PURPOSE: To report clinical experience with retrievable Gunther Tulip filters from implantation to retrieval and their status in nonretrieved situations. METHODS: Seventy-five Gunther Tulip filter implantations were performed in 71 patients (43 women; mean age 55 years). Indications for filter placement were pulmonary embolism (PE) or iliofemoral deep vein thrombosis (DVT) in patients with a contraindication to anticoagulation (43, 61%) or perioperative PE prophylaxis (28, 39%) in patients with confirmed iliofemoral DVT. Retrieval procedures were planned for each patient. Patients with nonretrieved filters were followed with plain radiography and duplex sonography. RESULTS: Technical success of filter insertion was 97.3% (73/75). Eighteen (25%) patients died from unrelated causes prior to retrieval attempts, and 6 other patients were too critically ill for a retrieval procedure. Of 49 (67%) planned retrieval attempts, 14 (19%) filters could not be removed owing to large trapped thrombi. The mean implantation period for the 35 (48%) retrieved filters was 8.2 days (range 1-13). Delivery tilt was observed in 12 (16%) filters and during retrieval attempts in 1 more case. For 9 nonretrieved filters, tilt and migration were observed in 22% at a mean follow-up of 30 months, but no venous thrombosis was assessed. CONCLUSIONS: Our data confirm the clinical efficacy of the Gunther Tulip filter during implantation and the feasibility of its retrieval. Further long-term follow-up should be conducted on nonretrieved filters to confirm our results.