VOLUME 1 NUMBER 2 2007
The Journal of INVASIVE FUNGAL infections
EDITOR-IN-CHIEF John R Perfect, Durham, NC, USA
Clinical Trials of Antifungal Prophylaxis in Surgical Critically Ill Patients Ed Horn, Sandra M Swoboda, and Pamela A Lipsett
Preemptive Antifungal Therapy in Critically Ill Surgical Patients Frédéric Grenouillet, Laurence Millon, Gilles Blasco, and Renaud Piarroux
Diagnostic and Therapeutic Approaches to Fungal Infections in Critical Care Settings: Different Options but the Same Strategy Rafael Zaragoza and Javier Pemán
www.invasivefungalinfections.com
Jointly sponsored by the University of Kentucky Colleges of Pharmacy and Medicine and Remedica. The University of Kentucky is an equal opportunity university. This journal is supported by an educational grant from Schering-Plough The Journal of Invasive Fungal Infections is supported by an educational grant from Schering-Plough Corporation
Editor-in-Chief Editor John Perfect Zeina A Kanafani Division of Infectious Diseases Director, Research Fellow, Division of Infectious Diseases, Mycology Research Unit, Duke University Duke University Medical Center, Durham, NC, USA Medical Center, Durham, NC, USA
Editorial Advisory Board David Andes Graeme Forrest Assistant Professor, Departments of Medicine Director of Antimicrobial Program at University and Medical Microbiology & Immunology, of Maryland and Assistant Director of Infectious University of Wisconsin, Madison, WI, USA Diseases in the Greenebaum Cancer Center Division of Infectious Diseases, University of Maryland Oliver Cornely School of Medicine, Baltimore, MD, USA Assistant Professor of Medicine, Klinik I für Innere Medizin, Klinikum der Universität Köln, Köln, Raoul Herbrecht Germany Professor of Hematology, Department of Hematology and Oncology, Ben de Pauw Hôpital de Hautepierre, Strasbourg, France Professor of Medicine, Department of Blood Transfusion Service and Transplant Immunology, Dimitrios P Kontoyiannis University Medical Center St Radboud, Director of Mycology Research Program, Nijmegen, The Netherlands The University of Texas, MD Anderson Cancer Center, Houston, TX, USA Francoise Dromer Head of the Molecular Mycology Unit and Nina Singh Director to the National Reference Center for Associate Professor of Medicine at University Mycoses & Antifungals, National Reference of Pittsburgh and Chief of Transplant Infectious Center for Mycoses & Antfungals, Institut Diseases, Infectious Diseases Section, Pasteur, Paris, France VA Medical Center, Pittsburgh, PA, USA
Jay Fishman Andrew Ullmann Associate Professor of Medicine, Harvard Infectious Disease Consultant, Klinikum Johannes Medical School and Associate Director of MGH Gutenberg-Universität, Mainz, Germany Transplantation Center, Director of Transplant Infectious Disease and Compromised Host Program, Jo-Anne van Burik Infectious Disease Division, Massachusetts General Associate Professor, Department of Medicine, Hospital, Boston, MA, USA University of Minnesota, Minneapolis, MN, USA Foreword Dear Colleagues, Contents Welcome to the second issue of the CME/CPE-accredited journal, Leading Articles The Journal of Invasive Fungal Infections. Clinical Trials of Antifungal Prophylaxis As in the first issue, this issue contains review articles from leading specialists in the field. These have been peer-reviewed for quality and in Surgical Critically Ill Patients CME/CPE-accredited to provide an ongoing educational resource. Ed Horn, Sandra M Swoboda, and Pamela A Lipsett 34 Three such leading articles are included, and all discuss antifungal therapeutic strategies in critically ill patients. Preemptive Antifungal Therapy Professor Pamela Lipsett and colleagues (Johns Hopkins University in Critically Ill Surgical Patients School of Medicine and Nursing, Baltimore, MD, USA) provide an Frédéric Grenouillet, Laurence Millon, excellent overview of all clinical trials conducted to date on antifungal Gilles Blasco, and Renaud Piarroux 42 prophylaxis in surgical critically ill patients. They also discuss meta- analyses of clinical trials of antifungal prophylaxis, and discuss some of the concerns surrounding this therapeutic strategy in patients in Diagnostic and Therapeutic Approaches intensive care. Comprehensive tables summarizing the clinical trials and to Fungal Infections in Critical Care Settings: meta-analyses are included. Different Options but the Same Strategy Following this, Dr Frédéric Grenouillet and colleagues (University Rafael Zaragoza and Javier Pemán 50 Hospital, Besançon, France) review preemptive antifungal therapy, again in surgical critically ill patients. Firstly, they discuss the issues Clinical Reviews regarding the definition of “preemptive” therapy, and the use of Candida colonization status of the patient in this classification. Aspergillosis 59 Focusing on invasive candidiasis, they discuss its pathophysiology and Coccidioidomycosis 61 risk factors, and consider studies that have aimed to define criteria that predict which patients are at high risk of infection. Lastly, they discuss a Candidiasis 63 management flowchart for fungal infection risk, in use at their unit Cryptococcosis 65 since 2004. The third and final leading article of this issue, from Drs Rafael Other Mycoses 66 Zaragoza and Javier Pemán (Hospital Universitario Peset and Hospital Therapeutics 67 Universitario La Fe, Valencia, Spain), addresses the diagnostic and therapeutic approaches to fungal infections in critically ill patients. Meeting Report Preemptive and prophylactic therapeutic strategies are reviewed. In addition, newer diagnostic tests, such as the detection of (1,3)-β-D- The American Society of Hematology glucan, which may guide the therapeutic strategy in critically ill patients 48th Annual Meeting and Exposition are discussed. The choice of antifungal agent and the potential for Orlando, FL, USA, December 9–12, 2006 69 combination therapy in these patients are also reviewed. As in the first issue, Dr Zeina Kanafani (Duke University Medical Center, Durham, NC, USA) has provided a thorough analysis of recent papers in our Clinical Reviews section. This is followed by a report from the 48th Annual Meeting and Exposition of the American Society of Hematology from Dr Ana Berceanu and Professor Raoul Herbrecht (Hôpitaux Universitaires de Strasbourg, Strasbourg, France), in which the presentations describing invasive fungal infections, primarily aspergillosis, in patients with hematological malignancies or hematopoietic stem cell transplantation are reviewed. We hope you find this issue of The Journal of Invasive Fungal Infections an educational and valuable tool. We welcome your feedback regarding the material presented as well as your suggestions for future topics to be covered. John R Perfect Editor-in-Chief
Editorial Policy Publisher’s Statement The Journal of Invasive Fungal Infections is an independent journal published by Remedica ©2007 Remedica Medical Education and Publishing. All rights reserved. No part of this Medical Education and Publishing. Editorial control is the sole responsibility of the Editor-in-Chief, publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any Editorial Advisory Board, and the Editors. Before publication, all material submitted to the journal means, electronic, mechanical, photocopying, recording, or otherwise without the prior permission is subjected to rigorous review by the Editor-in-Chief, Editorial Advisory Board, Editors, and/or of the copyright owners. While every effort is made by the publishers and editorial board to see independent reviewers for suitability of scientific content, scientific accuracy, scientific quality, and that no inaccurate or misleading data, opinions, or statements appear in this journal, they wish to conflict of interest. make it clear that the material contained in the publication represents a summary of the Aims and Scope independent evaluations and opinions of the authors and contributors. As a consequence, the The Journal of Invasive Fungal Infections is designed to bring a critical analysis of the world fungal board, publishers, and any supporting company accept no responsibility for the consequences of infection literature, written by clinicians, for clinicians, to an international, multidisciplinary audience. any such inaccurate or misleading data or statements. Neither do they endorse the content of the Our mission is to promote better understanding of fungal infections across the global healthcare publication or the use of any drug or device in a way that lies outside its current licensed system by providing an active forum for the discussion of clinical and healthcare policy issues. application in any territory. Leading Articles – These major review articles are chosen to reflect topical clinical and healthcare Editorial Team: Rhian Phillips, Scott Millar, Samantha K Bell Editorial Director: Reghu Venkatesan policy issues in invasive fungal infections. All contributions undergo a strict editorial review process. Design and artwork: AS&K Skylight Creative Services Publishers: Ian Ackland-Snow, Simon Kirsch Clinical Reviews – The most important papers from the best of the international literature on fungal The Journal of Invasive Fungal Infections (ISSN 1753-3783) is published four times a year. infections are systematically selected by the Editor-in-Chief. The Editors then prepare concise and Additional subscription information is available from the publishers: Remedica Medical Education critical analyses of each paper, and, most importantly, place the findings into clinical context. and Publishing Ltd, Commonwealth House, 1 New Oxford Street, London WC1A 1NU, UK. Meeting Reports – The Journal of Invasive Fungal Infections also provides incisive reportage from Telephone: +44 (0)20 7759 2999 Fax: +44 (0)20 7759 2951 Email: [email protected] the most important international congresses. ISSN 1753-3783 Clinical Trials of Antifungal Prophylaxis in Surgical Critically Ill Patients
Ed Horn1, Sandra M Swoboda2,3, and Pamela A Lipsett 2,3 1Clinical Pharmacy Specialist, Surgical Intensive Care Unit, Johns Hopkins Hospital, 2School of Nursing, and 3Department of Surgery, Johns Hopkins University Schools of Medicine and Nursing, Baltimore, MD, USA
Fungal infections represent a difficult group of nosocomial infections with a growing incidence in the critically ill surgical patient. LEADING ARTICLE Multiple strategies aimed at preventing fungal infections utilizing a myriad of antifungal agents have been evaluated. In clinical trials on fungal prophylaxis, as well as systematic reviews, antifungal agents have been shown to prevent invasive fungal disease in high-risk patient populations, without affecting overall mortality rates. Fungal prophylaxis is an option in patients at high-risk of developing fungal infections, but should be utilized with caution. Resistant pathogens present the greatest concern with regard to widespread prophylaxis. More data are needed to measure any possible reduction in mortality rate that may occur with antifungal prophylaxis, as well as the impact on drug resistance. J Invasive Fungal Infect 2007;1(2):34–41.
