Purpose: To evaluate the use of ultrasound (US) for detection of wooden foreign bodies implanted in cadaveric specimens.
Materials and methods: Wooden foreign bodies were randomly placed in the plantar soft tissues of three cadaver feet by using 5-mm-long incisions. Ten foreign bodies were 2.5 x 1.0 mm (length x diameter); 10 were 5.0 x 1.0 mm. Ten incisions were performed without implantation of foreign bodies. Three musculoskeletal radiologists independently performed US in blinded fashion and recorded the presence of a foreign body. Each used 7.5- and 10-MHz linear-array transducers and was informed of the possible presence of a foreign body.
Results: Sensitivity and specificity for detection of 2.5-mm-long foreign bodies were 86.7% and 96.7%, respectively. Sensitivity and specificity for detection of 5.0-mm-long foreign bodies were 93.3% and 96.7%, respectively. Overall sensitivity was 90.0%, with specificity of 96.7%, accuracy of 92.3%, positive predictive value of 98.0%, and negative predictive value of 83.0%.
Conclusion: US can be used effectively to locate wooden foreign bodies as small as 2.5 mm in length. Given that many foreign bodies are radiographically undetectable, the accuracy and availability of US make it an excellent modality for evaluation of radiolucent foreign bodies.