HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplementary Data Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Request Reprints Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (20) Citing Articles via Google Scholar Google Scholar Articles by Tan, K. Articles by Wood, N. Search for Related Content PubMed PubMed Citation Articles by Tan, K. Articles by Wood, N. Social Bookmarking                   What's this?

Investigation of the Potential Relationships Between Plasma Voriconazole Concentrations and Visual Adverse Events or Liver Function Test Abnormalities

From Clinical Pharmacology (Dr Tan, Dr Wood), Development Operations (Mr Brayshaw), Regulatory Affairs (Dr Tomaszewski), and Clinical R&D (Dr Troke), Pfizer Global Research and Development, Sandwich, Kent, United Kingdom.

Address for reprints: Keith Tan, Clinical R&D (IPC 096), Pfizer Global Research and Development, Ramsgate Road, Sandwich, Kent, CT13 9NJ, United Kingdom.

	ABSTRACT

TOP ABSTRACT METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
This study investigated the relationship between plasma voriconazole concentrations (pVC) and risk of visual adverse events (VAEs) or liver function test (LFT) abnormalities using longitudinal logistic regression. Seven-day mean pVC were calculated from 2925 plasma samples (1053 patients); in each 7-day period, the presence or absence of VAEs/abnormal LFTs was analyzed as a binary outcome variable. There was a relationship between pVC and risk of VAE (P = .011) and a weaker, but statistically significant, association with risk of aspartate transaminase (AST), alkaline phosphatase (ALP), or bilirubin but not alanine transaminase (ALT) abnormalities. The odds ratios of LFT abnormalities per 1 µg/mL pVC increase ranged from 1.07 to 1.17. Maximum weekly occurrences were 10%, 8%, 5%, and 14% for AST, ALT, ALP, and bilirubin abnormalities, respectively. Receiver-operating characteristic curve analysis indicates that individual pVC cannot be used to predict subsequent LFT abnormalities.

Key Words: Voriconazole • visual adverse event • liver function test • plasma concentration

In clinical studies, visual disturbances were the most commonly reported treatment-related adverse event with voriconazole.^4-7 These events occurred with an overall frequency of approximately 30% in patients but were transient, rarely required drug discontinuation, and were fully reversible.^1,4-10 A mechanistic study using electroretinograms in healthy volunteers dosed at 300 mg twice daily for a period of 28 days confirmed that the effect of voriconazole on visual function is fully reversible.^10,11 In addition, a recently completed study in which patients received voriconazole 200 mg twice daily for at least 6 months revealed no evidence of an effect of voriconazole on long-term visual function.¹²

In addition to visual disturbances, there were reports of liver function test (LFT) abnormalities associated with voriconazole use. These were usually mild to moderate in intensity, and most patients did not require treatment discontinuation.^4-8,13 Using the definitions summarized in Table I,⁹ the overall incidence of clinically significant transaminase abnormalities in patients participating in voriconazole therapeutic studies was 12.4% (Pfizer; data on file).

Increases in LFT abnormalities with voriconazole may be associated with dose, as demonstrated in a phase 1 healthy volunteer dose-escalation study¹³ and a dose escalation, safety, and pharmacokinetic (PK) study in patients at risk of fungal infection.⁷ In the healthy volunteer study, there were no LFT abnormalities in 14 subjects who received voriconazole at doses of 6 mg/kg every 12 hours intravenously on day 1, followed by 3 mg/kg every 12 hours intravenously for 6 days, followed by 200 mg every 12 hours orally for 7 days. However, when the same group received voriconazole at doses of 6 mg/kg every 12 hours intravenouslyonday1followedby5mg/kgevery12hoursintravenously for 6 days followed by 400 mg every 12 hours orally for 7 days, 5 of 14 subjects had either aspartate transaminase (AST) or alanine transaminase (ALT) values above the upper limit of normal (ULN).^13,14 In the patients at risk of fungal infection, there were no LFT abnormalities in 9 patients who received oral voriconazole at doses of 200 mg every 12 hours for 14 days. In a separate group dosed at 300 mg every 12 hours orally, 1 of 9 patients had an AST more than 3 times the ULN (146 IU/L) on the last day of treatment (day 14), which returned to normal range (9-34 IU/L) by day 27.⁷ In clinical trials comparing voriconazole with conventional or liposomal amphotericin B, the rates of LFT abnormalities were comparable between the voriconazole and liposomal or conventional amphotericin B groups.^5,8,9,14