Nosocomial infections are an important source of morbidity differences in definitions contribute to the ongoing and mortality in hospitalized patients. In the intensive care confusion and controversy in this area. In some institutions, unit (ICU), infections from Candida spp. are increasingly antifungal prophylaxis is performed without generalized common; indeed, candidemia is now the fourth leading consensus regarding the evidence from randomized clinical cause of bloodstream infection in the surgical ICU [1–4]. As trials of prophylaxis [5]. many as 50% of invasive candidiasis cases remain When considering a prophylaxis regimen, several criteria undiagnosed until autopsy [1]. Risk factors that are must be considered. The overall incidence and severity of associated with fungal infection in surgical ICU patients fungal disease must be high and/or fungal infections must include the following [2,3]: cause a substantial and measurable increase in morbidity and mortality rates in the patient population. The • Severity of illness. prophylactic regimen must have high tolerability, be cost- • Invasive devices or procedures. effective, and be confined to use in the appropriate high-risk • Broad-spectrum antibiotics. patients. Several published clinical trials have evaluated • Parenteral nutrition. the use of antifungal agents for prophylaxis (azoles, • Immunosuppression. amphotericins, and echinocandins) in selected critically ill • Length of stay in ICU. and surgical patients (Table 1) [6–16]. This paper will review • Fungal colonization. the use of the individual agents for prophylaxis in the critically ill adult surgical patient, focusing on recently Defining the surgical ICU patient who will benefit from published randomized, placebo-controlled trials and classic antifungal prophylaxis is difficult. First, consensus on the papers in this patient population. definition of prophylaxis versus preemptive treatment must be reached. Prophylaxis is defined as the use of an agent to Amphotericin B prevent infection prior to known colonization, whereas Amphotericin B has been utilized for the prevention of preemptive therapy is the introduction of an agent to fungal infections in several clinical trials. In surgical patients, patients with well-established risk factors, including a known recent trials have focused on the liver transplant population. degree of colonization with Candida spp. [4]. These subtle The use of targeted amphotericin B lipid emulsion at three different doses was prospectively studied in 30 consecutive Address for correspondence: Pamela A Lipsett, Johns Hopkins Hospital, high-risk liver transplant patients (with a stepwise reduction Osler 603, 600 N. Wolfe St, Baltimore, MD 21287, USA. in dose at 10-patient intervals) [6]. Targeted patients were Email: [email protected] those who required mechanical ventilation or renal support
34 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ANTIFUNGAL PROPHYLAXIS IN SURGICAL CRITICALLY ILL PATIENTS
through continuous hemofiltration for ≥5 days. All patients infection with Candida spp. Routine surveillance was received oral nystatin prophylaxis per standard protocol, performed, and the incidence of Candida spp. colonization and patients with identifiable pretransplant risk factors was significantly lower with ketoconazole, with no fungal for invasive fungal infection (fulminant hepatic failure, sepsis occurring in this group [10]. retransplantation, or treatment pre-ICU) received intravenous or oral fluconazole prophylaxis. Singhal et al. Itraconazole concluded that amphotericin B prophylaxis in high-risk Itraconazole is an azole available in both oral and intravenous patients after liver transplantation is well tolerated and may formulations. The spectrum of activity of itraconazole prevent invasive fungal infection [6]. includes both Candida and Aspergillus species. The spectrum Another study of liver transplant patients examined the of activity and oral availability of itraconazole make it an use of low-dose amphotericin B preparations for the attractive potential choice for antifungal prophylaxis. prevention of invasive infection [7]. High-risk individuals However, oral absorption may be limited under certain were identified, and patients received either intravenous conditions (increased gastric pH and gastrointestinal motility) conventional amphotericin B or liposomal amphotericin B and the intravenous formulation should be used cautiously in until discharge from the ICU, death, or the need for patients with renal insufficiency. In a randomized, placebo- treatment of suspected or confirmed infection. Eighty-three controlled study, Sharpe et al. evaluated the use of patients met the inclusion criteria. The incidence of invasive itraconazole (preoperative loading dose of 5 mg/kg followed fungal infection was low in both treatment arms – three in by 2.5 mg/kg daily) [11]. Nine of 38 patients in the placebo the lipid group and two in the conventional group. Candida group versus one of 33 patients in the itraconazole group spp. were isolated from the five subjects. The only significant (24% vs. 4%; p=0.04) acquired a fungal infection (either difference between the two study groups was that patients deep or suspected). However, there was no difference in who received lipid amphotericin B were more likely to mortality rate between the two groups. Thus, oral survive ICU discharge (79.6% vs. 59.5%; p=0.038), but this itraconazole is an acceptable prophylactic agent for the difference could not be specifically attributed to fungal prevention of fungal infections in liver transplant patients. infections. Thus, in these studies, there were no specific benefits of amphotericin B. Fluconazole To date, the agent with the most data on the potential Nystatin benefits of antifungal prophylaxis in surgical ICU patients is The use of oral nystatin versus placebo as antifungal fluconazole. Winston et al. examined the prophylactic use of prophylaxis for prevention of Candida spp. colonization was fluconazole versus placebo in liver transplant recipients [12]. studied in a clinical trial of 98 medical/surgical/trauma ICU Patients were randomized intraoperatively and treatment patients who were ventilated for >48 h [8]. In this trial, oral continued for 10 weeks post-transplant. Proven fungal nystatin prevented Candida spp. colonization in patients at infections of any type occurred less often in the fluconazole low risk of developing invasive disease; however; there were group compared with the placebo group (9% vs. 43%; no convincing data regarding the use of nystatin for p<0.001), with Kaplan–Meier estimates showing similar prevention of fungal infections. results. The incidence of invasive fungal infections was lower in the fluconazole group than the placebo group Azoles (6% vs. 23%; p<0.001). However, fluconazole had no effect Ketoconazole on overall survival compared with placebo (11% vs. 14%), Savino et al. prospectively randomized 292 surgical and although death related to fungal infection was lower in the trauma patients with an ICU stay of >48 h to one of four fluconazole group (2% vs. 13%; p=0.003). Furthermore, groups – placebo, clotrimazole, ketoconazole, or nystatin – colonization with non-albicans strains was not increased in to determine whether these regimens prevented patients treated with prophylactic fluconazole. colonization, fungal sepsis, or mortality. There were no Ables et al. conducted a small trial on the effects of differences between the groups in terms of these factors; fluconazole prophylaxis on mortality and infectious thus, prophylaxis did not provide a benefit [9]. Fungal outcomes [13]. Trauma and surgical ICU patients with an infection in this population was low (3%), and the study expected length of ICU stay of ≥48 h and the presence of was not powered to determine a difference between the more than one risk factor were randomized to fluconazole four groups in terms of this parameter. In another study of prophylaxis (intravenous loading dose followed by a daily ketoconazole, Slotman et al. performed a placebo-controlled dose) or placebo [13]. Infection was classified as trial in 57 surgical ICU patients with clinical risk factors for documented candidiasis (presence of positive cultures),
