JAC Advance Access originally published online on June 7, 2007
Journal of Antimicrobial Chemotherapy 2007 60(2):385-393; doi:10.1093/jac/dkm196
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Economic evaluation of targeted treatments of invasive aspergillosis in adult haematopoietic stem cell transplant recipients in the Netherlands: a modelling approach
1 Department of Health Organization Policy and Economics (HOPE), Faculty of Health Sciences, Maastricht University, PO Box 616, 6200 MD Maastricht, The Netherlands 2 Department of Medical Microbiology, University Medical Center Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands 3 Department of Pediatrics, University Medical Center Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands 4 Department of Internal Medicine, University Medical Center Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands 5 Department of Haematology, University Medical Center Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands 6 Department of General Internal Medicine, Bronovo Hospital, PO Box 96900, 2509 JH The Hague, The Netherlands 7 Department of Clinical Epidemiology and Medical Technology Assessment, University Medical Center Maastricht, PO Box 5800, 6202 AZ Maastricht, The Netherlands
* Corresponding author. Tel: +31-43-38-81723/81727; Fax: +31-43-36-70-960; E-mail: a.ament{at}beoz.unimaas.nl
Received 30 October 2006; returned 10 December 2006; revised 24 April 2007; accepted 6 May 2007
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
Objectives: The aim of this study was to assess the cost-effectiveness of a targeted treatment model of antifungal treatment strategies for adult haematopoietic stem cell transplant (HSCT) recipients in the Netherlands from a hospital perspective, using a decision analytic modelling approach.
Methods: The economic evaluation of desoxycholate amphotericin B, liposomal amphotericin B, voriconazole and caspofungin was undertaken. These drugs could be used alone, in various combinations or sequentially. In our model, first-line therapy consisted of either voriconazole or liposomal amphotericin B. If necessary, treatment was switched to a second-line treatment, including combination antifungal therapy. The theoretical population in this model consisted of adult HSCT recipients with proven or probable invasive aspergillosis (IA). Long-term survival was extrapolated from survival after 12 weeks of treatment and life expectancy.
Results: First-line antifungal treatment strategies with voriconazole were both more effective and less costly over first-line strategies employing liposomal amphotericin B at a dosage of 4 mg/kg/day. The strategy of voriconazole followed by caspofungin (voriconazole/caspofungin) was dominant over the strategies of voriconazole followed by liposomal amphotericin B (voriconazole/liposomal amphotericin B) or desoxycholate amphotericin B (voriconazole/desoxycholate amphotericin B). However, the voriconazole followed by the combination of liposomal amphotericin B and caspofungin strategy (voriconazole/liposomal amphotericin B + caspofungin) was more effective though more expensive than the voriconazole/caspofungin strategy resulting in an incremental cost-effectiveness ratio (ICER) of about 
107 000 for a life-year saved. At a dosage of 1 mg/kg/day of liposomal amphotericin B, the voriconazole/caspofungin strategy was more effective but more costly than the voriconazole/desoxycholate amphotericin B strategy with an ICER of 10 000 for each extra life-year saved. Between the voriconazole/liposomal amphotericin B + caspofungin and the voriconazole/caspofungin strategies, the ICER was 40 000.
Conclusions: Probabilistic analyses on net monetary benefit showed that the voriconazole/caspofungin strategy had the highest probability of being the most cost-effective strategy.
Keywords: antifungal treatment , cost-effectiveness , probabilistic sensitivity analysis
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
The neutropenic patient population is increasing due to the treatment of life-threatening diseases with newer, aggressive chemotherapeutic regimens and the use of haematopoietic stem cell and organ transplantation. During neutropenia induced by chemotherapy, the number of neutrophils, one of the primary host defences, is undetectably low rendering patients at high risk of developing infectious complications. Infection caused by the mould Aspergillus spp. is one of the most serious, because it is difficult to diagnose and it is associated with a high mortality despite adequate therapy.1–4 Acute invasive aspergillosis (IA) is often a rapidly progressive infection, and its refractoriness to therapy is due, in part, to the organism's rapid growth rate and to its tendency to invade blood vessels.5,6 The risk factors for the development of IA include prolonged neutropenia, the severity of the immunosuppression either by the disease or therapy, corticosteroid use, graft-versus-host-disease and the degree of exposure to Aspergillus spores.7–9
The incidence rate of IA in severely immunocompromised patients is between 3% and 30% in allogeneic haematopoietic stem cell transplant (HSCT) recipients.2,10–13 Wald et al.2 reported that the incidence rate of IA among these patients increased from 6% in 1987 to 11% in 1993.
