The percutaneous implantation of vena caval filters has gained acceptance as a means of preventing pulmonary embolism in patients with contraindications to or inefficiency of oral anticoagulants. In the past years, we predominantly used the Guenther filter. Its stainless steel wires are formed into a helical basket and anchoring struts.1 Our experience with 62 patients showed that this device prevents pulmonary embolism effectively. However, cranial and caudal migration of the filter, perforation of the caval wall by the struts, and breakage of the struts have been reported. In 39% of cases, thromboemboli were found inside the filter by computed tomography, but computed tomographic imaging does not allow differentiation between trapped emboli and genuine thrombi.2. Intravascular ultrasound has demonstrated reliable implementation in the diagnosis of peripheral arterial diseases.3,4 McCowan et al5 examined 6 different vena caval filters in vitro and in vivo, and found this method to be superior to both external ultrasound and cavography in detecting thrombi in the filter or vena cava. The present study was performed to assess the implementation of intravascular ultrasound imaging of Guenther basket filters, while paying particular attention to both the detection of trapped or genuine thromboemboli, and the diagnosis of filter perforation and breakage of struts. © 1992.