Voriconazole Replacing Liposomal Amphotericin B as First-Line Therapy in Suspected or Proven Fungal Infection in Acute Leukemia Patients: A Retrospective Audit of Clinical and Financial Outcomes in a UK District Hospital.

Background: In most centres in the United Kingdom, systemic antifungal therapy (AFT) is used as third-line therapy for fever complicating profound, prolonged neutropenia (PPN) during the treatment of acute leukemia. Voriconazole has been recommended as first-line AFT at the Kent and Canterbury Hospital (KCH) since October 2002; liposomal amphotericin B was the previous treatment of choice. The aim of the audit was to identify numbers of episodes of PPN and the numbers of suspected fungal infections in a 2-year period following the policy change. Subsequently the clinical and financial outcomes were examined. In addition, the impact on cost of AFT following centralisation of leukemia treatment from two district hospitals onto one site was examined.

Methods: A retrospective audit was conducted on data from hematology inpatients undergoing remission induction or consolidation therapy for acute leukemia at KCH between January 2003 and December 2004. The costs of voriconazole and liposomal amphotericin B treatment from 2002 to 2004, and 8 months prior and post centralisation of inpatient care (April 2004, which increased the population from 400,000 to 600,000), were examined.

Results: 84 episodes of PPN were identified in 41 patients undergoing treatment for acute leukemia; mostly acute myeloid leukemia (AML). Itraconazole prophylaxis from d1 of therapy and GCSF from d+5 was used in the majority of cases. 18 cases of suspected or radiologically proven fungal infection were identified. High-resolution computed tomography of the chest was performed in 10 cases and suspicious lesions identified in three. Voriconazole was used as first-line therapy in 17/18 cases. In 7 cases, treatment was switched to liposomal amphotericin B. Reasons for switching were rising C-reactive protein (1 patient), persistent fever (2 patients), radiological progression (1 patient) and side effects (3 patients). Of the 3 patients with radiological evidence of fungal infection, two had a complete resolution (1 voriconazole, 1 voriconazole/liposomal amphotericin B) and 1 patient died of refractory leukemia. There was a fall in total antifungal spend from £263K in 2002/3 to £229K in 2003 and a further 68% fall to £73K in 2004. We suspect this was due to increasing adherence to the new antifungal protocol and to improved practices following centralisation: in the 8 months pre-centralisation the antifungal spend across all hospitals was £102K falling by 74% to £26K in the 8 months on the single site.

Conclusion: Since introducing voriconazole as first-line AFT, centralising inpatient services, and adopting common policies for antimicrobial prophylaxis, there has been considerable financial benefit with no increase in morbidity and/or mortality.