The mortality of patients with IA is primarily determined by the underlying disease and its remission.3,7 Despite antifungal treatment, the mortality rate among allogeneic HSCT recipients is more than 85%.3,14,15
The average length of survival of HSCT recipients without IA was 2.3 years (Belgian National Cancer Institute). In the study of Wald et al.2 45% of the HSCT recipients without IA died within a year of transplant, while only 5% survived when the post-transplant period was complicated by IA. Management of IA ranges from antifungal prophylaxis, empirical treatment to treatment of confirmed infections.16 Since diagnosis of IA at an early stage of the infection is difficult, it has become standard clinical practice to start antifungal treatment empirically for those neutropenic patients with persistent fever who fail to respond to treatment with broad-spectrum antibacterial agents.17 Amphotericin B desoxycholate (D-AMB) has been the ‘gold standard’ for treating IA for over 40 years. Lipid formulations of amphotericin B (L-AMB) have been shown to be considerably less toxic than D-AMB and appear to be equally effective18 but their use is restrained by their high costs. Recently, two new antifungal agents voriconazole and caspofungin have become available offering more choice and making it possible to formulate different treatment strategies, including those involving combinations of drugs.9,19,20 In a recent study, voriconazole has been shown to be less expensive than D-AMB over a wide range of values for both unit costs and resource utilization.21
Modern diagnostic tests make it possible to consider restricting antifungal treatment to patients with proven or probable IA.22 Earlier diagnosis and therapy and an early switch to other antifungal drugs when there treatment is failing are likely to improve clinical outcomes of high risk patients, such as recipients of an HSCT and those with unchecked underlying disease.6,7,23–27
The aim of this study was to assess the cost-effectiveness of targeted antifungal treatment strategies for the Netherlands taking into account a switch to alternative antifungal drugs if needed.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
The panel of experts
All data and assumptions were agreed by consensus of the authors who represented Dutch University hospitals and met on three separate occasions during 2003 and 2004. The model was presented to the expert committee twice. The structure of the model was discussed in the first meeting. Consensus was then obtained for treatment strategies, the time from starting treatment to switching to an alternative drug after failure, serious side effects or other reasons, the duration of full-time treatment, and for the baseline values of the probabilities of switching treatment in a second meeting. Finally, the model outcomes and analyses were presented to the experts and their recommendations were incorporated into the model.
A decision tree was constructed which compared costs and effectiveness of alternative antifungal-targeted strategies. The theoretical population in the model consisted of adult HSCT recipients with proven or probable IA. The time horizon of the economic evaluation was 12 weeks because the data of probabilities used in the model were derived from treatment studies with a 12 week follow-up. The actual risk for developing IA is limited to this period but the time horizon does not mean that patients are treated that long. Treatment cost was restricted to the period of first- or second-line or a switch to other licensed antifungal therapy (OLAT) within 7 weeks of starting therapy (see further description costs and Figure 2). Effectiveness was expressed as patient survival after 12 weeks of initiating antifungal therapy. Life-years saved were calculated by multiplying survival probability with a life expectancy of 2.3 years for HSCT recipients who were successfully treated after 12 weeks of initiating IA treatment.2 Costs were quantified from the hospital perspective.28
The model allowed the antifungal drugs D-AMB, L-AMB, voriconazole and caspofungin to be used alone, in combination or sequentially.