In a study to examine the efficacy and safety of voriconazole in the treatment of acute invasive aspergillosis, Denning et al¹⁵ reported that 6 of 22 patients who developed abnormal liver function or liver failure had plasma voriconazole concentrations higher than 6 µg/mL. Furthermore, a review of hepatic laboratory abnormalities reported in clinical trials of voriconazole by Potoski and Brown¹⁶ concluded that although it is not certain, there may be an association between abnormal liver function and plasma voriconazole concentrations higher than 6 µg/mL. The authors felt that dosemodification guidelines needed to be developed based on plasma voriconazole concentrations. In contrast, Lutsar et al¹⁷ presented data that indicated that plasma voriconazole concentrations were not predictive of abnormal LFT findings or treatment outcomes.

Given that the PK of voriconazole is nonlinear with respect to dose¹³ and that there is wide interindividual variability in plasma concentrations at a given dose,^18,19 it is important to understand any potential relationship between voriconazole plasma concentrations and safety. This retrospective analysis was therefore carried out to investigate the relationship between plasma voriconazole concentrations and the risk of visual adverse events (VAEs) or LFT abnormalities in a large cohort of patients who were included in the phase 2 and 3 clinical trials.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Study Design
Safety and PK data from 10 phase 2 or 3 voriconazole therapeutic trials were analyzed retrospectively. Each clinical study protocol was approved prior to the start by independent ethics review committees whose constitutions were appropriate for the countries in which the studies were performed (for specific sites see the online appendix at http://jcp.sagepub.com/cgi/content/full/46/2/235/DC1/). A total of 2925 weekly (7day mean) plasma voriconazole concentrations were calculated using PK data from 1053 patients. This population included all patients from whom at least 1 plasma voriconazole concentration was obtained. The presence or absence of VAEs and time to onset from start of treatment were examined. The LFT outcome variables were recorded for each 7-day interval as the presence or absence of ALT, AST, alkaline phosphatase (ALP), or bilirubin abnormalities and were classified using the criteria summarized in Table I. These criteria are based on LFT values at the start of treatment, since some patients were recruited into studies with LFT values greater than the ULN.

Demography, Etiology, and Underlying Disease
The patient population consisted of 682 males (64.8%) and 371 females (35.2%). Most (81.8%) of the patients were white; the remainder were black (9.8%) and Asian/other (8.5%). Ages ranged from 11 to 85 years, with a mean of 44 years.

Patients were pooled from 10 therapeutic clinical trials including empirical therapy and treatment of infections such as aspergillosis, esophageal candidiasis, candidemia, scedosporiosis, and fusariosis. Most patients had serious underlying conditions, including hematologic malignancy, allogeneic or autologous hematopoietic cell transplantation, solid organ transplantation, or AIDS, and almost half were neutropenic (<500 cells/mm³).

Pharmacokinetics
Plasma voriconazole concentrations were measured by a validated high-performance liquid chromatography assay.²⁰ The lower limit of quantification was set at 0.1 µg/mL; concentrations below that limit were recorded as 0 in the analyses.

The raw PK data were summarized as weekly mean plasma voriconazole concentrations. While PK samples may have been taken over any part of the dosing interval, intraindividual variability in plasma voriconazole concentrations within the dosing interval and between doses was low compared with interindividual variability. Therefore, mean plasma concentrations were considered appropriate for the characterization of differences in exposure between individuals.

The weekly mean plasma concentrations were used as the PK explanatory variable in the PK/pharmacodynamic (PD) modeling. The number of weekly mean concentrations per patient ranged from 1 to 13, and the range of weekly concentrations was 0 to 20.4 µg/mL with a mean of 3.2 µg/mL.

Visual Adverse Events
The presence or absence of treatment-emergent visual adverse VAEs was analyzed as a binomial outcome variable using adverse event data recorded during the course of the studies. The term visual adverse event represents the composite of reports of visual disturbances captured during voriconazole treatment. The specific events captured included enhanced/altered visual perception, blurred vision, changes in color vision, photophobia, and other. The "other" category encompassed events that could not easily be categorized, such as "dots" and "scotomata" or imprecise terms such as visual disturbance that could not be categorized.

LFT Analyses
ALT, AST, ALP, and bilirubin concentrations were measured at baseline and monitored at predetermined time points during voriconazole administration in the various clinical studies. A binary variable was used to indicate an LFT abnormality for each 7-day interval.

The criteria used to define abnormal LFT values are summarized in Table I.