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 35 ED HORN, SANDRA M SWOBODA, AND PAMELA A LIPSETT Results (mortality) Mortality: Placebo 15% vs. clotrimazole 14% vs. ketoconazole 6% vs. nystatin 20%; p=NS Overall: 20%. None to IFI related 3-month survival: AmB 54.8% vs. LipAmB 69.4%; p=0.108 Not reported Mortality: 60% of those with fungal sepsis died Mortality: placebo 15.7% vs. itraconazole 3%; p=NS Overall: placebo 11% vs. fluconazole 14%; p>0.2 IFI related: placebo 13% vs. fluconazole 2% AmB: two episodes LipAmB: three episodes Placebo 25% vs. nystatin 0% invasion: Yeast placebo 17% vs. ketoconazole 0% Results (infection) No patients had evidence of fungal infection Placebo 3% vs. clotrimazole 1% vs. ketoconazole 2% vs. nystatin 7%; p=NS IFI: Proven placebo 43% vs. fluconazole 9%; p<0.001 IFI: placebo 23% vs. fluconazole 6%; p<0.001 DFI: placebo 16% vs. itraconazole 4%; p=0.225 SFI: placebo 8% vs. itraconazole 0%; p=0.141 spp. spp. Aspergillus Candida Candida spp. in trachea, stomach, urine, Candida from more than three normally non-sterile sites (throat, than three more from urine, sputum) rectum, normally non-sterile sites (throat, than three more from urine, sputum) rectum, spp. infection + isolating hyphae in BAL to SFI: clinical signs/symptoms with fever unresponsive antibiotics; highly suggestive broad-spectrum radiological findings without mycological/histological of evidence; clinical signs/symptoms in the presence fungal colonization Infection definitions IFI: histological evidence of tissue invasion; isolation one normally sterile site; isolation of from IFI: histological evidence of tissue invasion; isolation one normally sterile site; isolation of from of Presence and blood groin, rectum, than of sterile site specimen or more Positive culture two other sites DFI: positive histological evidence; one Proven clinical/radiological signs of blood culture; vagina, GI tract, wound SFI: positive skin, oropharynx, with signs of inflammation, ulceration, or urine culture plaque sinuses IFI: positive blood, pulmonary tissue, secretions, with other organs or peritoneal cavity cultures; signs/symptoms of infection not otherwise explained Positive culture of sterile site Positive culture spp. Primary outcomes Incidence of DFI IFI; SFI Proven Prevention of IFI Prevention of IFI Prevention Reduction of Candida colonization of yeast Prevention sepsis of yeast Prevention invasion Intervention (prophylaxis vs. preemptive) Prophylaxis; clotrimazole 10 mg PO TID or ketoconazole 200 mg PO daily or nystatin 2 million units every 6 h Prophylaxis; itraconazole 5 mg/kg pre- then operatively, 2.5 mg/kg PO BID for a total 56 days Prophylaxis; fluconazole 400 mg IV/PO daily for 10 weeks Prophylaxis; ABLC 1–5 mg/kg IV daily until ICU discharge Prophylaxis; AmB 15 mg IV daily or LipAmB 50 mg IV daily until ICU discharge Prophylaxis; nystatin 3 million units/day Prophylaxis; ketoconazole 200 mg PO daily until ICU discharge (maximum 21 days) Sample size 30 92 episodes, 83 patients 98 292 57 71 212 Patient population Liver transplant ICU care requiring for >5 days Medical/surgical ICU (low risk) Surgical/trauma ICU (non-burn, non-transplant) ICU Surgical Liver transplant Liver transplant ICU care requiring for >5 days Liver transplant Summary of trials examining fungal prophylaxis. Table 1. Table Normand et al. [8] Slotman et al. [10] Authors Singhal et al. [6] Shah et al. [7] Savino et al. [9] Sharpe et al. [11] Winston et al. Winston [12]
36 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 Ables et al. Trauma, intra- 125 Prophylaxis; Documented Documented candidiasis: positive culture from Placebo 20% vs. Placebo 20% vs. [13] abdominal, or intra- fluconazole 800 mg candidiasis respiratory, mucosal, or peritoneal specimens fluconazole 13% fluconazole 20% thoracic surgery; IV ×1, then 400 mg expected ICU LOS IV/PO daily until ICU Suspected IFI: compatible clinical illness (including SIRS) >48h; >1 risk factor discharge (maximum but no evidence of bacterial/viral cause 6 weeks) Superficial FI: fungal UTI, thrush, skin lesions Eggimann et al. Surgical patients 49 Prophylaxis; Frequency and time Intra-abdominal abscess/peritonitis caused by Candida Overall: [14] with recurrent GI fluconazole 400 mg to intra-abdominal Candida spp.; one or more blood cultures positive for peritonitis: placebo 50% vs. perforations or IV daily until Candida infections Candida spp.; urine culture positive for >100 000 CFU placebo 35% vs. fluconazole 30%; anastomotic complete resolution Candida spp.; clinical evidence of tissue infection with fluconazole 4%; p=0.23 leakage of abdominal disease Candida spp. isolated from biopsy p=0.02 (median 15–17 days) Canidida All Candida peritonitis-related:
T infections: placebo 20% vs. HE placebo 35% vs. fluconazole 0%; p J
UNLOF OURNAL fluconazole 9%; value not reported p=0.06 Garbino et al. Medical/surgical 220 Prophylaxis; Development of Candidemia: more than one positive blood culture + SFIs: placebo Overall incidence: [15] patients ventilated fluconazole 100 mg severe Candida histologically documented IFI or Candida 9.9% vs. placebo 16% vs. for 48 h with IV daily + SDD until infection endophthalmitis; more than two positive blood cultures fluconazole fluconazole 5.8%; I NVASIVE anticipated duration withdrawn from from two distinct sites 3.8%; p value RR 0.35; 95% CI of >72 h mechanical not reported 0.11–0.94. Overall: ventilation Severe candidemia: SIRS plus placebo 41% vs. candidemia
F fluconazole 41% UNGAL Pelz et al. Surgical ICU 260 Prophylaxis; Time to occurrence Definite: histological/microbiological evidence from two Proven IFI: Overall: [16] patients with an fluconazole 800 mg of fungal infection separate sterile sites (not bladder and urine) placebo 15.3% placebo 12.3% vs. A I expected LOS of enterally ×1, then during ICU stay vs. fluconazole 10.7%; NFECTIONS >72 h 400 mg enterally Presumed: positive blood culture or single sterile site fluconazole p=NS NTIFUNGAL daily until ICU (not indwelling drains or catheters); catheter with >15 8.5%, p<0.01 discharge CFU, as per CDC guidelines; deep surgical site infection requiring debridement; two positive urine cultures after Time to proven
V urinary catheter change infection: P OL
Kaplan–Meier IN ROPHYLAXIS 1 N Proven IFI: definite or presumed curve favors
O fluconazole over 2 2007 placebo; p<0.01
Piarroux et al. Surgical ICU 933 Preemptive; Prevention of Proven infections as per EORTC/MSG (European Proven Not reported
[19] fluconazole IV 800 mg proven candidiasis Organization for Research and Treatment of candidiasis: S ×1, then 400 mg Cancer/Mycoses Study Group) control: 7% URGICAL daily for 2 weeks preemptive: 3.8%; p=0.03 C
ABLC: amphotericin B lipid complex; AmB: amphotericin B; BAL: bronchoalveolar lavage; BID: twice daily; CDC: Centers for Disease Control and Prevention; CFU: colony-forming units; CI: confidence interval; RITICALLY DFI: deep fungal infection; GI: gastrointestinal; ICU: intensive care unit; IV: intravenous; IFI: invasive fungal infection; LipAmB: liposomal amphotericin B; LOS: length of stay; NS: not significant; PO: orally; RR: relative risk; SDD: selective digestive decontamination; SIRS: systemic inflammatory response syndrome; SFI: superficial fungal infection; TID: three times daily; UTI: urinary tract infection. I LL P ATIENTS 37 ED HORN, SANDRA M SWOBODA, AND PAMELA A LIPSETT
