BACKGROUND: Recombinant factor VIIa (rFVIIa) is used for many off-label indications without high quality evidence to support its efficacy. The aim of this study was to determine indications for use of off-label rFVIIa, efficacy and safety, and adherence to institutional guidelines. METHODS: We performed a retrospective review of off-label rFVIIa at two tertiary hospitals from 2007 to 2010. RESULTS: One hundred forty-five administrations were identified and analysed. Haemorrhage associated with cardiac surgery made up one-third of all rFVIIa usage, with trauma (20%) and other surgery (11%) the next most frequent indications. Compared with all others, cardiac surgery patients were older (60.0 years versus 47.4 years, P < 0.001) and had lower pre-rFVIIa transfusion requirements, a higher subjective response rate (88% versus 46%, P < 0.001) and lower mortality rates (6.1% versus 33%, P < 0.001), but higher rates of arterial thrombormbolic events (16.7% versus 2.1%, P = 0.002). Most patients received only one or two doses (n = 137; 95%), with no subject receiving a third or subsequent dose having an appreciable reduction in bleeding. Only a small number of patients (n = 15; 10.3%) had rFVIIa administered in accordance with our institutions' guidelines. CONCLUSION: Patients administered rFVIIa for haemorrhage not associated with cardiac surgery were severely unwell. Despite lack of evidence, administration of rFVIIa may be justified by the high mortality rate, but more than two doses are unlikely to provide further benefit. The high rate of arterial thromboembolism in cardiac surgical patients raises risk-benefit considerations. Adherence to our institutions' guidelines was poor.
BACKGROUND: Recombinant factor VIIa (rFVIIa) is used for many off-label indications without high quality evidence to support its efficacy. The aim of this study was to determine indications for use of off-label rFVIIa, efficacy and safety, and adherence to institutional guidelines. METHODS: We performed a retrospective review of off-label rFVIIa at two tertiary hospitals from 2007 to 2010. RESULTS: One hundred forty-five administrations were identified and analysed. Haemorrhage associated with cardiac surgery made up one-third of all rFVIIa usage, with trauma (20%) and other surgery (11%) the next most frequent indications. Compared with all others, cardiac surgery patients were older (60.0 years versus 47.4 years, P < 0.001) and had lower pre-rFVIIa transfusion requirements, a higher subjective response rate (88% versus 46%, P < 0.001) and lower mortality rates (6.1% versus 33%, P < 0.001), but higher rates of arterial thrombormbolic events (16.7% versus 2.1%, P = 0.002). Most patients received only one or two doses (n = 137; 95%), with no subject receiving a third or subsequent dose having an appreciable reduction in bleeding. Only a small number of patients (n = 15; 10.3%) had rFVIIa administered in accordance with our institutions' guidelines. CONCLUSION:Patients administered rFVIIa for haemorrhage not associated with cardiac surgery were severely unwell. Despite lack of evidence, administration of rFVIIa may be justified by the high mortality rate, but more than two doses are unlikely to provide further benefit. The high rate of arterial thromboembolism in cardiac surgical patients raises risk-benefit considerations. Adherence to our institutions' guidelines was poor.