Exploratory Analyses of the PK/PD Relationship and Receiver-Operating Characteristic Curves
The overall percentage occurrences of VAEs or ALT, AST, ALP, or bilirubin abnormalities were plotted against plasma voriconazole concentration categories. Receiver-operating characteristic (ROC) curves were also plotted to explore the trade-off between the sensitivity and specificity of predicting LFT abnormalities from preceding plasma voriconazole concentrations. ROC curves examine the relationship between the probability of a true-positive test (sensitivity) against that of a false-positive test (1–specificity) for chosen plasma concentration cut points. In addition, the positive and negative predictive values of each test were calculated.^21,22

Modeling the PK/PD Relationship
The PK/PD relationships between mean plasma voriconazole concentrations and VAEs or LFTs were investigated by longitudinal logistic regression using the generalized estimating equations approach.²³ The predicted probability of an event, without covariate adjustment, was plotted with respect to plasma voriconazole concentration. The corresponding 95% confidence intervals were also presented to indicate the degree of uncertainty of the estimated probability of an event for a given plasma voriconazole concentration. To investigate potential threshold effects, the continuous plasma concentration term was replaced by 10 plasma concentration categories (ie, 0 to <1, 1 to <2, 2 to <3,...,8 to <9, and 9+ µg/mL).

Covariates
The following factors were identified a priori as potential covariates to be explored: study, age, gender, weight, race, indication, primary underlying condition, hematologic risk factors, and study medication use. For each analysis, covariates were selected for the PK/PD model by the method of forward selection. The potential for a differing PK/PD relationship per covariate was also explored by fitting interaction terms in the model. Each PK/PD model was fitted with and without covariates. However, for all analyses, the underlying PK/PD relationship was similar in each case.

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
For each analysis, the PK/PD relationship did not differ according to covariates; that is, no significant plasma voriconazole concentration-by-covariate interactions were found.

View larger version (11K): [in this window] [in a new window]   Figure 1. Plasma voriconazole concentrations and visual adverse events. The percentage weekly occurrence shown is the number of events observed in weekly time periods over the total number of weekly time periods (numbers shown above each symbol) for each plasma voriconazole concentration category.

The longitudinal logistic regression analysis revealed a significant relationship between plasma voriconazole concentration and the odds of a VAE (P = .011) and showed that the overall incidence of VAEs varied between studies. The model predicted a 4.7% increase in the odds of a VAE for every 1 µg/mL increase in plasma voriconazole concentration. The predicted probability of a VAE increased from 18% to 31% for a change in plasma voriconazole concentration from0to9 µg/mL (Figure 1). Model predictions greater than 9 µg/mL were considered unreliable due to wide confidence intervals.

Liver Function Tests
ALT, AST, ALP, and bilirubin abnormalities occurred in <10% of patients, with the exception of the >9 µg/mL category for bilirubin (Figures 2A, 3A, 4A, 5A). For ALT, the slope of the model estimate is shallow and the confidence intervals are wide, indicating lack of evidence for a relationship between plasma voriconazole concentration and ALT abnormalities (Figure 2B). For AST, ALP, and bilirubin, the predicted probabilities of an abnormality increased from approximately 3% to 8%, 1% to 3%, and 3% to 10%, respectively, for a change in plasma voriconazole concentration from 0 to 9 µg/mL (Figures 3B, 4B, 5B). Model predictions greater than 9 µg/mL were considered unreliable due to wide confidence intervals. There is no apparent threshold plasma concentration above which the odds of an abnormality were increased for any of the LFTs analyzed.

View larger version (10K): [in this window] [in a new window]   Figure 2. Plasma voriconazole concentrations and risk of ALT abnormalities. The percentage weekly occurrence shown is the number of events observed in weekly time periods over the total number of weekly time periods (numbers shown above each symbol) for each plasma voriconazole concentration category.

View larger version (10K): [in this window] [in a new window]   Figure 3. Plasma voriconazole concentrations and risk of AST abnormalities. The percentage weekly occurrence shown is the number of events observed in weekly time periods over the total number of weekly time periods (numbers shown above each symbol) for each plasma voriconazole concentration category.

View larger version (10K): [in this window] [in a new window]   Figure 4. Plasma voriconazole concentrations and risk of ALP abnormalities. The percentage weekly occurrence shown is the number of events observed in weekly time periods over the total number of weekly time periods (numbers shown above each symbol) for each plasma voriconazole concentration category.

