Review of the rebound phenomenon in new anticoagulant treatments.

BACKGROUND: The launch of new anticoagulant treatments has sparked debate as to the optimal drug to use for primary prevention of venous thromboembolism (VTE) in major orthopaedic surgery, for the treatment of VTE, and for stroke prevention in atrial fibrillation, taking into account both efficacy and tolerability. Newer drugs such as fondaparinux and ximelagatran have shown improvements in short-term efficacy or convenience, but such effects may be offset by reduced tolerability. They also raise the question of a possible delayed increase in thromboembolic events after these drugs are stopped, the so-called 'rebound effect'. OBJECTIVE: To review pharmacological and clinical data describing the rebound effect associated with new and standard anticoagulant drugs. RESEARCH DESIGN AND METHODS: Computerized searches, covering the period 1960-2005 for the historical background and physiopathology review, and from 1999 to 2005 for the clinical trial analysis, were performed on BIOSIS, PubMed and clinicaltrials.gov databases. The terms 'rebound', 'anticoagulant', and 'heparin' were used. Only articles written in English were reviewed. Articles with drug interactions from other therapeutic classes or types of vascular surgery and commentary articles were excluded. RESULTS: Available data in relation to a possible rebound phenomenon following cessation of active treatment are very limited. Results coming mainly from orthopaedic surgery trials suggest an increased rate of venous or arterial thromboembolic events with newer anticoagulants, compared with standard anticoagulant therapy. An increase in the rate of serious arterial adverse events has, for example, been observed in VTE patients treated with ximelagatran relative to those receiving warfarin/placebo (short-term exposure: 0.75% vs 0.26%, p < 0.05; long-term exposure: 1.70% vs 0.70%, p