suspected infection (inflammatory response but no clear 15.4% to 8.5% (p<0.01) – a 7% absolute risk reduction evidence of infection), or superficial signs of infection. While that translates into a number needed to treat (NNT) to there was a trend towards a decreased incidence of infection prevent one fungal infection of 15 patients. Of the with fluconazole (13% vs. 19%; relative risk [RR] 0.66, 95% 11 patients in the fluconazole group with a proven fungal confidence interval [CI] 0.29–1.49), there was no effect on infection, four were infected prior to randomization but overall mortality. Furthermore, the inclusion of patients with were included in the analysis to take the most conservative an inflammatory response but no clear signs of fungal approach to the analysis. Taking this into consideration, infection make this study difficult to interpret. fluconazole reduced the incidence of fungal infections by a Eggimann et al. randomized 49 high-risk recurrent factor of 2- to 3-fold [17]. A time-to-event analysis showed gastrointestinal surgery patients to intravenous fluconazole that fluconazole had a significant advantage in delaying the or placebo [14]. Primary outcomes included frequency of, time to fungal infection (p<0.01). A multivariate Cox and time to, intra-abdominal candidiasis, as well as the regression model showed that the following factors emergence and persistence of colonization. There was a significantly increased the risk of fungal infection: significantly lower incidence of Candida peritonitis in the fluconazole group compared with the placebo group (4% • Acute Physiology and Chronic Health Evaluation III vs. 35%; p=0.02), and time to infection analysis revealed (APACHE III) score. that fluconazole prophylaxis was associated with a longer • Days to first dose of study drug. disease-free interval (p=0.04). Overall fungal infection rates • Fungal colonization at enrollment. were 8.7% in the fluconazole group and 35% in the placebo • Parenteral nutrition prior to enrollment. group (RR 0.25, 95% CI 0.06–1.06; p=0.06). There were no significant differences in mortality rate and length of stay in Thus, when considering antifungal prophylaxis, delaying hospital between fluconazole and placebo groups. However, therapy could decrease its effectiveness. The study was not this well-conducted trial shows that, in highly selective powered to assess the effect on mortality. This was the only surgical patients, fluconazole prophylaxis is effective and trial to exclusively use enteral fluconazole for prophylaxis. A should be considered for prevention of candidal peritonitis. follow-up study published by the same investigators showed In a larger randomized, placebo-controlled trial, Garbino et that enteral fluconazole is well absorbed in critically ill al. added fluconazole to a selective digestive decontamination surgical patients [18]. Serum concentrations of fluconazole (SDD) regimen in critically ill medical and surgical patients to were above the highest minimum inhibitory concentrations assess the effect on candidal infections [15]. Mechanically (MIC) for C albicans and C parapsilosis isolates in all but ventilated patients were randomized during or after the third 4% of patients, but were below the median MIC obtained day of their ICU stay to either 100 mg of intravenous from strains of C glabrata. fluconazole daily (n=103) or placebo (n=101). Candida The issue of preemptive therapy was assessed in a study infections occurred less frequently in patients treated with by Piarroux and colleagues [19], which utilized a fluconazole prophylaxis compared with those who received retrospective control group. During the prospective period, placebo (5.8% vs. 16%, RR 0.35, 95% CI 0.11–0.94). Ninety patients who developed a corrected colonization index percent (9/10) of the cases of candidemia occurred in the of ≥0.4 were placed on fluconazole (intravenous 800 mg placebo group. In addition, there was a decrease in both the loading dose followed by 400 mg/day for 2 weeks). When frequency and intensity of colonization in the fluconazole arm. compared with the retrospective control group, patients Crude mortality was not different between placebo (41%) given preemptive fluconazole had a lower rate of proven and fluconazole (39%) patients. candidiasis (7% vs. 3.8%; p=0.03). No emergence of azole- In one of the largest single-center, randomized, double- resistant Candida spp. was noted in the treatment group. blind, placebo-controlled trials of antifungal prophylaxis, enteral fluconazole (800 mg loading dose followed by Posaconazole 400 mg/day) or placebo was administered in 260 critically ill Posaconazole is a newer, extended-spectrum, orally surgical ICU patients with an anticipated surgical ICU stay of bioavailable azole that shows good activity against several ≥3 days [16]. Therapy was continued until ICU discharge, pathogenic fungi, including species of Candida that are less and the mean length of stay was 5 days. Infection susceptible to fluconazole. A recent randomized, double- definitions were consistent with both European and National blind trial involving 600 patients with graft-versus-host Institutes of Health consensus groups [16]. The intent-to- disease assessed the efficacy of posaconazole for prophylaxis treat analysis showed that enteral fluconazole prophylaxis against invasive fungal infections [20]. This agent was found reduced the incidence of proven fungal infections from to be as effective as fluconazole in preventing invasive
38 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ANTIFUNGAL PROPHYLAXIS IN SURGICAL CRITICALLY ILL PATIENTS
Table 2. Summary of meta-analyses. Study Number Number Infection reduction Attributable mortality Overall mortality of trials of patients Cruciani et al. [23] 9 1226 RR 0.30, RR 0.25, RR 0.60, 95% CI 0.10–0.82 95% CI 0.08–0.80 95% CI 0.45–0.81 Shorr et al. [24] 4 626 OR 0.44, Not reported OR 0.87, 95% CI 0.27–0.72 95% CI 0.59–1.28 Playford et al. [25] 12 1606 RR 0.46, Not reported RR 0.76, 95% CI 0.31–0.68 95% CI 0.59–0.97 Fluconazole – – RR 0.47, Not reported RR 0.77, 95% CI 0.32–0.70 95% CI 0.56–1.07 Ketoconazole – – RR 0.30, Not reported RR 0.68, 95% CI 0.07–1.31 95% CI 0.42–1.12 Vardakas et al. [26] 6 847 OR 0.20, Not reported OR 0.74, 95% CI 0.13–0.32 95% CI 0.52–1.05
CI: confidence interval; OR: odds ratio; RR: relative risk. fungal infections (p=0.07), and the rate of proven reduction of fungal infections in critically ill patients by or probable invasive aspergillosis was lower in the fungal prophylaxis with an azole antifungal. Meta-analysis is posaconazole-treated group compared with that in the difficult in this group of clinical trials because patient fluconazole group (2.3% vs. 7.0%; p=0.006). However, at characteristics and dosing regimens vary greatly depending present, the paucity of an intravenous formulation of on the population studied (ICU versus ward, medical versus posaconazole limits its potential as a therapeutic option in surgical, severity of illness). In the majority of the trials, invasive fungal infections in critically ill patients. ketoconazole (in early trials) and fluconazole were studied. Definitions for outcomes vary between trials and could over- Newer agents or underestimate treatment effects. As a class of agents, the echinocandins are increasingly The first analysis, by Cruciani et al., examined nine utilized in the treatment of critically ill patients with proven clinical trials of 1226 patients in both surgical and trauma or suspected fungal infections. A decision analysis was ICUs [23]. The dose and antifungal agent used for recently conducted based on considerations of likely prophylaxis varied between studies, with some using pathogens, incidence of fungal infections, and signs of a ketoconazole (200 mg administered enterally) and others clinical infection that has not responded to antibacterial using fluconazole (100–400 mg administered either therapy [21]. The results suggested that caspofungin is the enterally or intravenously). Only two of the trials utilized most effective agent but is too costly for a base case a loading dose for fluconazole (800 mg). Some of the analysis. The details of this analysis need to be examined in studies examining the effect of ketoconazole excluded more detail. On closer inspection of the specific details, it patients with hepatic dysfunction or encephalopathy. would appear that, while very useful in conceptualizing There was significant heterogeneity in the definitions different agents and balancing issues of incidence and of fungal infections and outcome measures. The overall resistance, the clinical use proposed in this analysis is not risk of candidemia was significantly reduced by azole prophylaxis. Another echinocandin, micafungin, has been antifungal prophylaxis (RR 0.30, 95% CI 0.11–0.83; studied as an agent for prophylaxis in neutropenic patients p=0.02). Overall mortality was significantly reduced in and is approved in the US for this indication [22]. this analysis (RR 0.61, 95% CI 0.45–0.82; p=0.001), as was mortality attributable to fungal infections (RR 0.25, Meta-analyses on fungal prophylaxis in the 95% CI 0.08–0.80; p=0.019). According to this meta- critically ill analysis, the NNT to prevent one episode of candidemia There have been four published meta-analyses and one was 27.9 patients (95% CI 24.0–44.6). Overall mortality Cochrane review examining clinical trials of antifungal and mortality attributable to fungal infections also had prophylaxis in the critically ill patient population (Table 2) low NNTs (11.5 [95% CI 7.5–25.0] and 35.2 [95% CI [23–26]. Many of these studies show some effect on the 30.6–71.2], respectively).