In our model (Figure 1), treatment started with either voriconazole or L-AMB. D-AMB was excluded as first-line treatment based on the experts opinion that less toxic alternatives will take the first-line place of D-AMB in clinical practice. However, voriconazole could also be followed by D-AMB, if necessary (voriconazole/D-AMB). Caspofungin was excluded as first-line treatment as it is not licensed for this indication and is not commonly used in this way. If necessary, treatment could be switched to an alternative or to a combination of drugs. Second-line therapy could be followed by a full course of OLAT. OLAT consisted of the remaining of the four above-mentioned antifungal drugs for which an equal proportion of the two remaining drugs was assumed. Consequently, seven treatment strategies were evaluated. Four were based on voriconazole as first-line treatment: voriconazole/D-AMB, voriconazole followed by L-AMB (voriconazole/L-AMB), voriconazole followed by caspofungin (voriconazole/caspofungin), voriconazole followed by a combination of L-AMB and caspofungin (voriconazole/L-AMB + caspofungin) and three were based on L-AMB as first-line treatment: L-AMB followed by voriconazole (L-AMB/voriconazole), or caspofungin (L-AMB/caspofungin), or the combination of voriconazole and caspofungin (L-AMB/voriconazole + caspofungin).
|
|
It was assumed that first- and second-line antifungal treatment would be discontinued early because of treatment failure, early death due to IA, early death due to underlying disease, serious side effects, or other reasons. Treatment failure was defined by progressive aspergillosis. First-line treatment failure resulted in an early switch to second-line treatment or to early IA death. Second-line treatment failure resulted in a switch to OLAT or to early IA death. For each first-line drug therapy, a limited number of serious side effects of treatment were included in the decision tree. Only nephrotoxicity and hypokalaemia were assumed to lead to a switch in antifungal treatment. A pathway for early HSCT death due to the underlying disease was inserted into the decision tree to take into account the lower drug consumption of patients who died early after starting antifungal treatment.
To perform the necessary calculations, each pathway in the model had to be defined by parameters such as the probabilities of an event to occur and the costs of clinical activities. All model parameters were based on a thorough review of the literature. Studies involving only adult HSCT recipients with proven or probable IA are scarce. Therefore, the results of studies involving more mixed populations (adults treated with chemotherapy for haematological malignancies) who had proven or probable IA were included.
When no data were available, or when data were not sufficiently reliable, parameter estimates were agreed by a consensus.
Probabilities. MEDLINE was used to retrieve relevant publications using the search criteria ‘invasive aspergillosis’, ‘drug therapy’ and ‘mortality’. In addition, articles were identified from the bibliographies of the papers retrieved. Published abstracts of economic evaluations of IA treatment strategies were also included. Articles and abstracts that reported the following criteria were then selected: (i) treatment of proven or probable IA among adults with haematological malignancies with voriconazole, L-AMB, D-AMB or caspofungin; (ii) the number of patients whose therapy was switched to alternative treatment because of failure, serious side effects or other reasons; (iii) the number of patients who died, as well as the time and cause of death.
Costs. The costs were calculated in euros and were assumed to have been incurred until death or discharge from the hospital up to a maximum of 12 weeks.
The costs of treatment were based on the cost per day and the duration of the treatment. Dosages were assessed according to the prescription in the articles of the literature review.29 The costs of the four antifungal drugs used in the model were based on Dutch national retail prices for 2005 and are shown in Table 1. Further details about the drug cost calculations are available on request.
|
Antifungal treatment could consist of a short, medium or full-course of first- and second-line treatment according to when a switch in therapy was made. The duration of OLAT was defined as a full-course treatment. We defined 5 days as being a short duration of treatment, 7 days as being a medium duration and 20 days as being the duration of a full-course of treatment. Treatment duration could extend to 34 days when treatment was switched twice because of serious side effects or other reasons (Figure 2).
Nephrotoxicity and hypokalaemia due to treatment with the initial antifungal drugs were assumed to increase the length of stay in the hospital by an extra 7 days at an estimated daily cost to a university hospital in the Netherlands of 500.30 The costs of serious side effects to second-line antifungal drugs were set at 500 for voriconazole, 0 for caspofungin, 1500 for L-AMB and 3000 for D-AMB according to the type of serious side effects caused by the drug.
