Abstract 1: Impact of Predicted Bleeding Risk on the Cost-Effectiveness of Direct Thrombin Inhibition vs. Heparin Monotherapy for Patients Undergoing PCI
Introduction Direct thrombin inhibitors (DTI) reduce bleeding and are cost-saving in patients (pts) undergoing PCI when compared to unfractionated heparin (UFH) plus routine glycoprotein IIb/IIIa inhibition (GPI). Little is known about the cost-effectiveness of DTI alone vs. UFH alone, however. Methods We combined data on patient-level bleeding risk with several external data sources to estimate the economic impact of substituting DTI for UFH among unselected PCI patients. We used a validated model to predict the risk of major bleeding for 81,628 NCDR® CathPCI Registry® patients who received UFH only in 2004-06. DTI costs were estimated based on current FDA dosing guidelines and acquisition costs. The cost of major bleeding ($8722) was estimated based on data from the Mid America Heart Institute. The benefit of DTI vs. UFH alone was based on bleeding risk reduction from ISAR-REACT 3. A Markov model based on published data was used to estimate the loss in life expectancy associated with major bleeding. Results The overall rate of major bleeding in the reference population was 2.2%, the estimated cost of a major bleed was $8722, projected DTI costs were $641/pt and heparin costs were negligible ($4). Assuming that DTI use reduces the risk of bleeding by 33%, use of DTI for all patients was estimated to increase costs by $573/patient and was cost saving only for patients with a predicted risk of major bleeding >20% (0.16% of the overall population). When life years lost due to bleeding were included, at willingness-to-pay thresholds of $50K and $100K per life year gained, DTI was cost effective for patients with a bleeding risk ≥8% (2.5% patients) and ≥5% (7.9% of patients), respectively. Conclusions For patients undergoing PCI, substitution of DTI for UFH monotherapy is projected to increase costs for virtually all patients, but may be reasonably cost-effective for patients with a high predicted risk (≥5%) of bleeding.