View larger version (10K): [in this window] [in a new window]   Figure 5. Plasma voriconazole concentrations and risk of bilirubin abnormalities. The percentage weekly occurrence shown is the number of events observed in weekly time periods over the total number of weekly time periods (numbers shown above each symbol) for each plasma voriconazole concentration category.

Longitudinal logistic regression analysis revealed a statistically significant relationship (P < .001) between plasma voriconazole concentration and the odds of an AST, ALP, and bilirubin abnormality. The model predicted a 13.1%, 16.5%, and 17.2% increase in the odds of an AST, ALP, and bilirubin abnormality, respectively, for every 1 µg/mL increase in plasma voriconazole concentration (Table II). There was no significant relationship between plasma voriconazole and the odds of an ALT abnormality (P = .171). The model predicted a 7.4% increase in the odds of an ALT abnormality for every 1 µg/mL increase in plasma voriconazole concentration (Table II).

A different set of covariates was found to be significant for each LFT analysis: ALT abnormalities were more strongly associated with the risk factor "bone marrow transplant" compared with no known risk factor (odds ratio [OR] = 3.4) and the underlying condition "other immunosuppression" compared with "cancer" (OR = 6.5). The odds of an AST abnormality varied across clinical studies, age, and gender. The risk of a bilirubin abnormality varied according to race, underlying disease, and hematologic risk factor. The odds of a bilirubin abnormality were highest in "bone marrow transplant" and "neutropenic" patients compared with no known hematologic risk factor (OR = 7.5 and 5.1, respectively). The risk of an ALP abnormality varied according to gender (OR = 3.0 M/F).

ROC Curve Analysis
The ROC curve for AST abnormalities for each of the voriconazole plasma concentration cut points (0.5 to 10 µg/mL) closely follows the 45° line of identity or no discrimination (Figure 6). This indicates that for all potential plasma concentration cut points, the proportion of abnormalities correctly predicted by the test (sensitivity) were similar to the proportion of wrong predictions of abnormality (1–specificity). The results for ALT, ALP, and bilirubin were similar (data not shown). Positive predictive values for AST, ALT, ALP, and bilirubin at the 10 µg/mL cut point were approximately 12%, 9%, 7%, and 16%, respectively.

View larger version (15K): [in this window] [in a new window]   Figure 6. ROC curve for predicting AST abnormalities from plasma voriconazole concentrations.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

Although none of the clinical efficacy studies were specifically designed for PK/PD evaluation, this retrospective analysis was carried out to determine whether there was a relationship between plasma voriconazole concentration and the risk of a VAE or an LFT abnormality in voriconazole-treated patients.

Visual events were the most common voriconazolerelated adverse events in patients and volunteers, and their frequency increased with increasing plasma voriconazole concentrations. In healthy volunteers, both the maximum plasma concentration (C_max)andthearea under the plasma concentration-time curve (AUC) were positively associated with VAEs (data not shown). It was not possible to use the data from this analysis to examine relationships between VAEs and C_max or AUC in patients. However, analysis of mean plasma concentration data in patients was consistent with the data in healthy volunteers, suggesting that VAEs were more likely in patients with higher mean plasma voriconazole concentrations.

In the LFT analyses, for most weekly periods over the interval studied, the median plasma voriconazole concentrations were higher in patients with ALT, ALP, and bilirubin abnormalities than in patients without abnormalities. However, over the plasma concentration bands summarized, the weekly occurrence of LFT abnormalities was low, with maximum occurrences of approximately 10%, 8%, 5%, and 14% for AST, ALT, ALP, and bilirubin, respectively.

Potoski and Brown¹⁶ proposed therapeutic drug monitoring of patients treated with voriconazole to reduce the incidence of LFT abnormalities. However, the overall incidence of LFT elevations for voriconazoletreated patients in the comparative studies was similar to those in the AmBisome or amphotericin B deoxycholate arms^8,9 and to those recorded for other agents and comparable patient groups.²⁴ In response, Lutsar et al¹⁷ considered that therapeutic drug monitoring would be unlikely to add any value over clinical judgment and diligent monitoring of LFTs.

Therapeutic drug monitoring can be a useful clinical approach with drugs such as cyclosporine, in which narrow plasma concentration thresholds differentiate between therapeutic and potentially nephrotoxic doses. Using ROC curve analysis, it is possible to determine threshold plasma concentrations of cyclosporine that optimize efficacy and minimize nephrotoxicity.²⁵ In the ROC curve for cyclosporine and nephrotoxicity, there are positive and significant deviations from the line of identity, indicating that cyclosporine concentration can be used to detect a high proportion of cases of potential nephrotoxicity, with a few false positives.