Model validation and verification
The model was presented to the panel twice. The structure of the model was discussed in the first meeting. Consensus about all parameter values was obtained in the second meeting. TreeAge Pro 2004 (TreeAge Software, Williamstown, MA, USA) was used to structure the decision model and to perform the calculations.
Baseline analysis.
The strategy yielding the lowest cost was considered to be the reference strategy. If strategies were both more effective and more costly, the relationship between costs and effectiveness [cost-effectiveness (CE)] was expressed as an incremental cost-effectiveness ratio (ICER).31 The net monetary benefit (NMB) for each strategy was calculated based on willingness to pay and a value for a life-year saved of 20 000.32
Sensitivity analysis. Extensive sensitivity analyses were performed using both probability and cost estimate ranges to determine the impact, i.e. potential effect of the variables on the optimal choice of reference strategy.33 The ranges of probabilities were determined by calculating the 95% confidence intervals when data were available. Other baseline values that influenced costs were varied as shown in Table 2.
|
Thresholds for CE and NMB, i.e. values of the variables at which the optimal path changed, will be indicated where appropriate.
Multi-variable model outcome uncertainty was further tested using probabilistic sensitivity analysis.34 Cost-effectiveness acceptability curves were constructed for the seven treatment strategies.35
Because this is a model approach no patients are involved.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
Literature review
Eleven articles met our search criteria. An overview of the reported data for a switch in treatment because of failure, side effects or due to other reasons is presented in Table 3.
|
The Kaplan–Meier curves in the Seattle study2 showed that 25% of 2282 HSCT recipients without diagnosed IA died within the first 3 months of transplantation and 17% in the next three months. Combining the survival curve of patients without IA with the time between transplantation and the diagnosis of IA, we calculated a probability of 0.20 for transplant-related mortality 3 months after IA diagnosis.
The probability of transplant-related mortality was higher in the first 14 days (mean 7 days) after transplantation, so we set the probability for early HSCT death at 0.06, based on the results of Wald et al.2
In the study of Denning et al.36 32 of 36 patients who failed to respond to voriconazole treatment died. Kontoyiannis et al.9 excluded patients who died in the first 7 days of treatment and reported 17 of the 28 patients who failed treatment died. To compensate for leaving out the early deaths, we augmented their reported data to 27 deaths in 38 patients with treatment failure. Hence, the calculated probability of dying after treatment failure was estimated to be 59/74 = 0.80.
Pooled data from reported trials were calculated and used as baseline values for one-way sensitivity analysis and probabilistic analysis in the decision tree.
The expected costs and effectiveness of the seven antifungal treatment strategies are depicted in Table 4. Total patient costs ranged from 6631 for the voriconazole/caspofungin strategy to 17 542 for the L-AMB/voriconazole + caspofungin strategy. Survival probability ranged from 0.504 for the L-AMB/caspofungin strategy to 0.594 for the voriconazole/L-AMB + caspofungin strategy, corresponding to 1.109–1.307 life-years saved for patients who followed these treatment strategies.
|
First-line antifungal treatment strategies with voriconazole dominated over strategies employing L-AMB as first-line treatment (Figure 3).
|
The strategy voriconazole/caspofungin was dominant over voriconazole/L-AMB and voriconazole/D-AMB. However, the voriconazole/L-AMB + caspofungin strategy was more effective though more expensive than the voriconazole/caspofungin strategy resulting in an ICER of
107 036 for each extra life-year saved. Multiple one-way sensitivity analyses showed that two variables exerted an important impact, namely the number of vials of L-AMB per day and the number of days of OLAT treatment. In both cases, the cheapest strategy changed from the voriconazole/caspofungin strategy to the voriconazole/D-AMB strategy: when the number of vials of L-AMB was <3.9 vials per day (baseline value 6) or the length of OLAT treatment was <15.6 days (baseline value 20).