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 39 ED HORN, SANDRA M SWOBODA, AND PAMELA A LIPSETT
In 2005, Shorr and colleagues re-examined the clinical superficial). The incidence of mortality was not significantly trials on the use of fluconazole for prophylaxis [24]. In this reduced by antifungal prophylaxis (OR 0.74, 95% CI meta-analysis, only the efficacy of fluconazole in critically ill 0.52–1.05). Analyses conducted in subsets of studies surgical patient populations was examined. This included (fluconazole prophylaxis, studies that excluded transplant four trials (previously reviewed) with 626 patients in total patients) revealed similar findings. In the secondary analysis, [13–16]. Again, doses and routes of fluconazole mortality was found to be significantly lower in the azole- administration varied from 100–400 mg/day (either treated patients (n=918; OR 0.68, 95% CI 0.49–0.95). intravenously or enterally), with two trials utilizing loading Adverse events were not different between groups (n=499; doses of 800 mg on the day of enrollment. The primary OR 1.36, 95% CI 0.86–2.15). Hepatotoxicity was the reason focus of these studies was the impact of fluconazole on for discontinuation of the study drug in only five patients candidal infections and mortality. Fluconazole was shown to in pooled trials. The authors’ conclusions echoed those of significantly reduce the incidence of fungal infections (odds previous meta-analyses: fluconazole reduced the incidence of ratio [OR] 0.44, 95% CI 0.27–0.72) but not mortality (OR infection, but not all-cause mortality. 0.87, 95% CI 0.59–1.28). The lack of effect on mortality These four meta-analyses, all published recently, could be multifactorial and attributed to small sample size. demonstrate the problems associated with both the topic of Based on this meta-analysis, fluconazole prophylaxis appears antifungal prophylaxis, and the use of meta-analysis to help to be successful for the prevention of invasive fungal define an important clinical issue. As noted previously, the infections in critically ill surgical patients. However, the definitions of an at-risk patient population and of what authors advise against the widespread use of fluconazole constitutes a fungal infection in non-neutropenic surgical prophylaxis because of the possible emergence of resistance patients have not been established. The trials are generally and non-albicans species. small, single-institution trials, primarily but not exclusively The third meta-analysis, by Playford et al., included studying the use of azoles for prophylaxis. The endpoints of 12 trials (1606 patients) that utilized either ketoconazole the clinical trials are not uniform. Thus, even if the use of (n=4) or fluconazole (n=8) for antifungal prophylaxis [25]. prophylaxis is reasonable, the agent, route, and dose are not Both patient population and outcomes varied among these clearly established. The above meta-analyses demonstrate trials. Total mortality, reported in 11 trials, was not that the results of an analysis will greatly depend on which significantly reduced by fluconazole (RR 0.77, 95% CI clinical trials are included and which are not. 0.56–1.07). No heterogeneity was noted in reporting this In summary, these studies conclude that antifungal outcome. When the four trials that examined ketoconazole prophylaxis in the adult critically ill surgical patient is likely to were included, there was a significant reduction in mortality reduce the incidence of fungal infections, but does not of almost 25% (RR 0.76, 95% CI 0.59–0.97). Invasive appear to affect mortality. fungal infections (reported in 10 of the trials) were significantly reduced by fluconazole (RR 0.47, 95% CI Potential problems associated with prophylaxis 0.32–0.71) but not by ketoconazole (RR 0.30, 95% CI There are several concerns surrounding routine prophylaxis 0.07–1.31). Overall risk reduction was almost 50% (RR with azole antifungal agents, including the emergence of 0.46, 95% CI 0.31–0.68). Other analyses of outcomes resistance of previously susceptible strains (C albicans), as well showed that pooled adverse events were not different in as an increase in non-albicans species, if given to all patients patients treated with either fluconazole (RR 0.73, 95% CI (i.e. not selecting high-risk patients). In a surgical ICU, 0.20–2.64) or ketoconazole (RR 1.24, 95% CI 0.17–8.74). Swoboda et al. examined fungal isolates pre- and post- The authors of this analysis suggest that antifungal fluconazole prophylaxis in high-risk patients [28]. Prior to prophylaxis with fluconazole should be considered in prophylaxis, C albicans and C glabrata were the predominant patients at increased risk of invasive fungal infections. strains that caused infection (45% and 30%, respectively). In The most recent meta-analysis, published by Vardakas the time period after fluconazole prophylaxis implementation, et al., examined six trials of azole prophylaxis in 847 critically based on the study by Pelz et al. [16], C glabrata was the ill surgical ICU patients [26]. Fluconazole was studied in five predominant strain causing candidemia (64%; p=0.05). trials, and ketoconazole in one trial. A secondary analysis on C albicans was also common, causing 21% of cases of mortality was conducted, which included the study by Jacobs candidemia. However, total non-albicans species were 55% et al. [27]. Antifungal prophylaxis was associated with (38/69) in “pre” period and 79% (11/14) in the “post” period significantly fewer overall fungal infections (OR 0.20, 95% (p=0.12). This study shows that, while fluconazole prophylaxis CI 0.13–0.32). These results did not differ according to the clearly reduced the incidence of fungal infections, there was type of fungal infection examined (invasive, candidemia, or not a significant shift to non-albicans species. One other
40 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ANTIFUNGAL PROPHYLAXIS IN SURGICAL CRITICALLY ILL PATIENTS
concern associated with azole prophylaxis is hepatotoxicity; 8. Normand S, Francois B, Darde ML et al. Oral nystatin prophylaxis of Candida spp. colonization in ventilated critically ill patients. Intensive Care Med 2005;31:1508–13. however, the meta-analyses showed no significant increases. 9. Savino JA, Agarwal N, Wry P et al. Routine prophylactic antifungal agents (clotrimazole, Thus, azole agents appear safe and well tolerated. ketoconazole, and nystatin) in nontransplant/nonburned critically ill surgical and trauma patients. J Trauma 1994;36:20–6. 10. Slotman GJ, Burchard KW. Ketoconazole prevents Candida sepsis in critically ill surgical Conclusion patients. Arch Surg 1987;122:147–51. 11. Sharpe MD, Ghent C, Grant D et al. Efficacy and safety of itraconazole prophylaxis for Surgical ICU patients are at increased risk of fungal infection fungal infections after orthotopic liver transplantation: a prospective, randomized, double- as a result of a variety of risk factors. A review of the blind study. Transplantation 2003;76:977–83. 12. Winston DJ, Pakrasi A, Busuttil RW. Prophylactic fluconazole in liver transplant recipients. evidence suggests that antifungal prophylaxis in this A randomized, double-blind, placebo-controlled trial. Ann Intern Med 1999;131:729–37. population decreases fungal infection rates by almost 50%, 13. Ables AZ, Blumer NA, Valainis GT et al. Fluconazole prophylaxis of severe Candida infection in trauma and postsurgical patients: a prospective, double blind, randomized, depending on the baseline incidence. Antifungal prophylaxis placebo-controlled trial. Infect Dis Clin Pract 2000;9:1066–72. should be reserved for the high-risk surgical patient, and its 14. Eggimann P, Francioli P, Bille J et al. Fluconazole prophylaxis prevents intra-abdominal candidiasis in high-risk surgical patients. Crit Care Med 1999;27:1066–72. use limited to the ICU stay only. Current data suggest that, 15. Garbino J, Lew DP, Romand JA et al. Prevention of severe Candida infections in when prophylaxis is used only in high-risk surgical patients, nonneutropenic, high-risk, critically ill patients: a randomized, double-blind, placebo- controlled trial in patients treated by selective digestive decontamination. Intensive Care that global shifts in Candida strains may not be seen. Med 2002;28:1708–17. 16. Pelz RK, Hendrix CW, Swoboda SM et al. Double-blind placebo-controlled trial of Further studies are needed to determine the impact of fluconazole to prevent candidal infections in critically ill surgical patients. Ann Surg prophylaxis on mortality and resistance patterns, both on a 2001;233:542–8. 17. Lipsett PA. Surgical critical care: fungal infections in surgical patients. Crit Care Med local and national scale. 2006;34:S215–24. 18. Pelz RK, Lipsett PA, Swoboda SM et al. Enteral fluconazole is well absorbed in critically ill Disclosures surgical patients. Surgery 2002;131:534–40. 