D-dimer measurements unhelpful for ruling in DIC

Physicians often use D-dimer assay results to corroborate the diagnosis of disseminated intravascular coagulation (DIC). We determined whether the results of this nonspecific assay helped to diagnose DIC in a group of patients admitted to the intensive care unit (a setting in which patients are often evaluated for this condition). We defined DIC in accordance with the following findings: systemic bleeding, prothrombin time (PT) more than 3 seconds above the reference range (or both); a fibrinogen level less than 150 mg/dL (1.5 g/L); and a platelet count less than 100 x 1O³/μL (100 X 10⁹/L). For each patient, we calculated the Acute Physiological and Chronic Health Evaluation (APACHE) II score for illness and measured the PT, fibrinogen level, platelet count, and (semiquantitative) D-dimer level. The D-dimer result was positive in 31 of 50 patients (62%); for all 50 patients, the PT, fibrinogen level, and platelet count were within the reference ranges. None of these patients had clinical or laboratory evidence of DIC. We found no correlation between D-dimer results and (1) clinically observed deep venous thrombosis or pulmonary embolus; and (2) PT, level of fibrinogen, or platelet count. The mean ± SD APACHE II scores of D-dimer-negative patients were lower than those of the D-dimer-positive patients (13.4 ± 1.4 vs 16.2 ± 1.1, P = .006) but did not correlate with the degree of D-dimer positivity. The mean ± SD fibrinogen level of the D-dimer-negative group (456 ± 295 mg/dL [4.6 ± 3.0 g/L]) was not significantly different from that of the D-dimer-positive group (393 ± 187 mg/dL [3.9 ± 1.9 g/L], P = .36). We conclude that D-dimer levels are elevated in most critically ill patients irrespective of their hemostatic or thrombotic symptoms. The D-dimer assay should not be used to diagnose DIC in this patient population because it adds no specificity; it should be used only to rule out DIC.