At a daily dose of 1 mg/kg for L-AMB, the voriconazole/D-AMB strategy had lowest costs. The voriconazole/caspofungin strategy and the voriconazole/L-AMB + caspofungin strategy were more effective but more costly with an ICER of 10 468 and 37 597, respectively, for each extra life-year saved.
For 4 mg/kg/day, there were no thresholds for baseline parameters of NMB with a willingness to pay of 20 000. The voriconazole/caspofungin strategy was the preferred strategy. At a dosage of 1 mg/kg/day of L-AMB/day, the voriconazole/caspofungin strategy remained the preferred strategy and the sensitivity analyses indicated four thresholds. Three times the voriconazole/D-AMB strategy became the preferred strategy. The voriconazole/L-AMB strategy became the preferred strategy when the probability to switch due to L-AMB failure was lower than 0.329 (baseline value 0.389).
For 4 mg/kg/day probabilistic analyses on NMB showed that even when the uncertainty of the estimates for probabilities used in the decision tree were taken into account the voriconazole/caspofungin strategy remained the most cost-effective strategy for treating IA in adult bone marrow transplant recipients. The acceptability curves showed that for a wide range of willingness to pay values the voriconazole/caspofungin mostly yielded the highest NMB (Figure 4a). At a dosage of 1 mg/kg/day of L-AMB, the voriconazole/D-AMB strategy yielded the highest NMB in most of the cases up to a willingness to pay of 27 000. At willingness to pay values between 27 000 and 88 000, the voriconazole/caspofungin strategy most often yielded the highest expected value for NMB, whereas the voriconazole/L-AMB + caspofungin was the most cost-effective strategy in most of the cases at willingness to pay values higher than 88 000 (Figure 4b).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
Our analysis suggests that the voriconazole/caspofungin strategy was the most economically favourable for treating IA among HSCT recipients.
Sensitivity analysis showed this strategy was dominant over the three first-line L-AMB strategies and the first-line voriconazole followed by second-line treatment with L-AMB. Even after extensive single and multiple variable sensitivity analyses this conclusion held. This is in line with earlier studies,37–40 despite the fact that the structure of the models in these studies was slightly different.
Sensitivity analysis showed the cost of treatment with L-AMB to be the threshold for the reference strategy. Kuti et al.41 showed that in economic evaluations, the outcome is highly sensitive to the acquisition price and dosage of the antifungal agent. At a dosage of 4 mg/kg/day, the acquisition costs of L-AMB were more than twice the daily treatment costs of either voriconazole or caspofungin. Our model showed that even when the daily treatment costs are comparable, when the dose of L-AMB is 1 mg/kg/day, first-line voriconazole strategies were still dominant over the first-line L-AMB strategies. This is because when costs are equal, the attractiveness of a given strategy is determined by the effectiveness of a drug, which is itself a product of its efficacy and side effects. Voriconazole offers greater effectiveness and fewer side effects than does L-AMB at this dose. However, this resulted in voriconazole/D-AMB becoming the reference strategy because of lower expected costs of OLAT treatment.
When the drugs included in the third-line treatment were restricted to those remaining out of the four available drugs, expensive first- and second-line treatment was followed by less costly OLAT treatment and vice versa. This could have influenced the outcomes of the model. We did not evaluate combinations as OLAT therapy, though this may well be employed in practice when first- and second-line treatment is considered to have failed.
The voriconazole/L-AMB + caspofungin strategy yielded the highest survival in life-years saved but the financial effort to gain one extra life-year is 107 036 which exceeds far beyond the ceiling ratio of 20 000 that is often used to make policy decisions in the Netherlands,31,42 making it unlikely that society is willing to pay for the incremental effectiveness of this strategy. With a dosage of 1 mg/kg/day of L-AMB, the ICER between the reference strategy (voriconazole/D-AMB) and voriconazole/L-AMB + caspofungin was 48 065. The recent documentation of breakthrough invasive zygomycosis in patients treated with voriconazole stresses the need to include this class of antifungal in the spectrum of drugs used in the second-line treatment strategy. As caspofungin is inactive against zygomycetes, the combination L-AMB and caspofungin appears to be an appropriate second-line treatment in those patients who progress during therapy on voriconazole, despite the increased costs.