19. Piarroux R, Grenouillet F, Balvay P et al. Assessment of preemptive treatment to prevent Dr Lipsett has received educational honoraria from Schering-Plough, severe candidiasis in critically ill surgical patients. Crit Care Med 2004;32:2443–9. speaker honoraria from Pfizer, and university grant support from 20. Ullmann AJ, Lipton JH, Vesole DH et al. Posaconazole or fluconazole for prophylaxis in severe graft-versus-host disease. N Engl J Med 2007;356:335–47. Astellas Pharma. Dr Horn and Ms Swoboda have no financial interests 21. Golan Y, Wolf MP, Pauker SG et al. Empirical anti-Candida therapy among selected to disclose. patients in the intensive care unit: a cost-effectiveness analysis. Ann Intern Med 2005;143:857–69. 22. van Burik JA, Ratanatharathorn V, Stepan DE et al. Micafungin versus fluconazole for prophylaxis against invasive fungal infections during neutropenia in patients undergoing References hematopoietic stem cell transplantation. Clin Infect Dis 2004;39:1407–16. Candida 1. Solomkin JS, Flohr AB, Quie PG et al. The role of in intraperitoneal infections. 23. Cruciani M, de Lalla F, Mengoli C. Prophylaxis of Candida infections in adult trauma and Surgery 88 1980; :524–30. surgical intensive care patients: a systematic review and meta-analysis. Intensive Care Med 2. Magill SS, Swoboda SM, Johnson EA et al. The association between anatomic site of 2005;31:1479–87. Candida colonization, invasive candidiasis, and mortality in critically ill surgical patients. 24. Shorr AF, Chung K, Jackson WL et al. Fluconazole prophylaxis in critically ill surgical Diagn Microbiol Infect Dis 55 2006; :293–301. patients: a meta-analysis. Crit Care Med 2005;33:1928–35. Candida 3. Pittet D, Monod M, Suter PM et al. colonization and subsequent infections in 25. Playford EG, Webster AC, Sorrell TC et al. Antifungal agents for preventing fungal Ann Surg critically ill surgical patients. 1994;220:751–8. infections in non-neutropenic critically ill and surgical patients: systematic review and 4. Eggimann P, Garbino J, Pittet D. Management of Candida species infections in critically ill meta-analysis of randomized clinical trials. J Antimicrob Chemother 2006;57:628–38. patients. Lancet Infect Dis 2003;3:772–85. 26. Vardakas KZ, Samonis G, Michalopoulos A et al. Antifungal prophylaxis with azoles in 5. Lipsett PA. Clinical trials of antifungal prophylaxis among patients in surgical intensive care high-risk, surgical intensive care unit patients: a meta-analysis of randomized, placebo- units: concepts and considerations. Clin Infect Dis 2004;39:S193–9. controlled trials. Crit Care Med 2006;34:1216–24. 6. Singhal S, Ellis RW, Jones SG et al. Targeted prophylaxis with amphotericin B lipid complex 27. Jacobs S, Price Evans DA, Tariq M et al. Fluconazole improves survival in septic shock: a in liver transplantation. Liver Transpl 2000;6:588–95. randomized double-blind prospective study. Crit Care Med 2003;31:1938–46. 7. Shah T, Lai WK, Gow P et al. Low-dose amphotericin for prevention of serious fungal 28. Swoboda SM, Merz WG, Lipsetta PA. Candidemia: the impact of antifungal prophylaxis in infection following liver transplantation. Transpl Infect Dis 2005;7:126–32. a surgical intensive care unit. Surg Infect (Larchmt) 2003;4:345–54.
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 41 Preemptive Antifungal Therapy in Critically Ill Surgical Patients
Frédéric Grenouillet1, Laurence Millon1, Gilles Blasco2, and Renaud Piarroux1 1Mycology Department and 2Surgical Intensive Care Unit, University Hospital, Besançon, France
The incidence of invasive candidiasis in critical care settings has increased exponentially over the past two decades. The crude mortality rates of these infections range from 35% to 60%. Most frequently, the source of invasive candidiasis is LEADING ARTICLE endogenous and the colonization stage plays a key role in the onset of the infection. Many studies have been performed to define the most pertinent risk factor criteria to help characterize patients who are at higher risk of candidiasis in the surgical intensive care unit (SICU). Thus, these patients could be given preventative treatment. Of these risk factors, isolation of Candida species from several body sites prior to infection seems to be one of the best predictors of infection. This review focuses on the use of colonization data in a strategy to prevent Candida infections in SICU patients, and on the preemptive approach, a concept that is midway between prophylaxis and curative treatment. J Invasive Fungal Infect 2007;1(2):42–9.
The incidence of invasive candidiasis in critical care settings spread of candidal colonization from the abdominal cavity has increased exponentially over the past two decades as a to other body sites prior to the occurrence of candidemia result of improvements in life-support systems. Candida was first described by Solomkin et al. [14,15], and it is isolates account for a rising proportion of nosocomial likely that this colonization is a prerequisite for invasive bloodstream infections, especially in surgical patients [1–3]. infection [16–19]. Although some studies have failed to These infections are particularly severe and their associated identify Candida colonization as a risk factor for invasive mortality rates, which range from 35–60%, are similar to candidiasis [3], isolation of Candida species from several those observed in patients with septic shock [2,4]. Most body sites prior to infection seems to be one of the best studies conclude that the mortality rate attributable to predictors of infection [2]. Pittet et al. showed that the candidemia is approximately 30% [5]. Many investigations intensity of colonization by Candida species was have clearly demonstrated that candidiasis lengthens patient significantly related to the risk of proven candidiasis. They stay in the intensive care unit (ICU) and the overall time proposed a calculation of two indices: a colonization index spent in hospital, and increases hospital charges [6–8]. (CI) and a corrected colonization index (CCI) with high Preventing the occurrence of these infections is positive predictive values [20]. In a “before and after” tempting, but treatments can be costly, both financially, intervention study, we recently described the potential and with regard to fungal resistance and side effects interest of systematic weekly CCI determination in SICU [9–11]. Consequently, studies have been performed to patients for implementation of a targeted preemptive define additional risk-factor criteria that could help to antifungal therapy [21]. This review will focus on the use select subgroups of patients from the high-risk surgical of colonization data (and other risk factors) in a strategy to intensive care unit (SICU) for treatment [3,9,12,13]. Taking prevent Candida infections in SICU patients, and on the the pathophysiology of invasive candidiasis into concept of a preemptive approach, midway between consideration could help to explore alternative modes of prophylaxis and curative treatment. preventing these infections. Candida species are frequently present as part of the mucosal microflora in the normal The preemptive concept human host, in particular, in the gastrointestinal tract. The There is considerable controversy among clinicians and mycologists over the definition of the term “preemptive”. Address for correspondence: Frédéric Grenouillet, Mycology According to Eggimann et al., preemptive therapy is defined Department, CHU Jean Minjoz, 25030 Besançon, France. as early antifungal treatment given to patients with several Email: [email protected] risk factors for invasive candidiasis and evidence of
42 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 PREEMPTIVE ANTIFUNGAL THERAPY IN CRITICALLY ILL SURGICAL PATIENTS
Figure 1. Conceptual basis of antifungal therapies in critical care settings
Antifungal therapies
Prophylactic Preemptive Empirical Curative
Targeted Targeted No colonization prophylactic empirical data needed
Proof of infection
Sign of sepsis
Candida colonization
Risk factors
