for the diagnosis of DIC 2 decades ago. The JMHW criteria include semiquantitation of fibrin degradation products as 1 component of the scoring system. The complexity of the algorithm and the current use of D-dimer assays limits the applicability of this scoring system. The International Society on Thrombosis and Haemostasis (ISTH) recently proposed a DIC scoring system based on 4 laboratory parameters and the presence of a predisposing condition. Elevation of a fibrin-related marker, such as D-dimer, represents a key element of the ISTH algorithm, which also scores elevations in the prothrombin time (PT) and decreases in the platelet count and fibrinogen concentration.
Quantitative, rapid D-dimer assays with clinical performance characteristics comparable to conventional enzyme-linked immunosorbent assays have become widely available during the last several years. The immunoturbidimetric D-dimer assays represent a relatively new class of automated D-dimer tests that are based on photo-optical detection of microlatex particle agglutination.
Little is known about the performance of these sensitive D-dimer assays in the context of patient evaluation for suspected DIC. Therefore, we evaluated the analytic and clinical performance of the STA LIATEST (Diagnostica Stago, Parsippany, NJ) immunoturbidimetric D-dimer assay in healthy people, in hospitalized patients not suspected of having DIC, and in patients who have had D-dimer assays ordered for suspected DIC. Because the measurement of D-dimer has not been harmonized among marketed assays, cutoff values for scoring D-dimer elevations in the ISTH algorithm need to be assay-specific. By using receiver operating characteristic (ROC) curve analysis, we identified a prospective cutoff that maximizes sensitivity and specificity of the immunoturbidimetric D-dimer assay. By using this cutoff, we compared the diagnostic performance of the immunoturbidimetric D-dimer assay with the ISTH scoring system.