In contrast, the evidence presented here suggests that the absolute risk of an LFT abnormality in voriconazoletreated patients is low. There was no statistically significant relationship between plasma voriconazole concentrations and ALT abnormalities, whereas statistically significant, but weak, associations were identified between plasma voriconazole concentrations and AST, ALP, and bilirubin abnormalities. Furthermore, unlike cyclosporine,²¹ ROC curve analysis indicated that plasma voriconazole concentrations cannot be used to discriminate between true-positive and falsepositive LFT abnormalities using cut points of voriconazole plasma concentrations from 0.5 to 10 µg/mL. Therefore, individual plasma voriconazole concentrations cannot be regarded as a good predictor for LFT abnormalities.

There is also little value in measuring plasma voriconazole concentrations in the absence of an abnormal LFT. For example, the model predicts a probability of an AST abnormality of 7%, even when the plasma voriconazole concentration is as high as 10 µg/mL. The ROC analysis shows that discontinuing or altering therapy solely because the plasma voriconazole concentration was 10 µg/mL would be unnecessary in 88% of cases since this strategy was associated with a positive predictive value of only 12%. Hence, routine LFT measurement would provide more valuable clinical information than monitoring plasma voriconazole concentrations.

In conclusion, the relationships between plasma voriconazole concentrations and VAE and LFT abnormalities have been explored quantitatively. ROC analysis indicated that therapeutic drug monitoring is unlikely to add any value over routine LFT monitoring and clinical judgment in guiding dose adjustment in patients treated with voriconazole.

	ACKNOWLEDGEMENTS

TOP ABSTRACT METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
This work was carried out with the financial support of Pfizer Inc. The authors acknowledge Michael Oakes for his contribution to the analyses of these studies.

DOI: 10.1177/0091270005283837

	REFERENCES

TOP ABSTRACT METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 

1. Ghannoum MA, Kuhn DM. Voriconazole: better chances for patients with invasive mycoses. Eur J Med Res. 2002;7: 242-256.[Web of Science][Medline] [Order article via Infotrieve]

2. VFEND: Summary of Product Characteristics. Surrey, UK: Pfizer Ltd; 2002.

3. Vfend [package insert]. New York, NY: Pfizer Inc; 2003.

4. Ally D, Schürmann W, Kreisel G, et al. A randomized double blind, double dummy multicenter trial of voriconazole and fluconazole in the treatment of oesophageal candidiasis in immunocompromised patients. Clin Infect Dis. 2002;33: 1447-1455.

5. Herbrecht R, Denning DW, Patterson TF, et al. Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. N Engl J Med. 2002;347: 408-415.[Abstract/Free Full Text]

6. Perfect J, Marr K, Walsh T, et al. Voriconazole treatment for less common, emerging, or refractory fungal infections. Clin Infect Dis. 2003;36: 1122-1131.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]

7. Lazarus HM, Blumer JL, Yanovich S, et al. Safety and pharmacokinetics of oral voriconazole in patients at risk of fungal infection: a dose escalation study. J Clin Pharmacol. 2002;42: 395-402.[Abstract]

8. Walsh TJ, Pappas P, Winston DJ, et al. Voriconazole compared with liposomal amphotericin B for empirical antifungal therapy in patients with neutropenia and persistent fever. N Engl J Med. 2002;346: 225-234.[Abstract/Free Full Text]

9. Walsh TJ, Finberg RW, Arndt C, et al. Liposomal amphotericin B for empirical therapy in patients with persistent fever and neutropenia. N Engl J Med. 1999;340: 764-771.[Abstract/Free Full Text]

10. Boucher HW, Groll AH, Chiou CC, et al. Newer systemic antifungal agents: pharmacokinetics, safety and efficacy. Drugs. 2004;64: 1997-2020.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]

11. Tomaszewski K, Purkins L. Visual effects with voriconazole [abstract]. In: Program and Abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, IL, December 16-19, 2001. Washington, DC: American Society for Microbiology; 2001: A-639.

12. Tomaszewski K, Laites A, Goodrich J. The long-term visual safety of voriconazole [abstract]. Clin Microbiol Infect. 2005;11(S2): 347-348. Abstract P1109.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]

13. Purkins L, Wood N, Ghahramani P, et al. Pharmacokinetics and safety of voriconazole following intravenous- to oral-dose escalation regimens. Antimicrob Agents Chemother. 2002;46: 2546-2553.[Abstract/Free Full Text]

14. Antiviral Drugs Advisory Committee. Briefing document for voriconazole. 2001. Available at: http://www.fda.gov/ohrms/dockets/ac/01/briefing/3792b2_01_Pfizer.pdf. Accessed May 20, 2005.