significant Candida colonization. Prophylaxis refers to the closely resembles the preemptive approach used in lung antifungal therapy prescribed for patients who are at high transplant patients who have persistent Aspergillus airway risk for Candida infection (e.g. organ-transplant recipients colonization and are at a risk for aspergillosis [24]. and immunocompromised patients with expected or current Different terminologies used by our group to characterize long-term neutropenia) but have no evidence of antifungal therapies are shown in Fig. 1. Throughout this colonization [22]. Here, the definition of “preemptive” review, we will use the term “preemptive” for early, targeted would be independent of the clinical status of the patient. antifungal therapy, based on Candida colonization data, Ostrosky-Zeichner et al. define the term “preemptive” as regardless of the clinical status of the patient. early treatment of the infection based on clinical, laboratory, or radiological surrogate markers of disease in high-risk Risk factors for invasive candidiasis hosts, before clinical signs or symptoms of full-blown disease Many authors have extensively described the risk factors for develop [23]. For these investigators, Candida colonization invasive candidiasis. Table 1 reviews these, focusing on the is not a surrogate marker indicative of early infection, but independent factors in critically ill surgical patients as represents a risk factor at most. They favor the term determined by multivariate analysis [2,3,12,13,18,20,25–31]. “prophylaxis” or “targeted prophylaxis” for antifungal Most of these risks are common interventions or conditions therapies given to patients highly colonized with Candida that are seen in the intensive care setting. Unfortunately, and who have several risk factors; they do not include data are often lacking to determine whether there is a causal Candida colonization data in their development of relationship to the diseases or whether they are just preventive strategies in SICUs [12,23]. associated markers indicating severity of illness and other We are in agreement with the definition of Eggimann et predisposing conditions. al. as it includes high Candida colonization, a prerequisite to the development of infection, and several other risk factors Pathophysiology of invasive candidiasis [21,22]. Thus, the term “targeted prophylaxis” can be Most frequently, progressive colonization from the abdominal included in this definition as well as the notion of “targeted cavity to other body sites precedes candidemia. The empirical therapy”, because in each case, antifungal agents development of these infections from an exogenous source are implemented due to colonization by Candida. (via vascular accesses) is not frequent [14], with a proven Preemptive therapies prevent the development of infection endogenous source in 85–90% of candidemia [16,19]. The in patients who harbor a potential pathogen that is either at key role of the colonization stage in the development of a not-yet-infecting stage, or patients in whom the infection invasive candidiasis has been demonstrated in several studies is at an early stage. This approach to Candida infection dealing with surveillance of at-risk patients using mycological
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 43 FRÉDÉRIC GRENOUILLET, LAURENCE MILLON, GILLES BLASCO, AND RENAUD PIARROUX
Table 1. Risk factors for candidemia and invasive candidiasis. the infecting stage. While filamentation facilitates tissue Independent risk factors in critically ill surgical patients invasion, the two morphological forms, yeast and hyphae, Candida colonization [20,25] are responsible for infection [34]. Broad-spectrum antibiotics [12,25,26] Colonization as the main risk factor (In particular, the number of antibiotics and duration of antibiotic therapy) of candidiasis Central venous catheter [3,26] Pittet et al. studied the role of colonization as a predictive Acute renal failure and hemodialysis [3,12,25,26] factor for invasive candidiasis in ICU patients [20]. Colonization was defined as the isolation of Candida from at least three Prior surgery [3,13] samples from the same or different body sites during a period Total parenteral nutrition (TPN) [3,13,27] of at least 2 consecutive days. This study, conducted over a Severity of illness [18,28] 6-month period, included 29 colonized patients of the 650 Other described risk factors* [2,29–31] admitted to the ICU; 11/29 (38%) developed invasive Length of stay in surgical intensive care unit candidiasis whereas 18/29 (62%) remained colonized without Diabetes occurrence of any Candida infection. Mechanical ventilation A colonization index (CI) was calculated as the ratio of Multiple blood transfusions the number of Candida-positive samples to the total number Bladder catheter of analyzed samples and the CI average was statistically Recurrent, persistent gastrointestinal perforation higher in patients who developed candidal infection. The High-risk gastrointestinal surgery threshold of 0.5 was reached in a median number of 6 days Severe acute pancreatitis (range 2–21 days) before the diagnosis of invasive Solid organ transplantation (in particular liver, pancreatic, candidiasis. The sensitivity, specificity, and positive and and small-bowel transplant) negative predictive values of this CI were 100%, 69%, Corticosteroids 44%, and 100%, respectively. A corrected colonization Cancer, immunosuppressive drugs index (CCI) was also defined as the ratio of highly positive samples to the total number of analyzed samples. The Anti-histamine H2 receptor blockers Diarrhea sensitivity, specificity, positive and negative predictive values of CCI were all 100% when using a threshold of 0.4. *Other risk factors include risks determined by univariate analysis and those However, because the excellent predictive values of these described in non-critically ill surgical patients. two indices were not known at the time of the study, antifungal therapy was not implemented in patients before culture of samples from multiple body sites [16,19,20,25]. In occurrence of proven infection. most cases, genotyping analyses have shown that the Although the study of Pittet et al. is very well known, the infecting strain is the actual strain or one of the strains that role of colonization as the most important risk factor for colonizes the gastrointestinal tract [16,18,19,32]. infection is still under debate [20,23]. This is, in part, due to The first event in the colonization process is the ecological data from one multicenter study, which failed to identify modification of the endogenous florae, which can be induced colonization as a definite risk factor for candidiasis. The and/or amplified by diabetes, immunosuppression, broad- definition of colonization in this study was, however, based spectrum antibiotic therapy, or extended hospitalization only on urine and rectal colonization without investigation [3,20]. All of these factors alter the balance of endogenous of other body sites, and this may be partially responsible for microorganisms, and thus may favor the proliferation of the surprising result [3]. Candida and the subsequent invasion of gastrointestinal and Other studies argue that colonization is a key point. oropharyngeal mucosa. The second process is the translocation During clinical trials that have shown the benefit of of Candida through the mucosal barrier and tissue invasion, prophylaxis in ICU settings, there is often a decrease in occurring more frequently if the gastrointestinal mucosa Candida colonization in fluconazole-treated patients has been previously weakened by surgery, chemotherapy, or [29,35]. Another study underlined the high negative other cytotoxic treatments. The third event is the secondary predictive values of number of colonized body sites and of hematogenous dissemination, with possible formation of CI, which strongly suggest that in the absence of fungal micro-abscesses in tissues and organs [9,15,33]. colonization, infection is unlikely [36]. For C albicans, filamentation and lytic enzyme secretion It has been reported that CI determination in patients are often involved in the transition from the colonizing to with several other risk factors can be of help in predicting
44 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 PREEMPTIVE ANTIFUNGAL THERAPY IN CRITICALLY ILL SURGICAL PATIENTS
further candidiasis [37,38], although both the number of on statistical modelization, where prophylaxis would be colonized body sites and the CI have low predictive positive implemented only in patients with a risk of candidiasis of values in patients with a lower risk of candidiasis [31,36,39]. >10% [30]. In a recently published study, urinary, respiratory, and In order to select these patient subgroups, several studies rectum/ostomy Candida colonizations were significantly were performed to define risk factor-based clinical prediction associated with a higher risk of invasive candidiasis, whereas rules. Paphitou et al. showed in a single center, retrospective oropharyngeal and gastric colonizations were not [40]. study, that surgical/trauma intensive care patients with a These significant data need to be confirmed before reducing SICU stay of at least 4 days, who had any combination of surveillance screenings to only urine, respiratory tract, and diabetes, new-onset hemodialysis, parenteral nutrition, and rectum/ostomy. The ranking position given to colonization broad-spectrum antibiotics, had an invasive candidiasis rate data in care management clearly depends on the type of of 16%, while the rate in other patients was 5% [12]. From a patients screened, on the exact risk level for candidiasis, and multicenter retrospective study, Ostrosky-Zeichner et al. on body sites examined. created a prediction rule for candidiasis, which was a combination of two categories of risk factors for patients that Prophylaxis or preemptive therapy stayed in the unit for >3 days and were