The probability of developing renal toxicity is lower after treatment with L-AMB as compared to D-AMB.18 Costs regarding 2.2 year of dialysis for a patient who survived IA infection after HSCT and who developed renal insufficiency are about 150 000.43 For this price, 117 patients can be treated with the voriconazole/L-AMB strategy instead of the voriconazole/D-AMB strategy based on a dosage of 4 mg/kg/day of L-AMB or 1402 patients with 1 mg/kg/day L-AMB.
The external validity of any study of this nature is heavily reliant on the assumptions made to build the model. Costs were based exclusively on Dutch retail prices and took no account of discounting. Discounting can be disregarded in cases where the time horizon is less than a year, which is the case here. Moreover, we consider only Dutch circumstances so our conclusions are only valid for countries with a comparable healthcare system and comparable costs of drugs.
We were limited by the paucity of data published. For example, although estimates of many of our important model parameters were based on the multicentre study of Herbrecht et al.24 which represented populations in North America and Europe, our main conclusions hold for a broad range of alternative assumptions as shown by the sensitivity analyses and strengthen the validity of our model.
We performed this study from a hospital perspective taking only hospital costs into account and disregarded other societal or personal costs, for example production loss and travel costs. However, at the present time this is only of theoretical concern as therapy for IA is mostly undertaken in hospital. We also did not consider caspofungin as a first-line therapy because it is not licensed for this indication. This strategy should be investigated should results of a therapy with this drug for first-line therapy prove favourable.
The costs of the antifungal drugs were based on the number of vials needed assuming an average weight per person of 70 kg and not the actual dose that might be given. The yields a discontinuous cost function but does not influence our conclusions and sensitivity analyses showed the robustness of our results. Recalculation of the model under the assumption of a different average weight y per person would also not influence our conclusions, because all strategies cost roughly x% more leading to the same priority ranking.
In conclusion, our approach suggests that a strategy of first-line therapy with voriconazole reserving caspofungin for second-line treatment is preferred from an economic viewpoint for treating IA in adult HSCT recipients in the Netherlands.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
We thank Mrs E. Brounts for her assistance in gathering the literature and improving the layout of the manuscript. We also thank UCB Pharma B.V. and Gilead Sciences for their unrestricted financial support of the study.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionTransparency declarationsReferences |
|---|
1 Kaiser L, Huguenin T, Lew PD, et al. Invasive aspergillosis. Clinical features of 35 proven cases at a single institution. Medicine (1998) 77:188–94.[CrossRef][Medline]
2 Wald A, Leisenring W, van Burik JA, et al. Epidemiology of Aspergillus infections in a large cohort of patients undergoing bone marrow transplantation. J Infect Dis (1997) 175:1459–66.[Web of Science][Medline]
3 Lin SJ, Schranz J, Teutsch SM. Aspergillosis case-fatality rate: systematic review of the literature. Clin Infect Dis (2001) 32:358–66.[CrossRef][Web of Science][Medline]
4 Denning DW, Stevens DA. Antifungal and surgical treatment of invasive aspergillosis. Rev Infect Dis (1990) 12:1147–201.[Web of Science][Medline]
5 Stevens DA, Kan VL, Judson MA, et al. Practice guidelines for diseases caused by Aspergillus. Clin Infect Dis (2000) 30:696–709.[CrossRef][Web of Science][Medline]