15. 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: 563-571.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]

16. Potoski BA, Brown J. The safety of voriconazole. Clin Infect Dis. 2002;35: 1273-1275.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]

17. Lutsar I, Hodges MR, Tomaszewski T, et al. Safety of voriconazole and dose individualization. Clin Infect Dis. 2003;36: 1087-1088.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]

18. Purkins L, Wood N, Greenhalgh K, et al. The pharmacokinetics and safety of intravenous voriconazole: a novel wide-spectrum antifungal agent. Br J Clin Pharmacol. 2003;56(S1): 2-9.

19. Purkins L, Wood N, Greenhalgh K, et al. Voriconazole, a novel wide-spectrum triazole: oral pharmacokinetics and safety. Br J Clin Pharmacol. 2003;56(S1): 10-16.

20. Stopher DA, Gage R. Determination of a new antifungal agent, voriconazole, by multidimensional high-performance liquid chromatography with direct plasma injection onto a size-exclusion column. J Chromatogr B Biomed Sci Appl. 1997;691: 441-448.[CrossRef][Medline] [Order article via Infotrieve]

21. Altman DG. Practical Statistics for Medical Research. New York, NY: Chapman and Hall; 1991: 409-418.

22. Hanley JA, McNeil BJ. The meaning and use of the area under the receiver operating characteristic (ROC) curve. Radiology. 1982;143: 29-36.[Abstract/Free Full Text]

23. Liang KY, Zeger SL. Longitudinal data analysis using generalised linear models. Biometrika. 1986;73: 13-22.[Abstract/Free Full Text]

24. Bohme A, Karthaus M. Treatment of systemic fungal infections in patients with hematologic malignancies. Antibiot Chemother. 2000; 50: 79-93.[Medline] [Order article via Infotrieve]

25. Min DI, Perry PJ, Chen HY, et al. Cyclosporine trough concentrations in predicting allograft rejection and renal toxicity up to 12 months after renal transplantation. Pharmacotherapy. 1998;18: 282-287.[Web of Science][Medline] [Order article via Infotrieve]
CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

This article has been cited by other articles:

				B. H. Segal Aspergillosis N. Engl. J. Med., April 30, 2009; 360(18): 1870 - 1884. [Full Text] [PDF]

				D. Andes, A. Pascual, and O. Marchetti Antifungal Therapeutic Drug Monitoring: Established and Emerging Indications Antimicrob. Agents Chemother., January 1, 2009; 53(1): 24 - 34. [Full Text] [PDF]

				A. Howard, J. Hoffman, and A. Sheth Clinical Application of Voriconazole Concentrations in the Treatment of Invasive Aspergillosis Ann. Pharmacother., December 1, 2008; 42(12): 1859 - 1864. [Abstract] [Full Text] [PDF]

				P. Liu, G. Foster, K. Gandelman, R. R. LaBadie, M. J. Allison, M. J. Gutierrez, and A. Sharma Steady-State Pharmacokinetic and Safety Profiles of Voriconazole and Ritonavir in Healthy Male Subjects Antimicrob. Agents Chemother., October 1, 2007; 51(10): 3617 - 3626. [Abstract] [Full Text] [PDF]

				S. Weiler, H. Zoller, I. Graziadei, W. Vogel, R. Bellmann-Weiler, M. Joannidis, and R. Bellmann Altered Pharmacokinetics of Voriconazole in a Patient with Liver Cirrhosis Antimicrob. Agents Chemother., September 1, 2007; 51(9): 3459 - 3460. [Full Text] [PDF]

				C. Eiden, H. Peyriere, M. Cociglio, S. Djezzar, S. Hansel, J.-P. Blayac, D. Hillaire-Buys, and for the Network of the French Pharmacovigilance Ce Adverse Effects of Voriconazole: Analysis of the French Pharmacovigilance Database Ann. Pharmacother., May 1, 2007; 41(5): 755 - 763. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplementary Data Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Request Reprints Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (20) Citing Articles via Google Scholar Google Scholar Articles by Tan, K. Articles by Wood, N. Search for Related Content PubMed PubMed Citation Articles by Tan, K. Articles by Wood, N. Social Bookmarking                   What's this?