Predictive value of D-dimer assay in superficial thrombophlebitis of the lower limbs

Objective

The negative predictive value of D-dimer (DD) assay in patients with venous thromboembolic disease is well established for deep vein thrombosis and pulmonary embolism. Little is known about the value of DD assay in patients with superficial thrombophlebitis (ST). The purpose of this study was to assess the value of DD assay in patients with ST of the lower limb.

Method

The study group was composed of 100 consecutive patients, irrespective of age. Patients with clinical manifestations suggestive of ST of the lower limbs with positive duplex color Doppler evidence confirming the diagnosis and DD assay results (Vidas D-Dimer Exclusion) within 24 hours were included in the study. Patients with thrombosis in another site in addition to the superficial vein of the lower limb, those taking anticoagulants for more than 48 hours, and those with a condition known to potentially elevate DD levels were excluded. The volume of the thrombus was determined echographically and reported as mean diameter and length.

Results

Sixty-two women and 38 men were included. Mean age (± 5) was 58 years ± 13.48 (range 18–90; median: 57). The ST involved the Great saphenous (n = 74), the small saphenous (n = 11) or another vein (n = 15). Mean thrombus volume was 4453 mm³ ± 7101 (range 94–38484; median: 1751). Mean DD level was 829 ng/ml ± 516.72 (range 100–2567; median: 715.5). DD assay was negative (< 500 ng/ml) in 32 patients (32%) and positive in 68 (68%). For these three items, there was no significant difference between ST with and without varicose veins. DD assay was always positive (≥ 500 ng/ml) in all patients aged over 70 years (n = 22). In patients aged less than 70 years (n = 78), DD assay was positive in 46 (59%) and negative in 32 (41%). DD level was positively correlated with thrombus volume in patients aged less than 70 years (P < 0.0001). ROC analysis, sensitivity as a function of specificity by thrombus volume for the entire population, determined the usefulness of a negative DD assay. Considering the critical threshold at 5914 mm³, sensitivity was 1.0 (95CI 0.89–1.0), with 0.29 specificity (95CI 0.19–0.42), 1.00 negative predictive value and 0.75 positive predictive value. However, the thrombus volume was less than this threshold value in three of the nine cases of ST with extension to the terminal portion of the saphenous.

Conclusion

A positive DD assay was observed in 68% of patients with ST, with no significant difference with or without varicose veins. The test was positive in all patients aged over 70 years and in 59% of those aged under 70 years. There was a correlation between DD level and thrombus volume, yielding a threshold volume (5914 m³) above which all DD tests were positive. Nevertheless, this threshold volume was too great to include all ST extending to the terminal portion of the saphenous. Measurement of DD level is thus not contributive to the diagnosis of ST.