6 Ellis D. Amphotericin B: spectrum and resistance. J Antimicrob Chemother (2002) 49(Suppl_1):7–10.[Abstract]
7 Leenders ACAP, Daenen S, Jansen RLH, et al. Liposomal amphotericin B compared with amphotericin B deoxycholate in the treatment of documented and suspected neutropenia-associated invasive fungal infections. Br J Haematol (1998) 103:205–12.[CrossRef][Web of Science][Medline]
8 Reichenberger F, Habicht JM, Gratwohl A. Diagnosis and treatment of invasive pulmonary aspergillosis in neutropenic patients. Eur Respir J (2002) 19:743–55.[CrossRef][Web of Science][Medline]
9 Kontoyiannis DP, Hachem R, Lewis RE, et al. Efficacy and toxicity of caspofungin in combination with liposomal amphotericin B as primary or salvage treatment of invasive aspergillosis in patients with hematologic malignancies. Cancer (2003) 98:292–9.[CrossRef][Web of Science][Medline]
10 Ninin E, Milpied N, Moreau P, et al. Longitudinal study of bacterial, viral, and fungal infections in adult recipients of bone marrow transplants. Clin Infect Dis (2001) 33:41–7.[CrossRef][Web of Science][Medline]
11 Pound MW, Drew RH, Perfect JR. Recent advances in the epidemiology, prevention, diagnosis, and treatment of fungal pneumonia. Curr Opin Infect Dis (2002) 15:183–94.[Web of Science][Medline]
12 Marr KA, Carter RA, Crippa F, et al. Epidemiology and outcome of mould infections in hematopoietic stem cell transplant recipients. Clin Infect Dis (2002) 34:909–17.[CrossRef][Web of Science][Medline]
13 Kontoyiannis DP, Bodey GP. Invasive aspergillosis in 2002: an update. Eur J Clin Microbiol Infect Dis (2002) 21:161–72.[CrossRef][Web of Science][Medline]
14 Denning DW. Invasive aspergillosis. Clin Infect Dis (1998) 26:781–803.[Web of Science][Medline]
15 Perea S, Patterson TF. Antifungal resistance in pathogenic fungi. Clin Infect Dis (2002) 35:1073–80.[CrossRef][Web of Science][Medline]
16 Dixon S, McKeen E, Tabberer M, et al. Economic evaluations of treatments for systemic fungal infections. PharmacoEconomics (2004) 22:421–33.[CrossRef][Web of Science][Medline]
17 Bennett JE, Powers J, Walsh T, et al. Forum report: issues in clinical trials of empirical antifungal therapy in treating febrile neutropenic patients. Clin Infect Dis (2003) 36:S117–S122.[CrossRef][Web of Science][Medline]
18 Barret JP, Vardulaki KA, Conlon C, et al. A systematic review of the antifungal effectiveness and tolerability of amphotericin B formulations. Clin Ther (2003) 25:1295–1320.[CrossRef][Web of Science][Medline]
19 Deresinski SC, Stevens DA. Caspofungin. Clin Infect Dis (2003) 36:1445–57.[CrossRef][Web of Science][Medline]
20 Steinbach WJ, Stevens DA. Review of newer antifungal and immunomodulatory strategies for invasive aspergillosis. Clin Infect Dis (2003) 37:S157–87.[CrossRef][Web of Science][Medline]
21
Wenzel R, Del Favero A, Kibbler C, et al. Economic evaluation of voriconazole compared with conventional amphotericin B for the primary treatment of aspergillosis in immunocompromised patients. J Antimicrob Chemother (2005) 55:352–61.
22 Severens JL, Donnelly JP, Meis JF, et al. Two strategies for managing invasive aspergillosis: a decision analysis. Clin Infect Dis (1997) 25:1148–54.[Web of Science][Medline]
23
Denning DW. Echinocandins: a new class of antifungal. J Antimicrob Chemother (2002) 49:889–91.
24
Herbrecht R, Denning DW, Patterson TF, et al. Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. N Engl J Med (2002) 347:408–15.
25 Patterson TF. Early use of antifungal therapy in high-risk patients. Curr Opin Infect Dis (2002) 15:561–3.[Web of Science][Medline]
26 Ellis M. Amphotericin B and invasive aspergillosis—how do the data guide us? J Med Microbiol (2002) 51:95–7.[CrossRef][Web of Science][Medline]
27
Herbrecht R, Patterson TF, Bennett JE. Voriconazole versus amphotericin B for invasive aspergillosis; the authors reply. N Engl J Med (2002) 347:2080–1.
