Economic evaluation of posaconazole versus fluconazole prophylaxis in patients with graft-versus-host disease (GVHD) in the Netherlands.

The objective of this study was to evaluate the cost-effectiveness of posaconazole versus fluconazole for the prevention of invasive fungal infections (IFI) in graft-versus-host disease (GVHD) patients in the Netherlands. A decision analytic model was developed based on a double-blind randomized trial that compared posaconazole with fluconazole antifungal prophylaxis in recipients of allogeneic HSCT with GVHD who were receiving immunosuppressive therapy (Ullmann et al., N Engl J Med 356:335-347, 2007). Clinical events were modeled with chance nodes reflecting probabilities of IFIs, IFI-related death, and death from other causes. Data on life expectancy, quality-of-life, medical resource consumption, and costs were obtained from the literature. The total cost with posaconazole amounted to 9,428 (95% uncertainty interval 7,743-11,388), which is 4,566 (2,460-6,854) more than those with fluconazole. Posaconazole prophylaxis resulted in 0.17 (0.02-0.36) quality adjusted life year (QALY) gained compared to fluconazole prophylaxis, corresponding to an incremental cost effectiveness ratio (ICER) of 26,225 per QALY gained. A scenario analysis demonstrated that at an increased background IFI risk (from 9% to 15%) the ICER was 13,462 per QALY. Given the underlying data and assumptions, posaconazole prophylaxis is expected to be cost-effective relative to fluconazole in recipients of allogeneic HSCT developing GVHD in the Netherlands. The cost-effectiveness of posaconazole depends on the IFI risk, which can vary by hospital.

Economic evaluation of posaconazole vs. standard azole prophylaxis in high risk neutropenic patients in the Netherlands.

BACKGROUND: Acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) patients experience prolonged neutropenia after treatment with intensive chemotherapy, leading to a high risk of invasive fungal infections (IFI). The present study evaluates the cost effectiveness of posaconazole vs. standard azoles for the prevention of IFIs in neutropenic patients in the Netherlands. METHODS: A decision-tree model was developed using data from a randomized trial that compared posaconazole and standard azole (fluconazole or itraconazole) prophylaxis in neutropenic patients receiving remission-induction chemotherapy for AML/MDS (Cornely et al., N Engl J Med 2007;356:348-359). Following initiation of prophylaxis, clinical events are modeled with chance nodes reflecting probabilities of IFIs, IFI-related death, and death from other causes. Patients surviving the prophylaxis are assumed to have a life expectancy according to the underlying condition. This allows translation of the trial outcomes to a lifetime horizon. Data on life expectancy, quality of life, medical resource consumption and costs were obtained from the literature. Model outcomes include cost per life year (LY) gained and cost per quality adjusted life year (QALY) gained. RESULTS: The total cost (treatment of breakthrough IFI + prophylaxis) for posaconazole amounted to 4412 euros (95% uncertainty interval 3403 euros - 5666 euros), which is -183 euros (-1985 euros to 1564 euros) less than costs with standard azoles. Posaconazole prophylaxis resulted in 0.08 (0.02-0.15) QALYs gained in comparison with prophylaxis with standard azoles. Results from a probabilistic sensitivity analysis indicate that there is a 90% probability that the cost per QALY gained with posaconazole is below 20,000 euros. Additional scenario analyzes with different assumptions confirmed these findings. CONCLUSION: Given the underlying data and assumptions, the economic evaluation demonstrated that posaconazole prophylaxis is expected to be cost-effective compared with fluconazole/itraconazole in neutropenic AML/MDS patients after intensive chemotherapy.

Severe jaundice, due to vanishing bile duct syndrome, as presenting symptom of Hodgkin's lymphoma, fully reversible after chemotherapy.

Liver involvement in Hodgkin's lymphoma is common and is caused by hepatic infiltration, biliary obstruction by lymphoma, hepatitis, sepsis or complications of chemotherapeutic treatment. Jaundice caused by the vanishing bile duct syndrome related to Hodgkin's lymphoma is very rare. The mechanism is poorly understood but a paraneoplastic effect seems most likely as liver biopsy samples show cholestasis in the absence of lymphoma cells. Despite adequate treatment almost all reported patients died of liver failure or disease progression. Disease progression is explained partly by the difficulties encountered in the administration of potential hepatotoxic chemotherapy in severely cholestatic patients. We describe a 17-year-old man with vanishing bile duct syndrome and Hodgkin's lymphoma who was treated successfully with chemotherapy. The markedly elevated serum bilirubin levels completely normalized. Our case demonstrates that although dosing of chemotherapy in this situation can be very difficult, a good clinical outcome is possible, which makes the attempt at curative treatment worthwhile.

