Comparison of three techniques for evaluation of de novo asymptomatic pulmonary arterial thrombosis following deep vein thrombosis in total knee arthroplasty

Purpose: Deep venous thrombosis (DVT) following total knee arthroplasty (TKA) frequently results in pulmonary arterial thrombosis (PAT). Using multislice-CT (MSCT), we evaluated the incidence of de novo asymptomatic PAT following DVT in subjects undergoing TKA and compared these results by ventilation-perfusion (VP) lung scintigram and ultrasonography of the lower extremities.

Materials and methods: Twenty-five asymptomatic subjects (5 males, 53-82 years old, receiving anticoagulant therapy) underwent enhanced MSCT, VP lung scintigram, and ultrasonography of the lower extremities before and 7 days after TKA.

Results: Among 25 subjects, 2 were found to have asymptomatic PAT and DVT, respectively, by pre-TKA MSCT. In the remaining 23 subjects, the post-TKA MSCT detected PAT in 7 subjects (30.4%) and DVT in 6 (26.1%). Those 16 subjects without evidence of PAT on the post-TKA MSCT also had no VP mismatches on their VP lung scintigrams. Additionally, all subjects without evidence of DVT on the post-TKA MSCT also showed no evidence of DVT by ultrasonography. Of the 7 subjects in whom PAT was detected by post-TKA MSCT, VP mismatch was shown by lung scintigraphy in 2 and DVT was confirmed by MSCT in both. No patient had VP mismatch without PAT, and 2 subjects with PAT and VP mismatch had DVT. Conversely, DVT was not revealed by ultrasonography among the 6 subjects (26.1%) in whom DVT was detected in the lower extremities by post-TKA MSCT. However, blood flow abnormalities were shown by ultrasonography of the lower extremities in 4 of the 6 subjects (17.4%).

Conclusion: MSCT was the most sensitive of the 3 methods and could simultaneously evaluate PAT and DVT. The apparent discrepancy between these techniques may be due to MSCT's ability to detect thrombi that do not interfere with blood flow.