28 Drummond MF, O'Brien BJ, Stoddart GL, et al. Methods for the Economic Evaluation of Health Care Programmes (1997) Second edition. Oxford: Oxford University Press.
29 Ellis M, Spence D, de Pauw B, et al. An EORTC international multicenter randomized trial (EORTC number 19923) comparing two dosages of liposomal amphotericin B for treatment of invasive aspergillosis. Clin Infect Dis (1998) 27:1406–12.[Web of Science][Medline]
30 Oostenbrink J, Koopmanschap M, Rutten F. Handleiding voor kostenonderzoek (2000) Amstelveen: College voor Zorgverzekeringen.
31 van Hout BA, Al MJ, Gordon GS, et al. Costs, effects and C/E-ratios alongside a clinical trial. Health Econ (1994) 3:309–19.[Web of Science][Medline]
32
Stinnett AA, Mullahy J. Net health benefits: a new framework for the analysis of uncertainty in cost-effectiveness analysis. Med Decis Making (1998) 18:S68–80.
33 Briggs A, Sculpher M, Buxton M. Uncertainty in the economic evaluation of health care technologies: the role of sensitivity analysis. Health Econ (1994) 3:95–104.[Web of Science][Medline]
34
Doubilet P, Begg CB, Weinstein MC, et al. Probabilistic sensitivity analysis using Monte Carlo simulation. A practical approach. Med Decis Making (1985) 5:157–77.
35 Fenwick E, O'Brien BJ, Briggs A. Cost-effectiveness acceptability curves—facts, fallacies and frequently asked questions. Health Econ (2004) 13:405–15.[CrossRef][Web of Science][Medline]
36 Denning DW, Ribaud P, Milpied N, et al. Efficacy and safety of voriconazole in the treatment of acute invasive aspergillosis. Clin Infect Dis (2002) 34:563–71.[CrossRef][Web of Science][Medline]
37 Lewis JS, Boucher HW, Lubowski T, et al. Economic advantage of voriconazole (VRC) for primary treatment of invasive aspergillosis compared to conventional amphotericin B (CAB). In: Abstracts of the Forty-third Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, IL, 2003. Washington, DC, USA: American Society for Microbiology. Abstract A-1359, p. 26.
38 Marbaix S, Marciniak A, De Mees V, et al. Cost-effectiveness of voriconazole compared with conventional amphotericin B in first line treatment of invasive aspergillosis in Belgium. Value Health (2003) 6:750.
39 Grau S, Mateu-de Antonio J, Soto J, et al. Cost-effectiveness analysis of voriconazole versus amphotericin B in the treatment of invasive aspergillosis in Spain. Value Health (2003) 6:750–1.
40 Jansen JP, Meis JF, Van ‘t Wout JW, et al. Economic evaluation of voriconazole in the treatment of invasive aspergillosis. Abstracts of the Ninth Congress of the European Confederation of Medical Mycology and Seventh Trends in Invasive Fungal Infections (ECMM/TIFI), Amsterdam, 2003. Abstract P022.
41 Kuti JL, Kotapati S, Williams P, et al. Pharmacoeconomic analysis of amphotericin B lipid complex versus liposomal amphotericin B in the treatment of fungal infections. PharmacoEconomics (2004) 22:301–10.[CrossRef][Web of Science][Medline]
42 Laupacis A, Feeny D, Detsky AS, et al. How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations. CMAJ (1992) 146:473–81.[Abstract]
43 de Wit GA, Ramsteijn PG, de Charro FT. Economic evaluation of end stage renal disease treatment. Health Pol (1998) 44:215–32.[CrossRef]
44 Bowden R, Chandrasekar P, White MH, et al. A double-blind, randomized, controlled trial of amphotericin B colloidal dispersion versus amphotericin B for treatment of invasive aspergillosis in immunocompromised patients. Clin Infect Dis (2002) 35:359–66.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||