Candida krusei transmission among hematology patients resolved by adapted antifungal prophylaxis and infection control measures.

A sudden increase in neutropenic hematology patients with Candida krusei colonization and bacteremia prompted a longitudinal epidemiological investigation. We identified 39 patients; 13 developed candidemia, and three died; 25 patients carried the same genotype. We intervened by changing antifungal prophylaxis and implementing strict infection control measures. The incidence dropped immediately.

Strict infection control measures do not prevent clonal spread of coagulase negative staphylococci colonizing central venous catheters in neutropenic hemato-oncologic patients.

Coagulase negative staphylococci (CoNS) are a main cause of catheter related infections (CRI). Earlier studies (1994-1996) revealed a high incidence of CRI (6 per 1000 catheter days) among neutropenic hemato-oncologic patients in the Erasmus MC Hematology Department (Rotterdam, The Netherlands). This was mainly explained by expansion of two methicillin resistant Staphylococcus epidermidis (MRSE) clones (Nouwen et al., J. Clin. Microbiol. 36 (1998) 2696-2702). In a new, 16-bed unit in the same institution, we investigated the effect of strict clinical isolation measures on the incidence of CRI. During two 6-month screening periods (period I: April 1998-December 1998 and period II: April 1999-October 1999) all patients receiving a central venous catheter were prospectively monitored for the development of CRI. During period I every visitor of the cubicles had to wear hair caps, masks, gowns and gloves. During period II these procedures were abolished, but hands were cleansed using alcohol and masks were worn during both periods in case of coughing and sneezing. All CoNS strains isolated from blood cultures were genetically classifies by pulsed field gel electrophoresis (PFGE). The incidence of CRI during period I was 13.0 per 1000 catheter days, in comparison to 9.6 in period II (P=0.84). During this latter period, 19 CRI were diagnosed, 14 catheter related bacteremia episodes (CRB) and five local infections. Seventy-two percent (n=9) of CRB were due to a CoNS. The mean catheter survival until appearance of a CRI increased from 43 days during period I to 78 days in period II (P=0.39). The mean catheter survival until infection related removal was increased from 43 days to 133 days (P=0.12). During period I less experienced intervention radiologists introduced the catheters, which may have limited the efficacy of the strict hygiene measures. Thus, abolishing strict isolation precautions had no negative effect on the incidence of CRI. After genotyping of 38 MRSE strains isolated from blood and central venous catheter cultures of 12 patients in period II, eight PFGE types were found. Three types were found in more than one patient, but based on epidemiological data patient-to-patient spread could not be proven. No genotypic identity between patient and personnel CoNS isolates was shown and the two major clonal types that were present between 1994 and 1996 were not encountered. However, from December 1998 onwards new MRSE clones could be identified (types E and J). In conclusion, despite a constant rate of CRI and implementation of optimal patient care, clonal spread of MRSE strains was not prevented by strict hygiene measures.

Galactomannan detection in computerized tomography-based broncho-alveolar lavage fluid and serum in haematological patients at risk for invasive pulmonary aspergillosis.

We determined the value of galactomannan (GM) detection in computerized tomography (CT)-based broncho-alveolar lavage (BAL) fluid and serum for the diagnosis of invasive pulmonary aspergillosis (IPA) in haemato-oncological patients with neutropenia. CT of the thorax and BAL were performed systematically at predefined clinical indications. GM was determined by sandwich enzyme-linked immunosorbent assay; the clinicians were unaware of the results. Of 160 patients, 17 patients (10.6%) presented with proven, probable or suspected IPA. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of GM detection in CT-based BAL fluid were all 100%. For GM detection in serially sampled serum, the sensitivity was 47%, the specificity 93%, the PPV 73% and the NPV 82%. A non-blinded follow-up study was performed to validate these results. In this study, 22 of 198 patients (11.1%) presented with IPA, and the sensitivity, specificity, PPV and NPV of GM detection in CT-based BAL fluid were 85%, 100%, 100% and 88% respectively. None of BAL fluids obtained after antifungal treatment of 3 d or more were positive. These results indicate that, when CT is used systematically and at an early stage, GM detection in CT-based BAL fluid has a high PPV for diagnosing IPA early in untreated patients.