expected to stay for to prevent candidiasis an additional 2 days. This rule identified a subgroup of For many clinicians, preventing the occurrence of candidiasis 10.6% of ICU patients, having a risk for candidiasis of in SICU patients is very enticing. The effectiveness of approximately 10%; however, this rule was able to predict antifungal prophylaxis has been proven in neutropenic and only one-third of the candidiasis [48]. These rules need to be solid-organ transplant patients, and in selected subgroups of prospectively validated. More recently, a multicenter, patients with cancer. The use of prophylaxis in ICU non- prospective, observational study defined the “Candida neutropenic patients is still under debate. The latest score”, which would guide implementation of early guidelines from the Infectious Diseases Society of America antifungal treatment in non-neutropenic ICU patients. This state that institutions in which high rates of invasive score included multifocal Candida colonization, prior surgery, candidiasis in the adult or neonatal ICU persist despite parenteral nutrition (each factor with a value ≈1) and severe standard infection-control procedures should consider sepsis (value ≈2). According to these authors, patients with a fluconazole prophylaxis for carefully selected patients in “Candida score” of >2.5 should receive an early antifungal these care areas [41]. treatment (sensitivity 81%, specificity 74%). The “Candida Three recent meta-analyses focused on the use of score” is useful and applicable either for medical or surgical antifungal prophylaxis in the ICUs [42–44]. As no patients, and underlines the relative importance of fungal multicenter, randomized studies of antifungal agents versus colonization in the pathogenesis of candidiasis [13,39]. placebos exists, the authors of these analyses reviewed all A similar approach was previously developed by our group. monocenter studies dealing with this subject. The overall In a “before and after” intervention study of 2-year conclusions of the three meta-analyses were concordant. prospective and 2-year historical control cohorts, Piarroux et al. They all showed a significant decrease in the incidence of assessed a preemptive strategy to prevent candidiasis in SICU candidiasis, with an odds ratio of infection for patients with patients with a SICU stay of ≥5 days and several risk factors prophylaxis varying from 0.2 to 0.44, depending on the [21]. This strategy, based on systematic mycological screening particular studies included in the meta-analyses and on the at admittance and weekly thereafter, led to implementation of definition of infection. However, results concerning crude an early preemptive antifungal therapy in patients with a CCI and attribuable mortality rates were discordant; only of ≥0.4. The incidence of SICU-acquired proven candidiasis Cruciani et al. showed a significant decrease in these significantly decreased from 2.2% to 0% (p≤0.001; Fisher mortality rates with prophylaxis [42]. All of these analyses test). The main results of this study are presented in Table 2 also underlined the lack of data concerning the ecological and show that a targeted preemptive strategy can efficiently consequences of prophylactic use of antifungals in the ICU. prevent acquisition of proven candidiasis in SICU patients. The potential ecological impact of azole prophylactic use has This study underlines that CCI can be very useful for remained contentious since the first shifts toward azole- identifying subgroups of SICU patients exposed to several risk resistant isolates or species, such as C krusei or C glabrata, factors, in order that they can receive appropriate antifungal were demonstrated in several groups of patients who preemptive treatment [21]. This study also suggests that a CCI received fluconazole prophylaxis [45–47]. To reduce this risk, of <0.4 is probably an indication that no prophylactic or in addition to the exponential increase of antifungal ICU preemptive therapeutic intervention is necessary in clinically expenses, several authors have proposed a concept based stable, low-risk patients, even when the CI is ≥0.5 [21]. This
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 45 FRÉDÉRIC GRENOUILLET, LAURENCE MILLON, GILLES BLASCO, AND RENAUD PIARROUX
Table 2. Characteristics of patients included during the two periods of the “before and after” intervention study conducted by Piarroux et al. [21]. Characteristics of patients Retrospective cohort Prospective cohort p Without premptive strategy With preemptive strategy (n=455) (n=478) Age, mean years±SD 50.7±19.6 52.0±19.0 0.33 SAPS II score, mean points±SD 39.0±15.0 43.2±15.5 <0.001 SICU stay, median day (inter-quartile range) 15 (9–23) 12 (7–21) 0.001 Male/female ratio 295/160 331/147 0.15 Number of antibiotics 2.8±2.2 2.8±2.0 0.82 Antibiotic therapy duration, mean days±SD 9.4±9.1 8.8±7.3 0.25 Receipt of anti-anaerobics/no receipt 259/196 385/93 <0.001 Bacteremia 55 (12.1%) 49 (10.3%) 0.37 Underlying disorders Abdominal surgery 94 (20.7%) 109 (22.8%) 0.63 Vascular surgery 19 (4.2%) 32 (6.7%) 0.09 Severe neurotrauma and trauma 197 (43.3%) 197 (41.2%) 0.52 Neurosurgery 120 (26.4%) 106 (22.2%) 0.13 Others 25 (5.5%) 34 (7.1%) 0.31 Overall SICU mortality 76 (16.7%) 73 (15,3%) 0.55 Non-septic conditions 47 (10.3%) 49 (10,3%) 0.96 Septic conditions* 29 (6.4%) 24 (5.0 %) 0.37 Bacterial infections 8 (1.8%) 6 (1.3%) 0.53 Candidiasis 7 (1.5 %) 2**(0.4%) 0.10 Aspergillosis 2 (0.4%) 0.49 Undocumented sepsis† 14 (3.1%) 14 (2.9%) 0.89 Occurrence of proven candidiasis 32 (7%) 18 (3.8%) 0.03 Diagnosed at admittance (imported cases) 22 (4.8%) 18 (3.8%) 0.42 SICU-acquired†† 10 (2.2%) 0 <0.001
SAPS: Simplified Acute Physiology Score; SD: standard deviation; SICU: surgical intensive care unit. *Patients with sepsis, as defined by Bone et al. [49] at time of death; **Imported candidiasis cases; †No isolation of pathogen from blood culture or a naturally sterile body site despite clinical features evoking infection; ††Diagnosed in SICU or during the month after SICU discharge. confirms the poor positive predictive value of CI alone – a criteria are needed for identification of patients and units, in result with which other previous studies concur [36,50]. which systematic mycological surveillance and subsequent However, a CI value of ≥0.5 in a clinically unstable septic early antifungal treatment could be implemented. As patient justifies implementation of an antifungal therapy. As pointed out by Ostrosky-Zeichner, the “before and after” underlined by others [51], the integration of colonization data intervention assessment of this strategy suffers from the into patient care management requires standardized processing geographical limitations imposed by a single unit with a of peripheral mycological samples in laboratories, especially for defined baseline of disease, a specific patient base, and a the accurate quantification of yeasts. A management flowchart particular epidemiological makeup [52]. Obviously, the for fungal risk in SICU patients, including colonization index more frequent the occurrence of systemic candidiasis in a data and other risk factors, was proposed in 2003 by Eggiman given population and in a particular SICU, the more et al. [22]. Our approach is very similar; the care management pertinent a preemptive strategy should be. In contrast, in procedures we have used since 2004 are shown in Fig. 2. subclasses of patients with a very low risk of systemic candidiasis, a systematic surveillance of colonization Preemptive strategy in SICUs: what remains to followed by treatment of highly colonized patients should be done? be avoided, because it may be either inappropriate or not Several points need to be clarified before the preemptive cost-effective [9,31,39,51]. Thus, because of the relatively approach can be widely used in SICUs. Firstly, more precise rare occurrence of invasive candidiasis and the large
46 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 PREEMPTIVE ANTIFUNGAL THERAPY IN CRITICALLY ILL SURGICAL PATIENTS
Figure 2. Management flowchart for fungal risk in SICU patients at Besançon University Hospital, France.
SICU patient at risk for severe candidiasis
Colonization by Candida spp.
No Yes
High-risk patient subgroups Presence of several risk factors
Yes No No Yes
Antifungal prophylaxis No Sepsis?
Yes
Monitoring of Candida coloization (weekly) Antifungal Oropharyngeal and rectal swabs preemptive Gastric and tracheal aspirates, urine therapy Drain effuents, ostomy drainage
Corrected colonization index (CCI) ≥0.4 (= highly positive samples/total number of analyzed samples)
<0.4
Colonization index (CI) (= positive samples/total number of analyzed samples)
<0.5 ≥0.5
SICU: surgical intensive care unit. Patient subgroups with high risk for candidiasis (i.e. patients with recurrent gastrointestinal perforations or anastomotic leakages, liver transplants), for whom prophylaxis was demonstrated as beneficial, is reviewed elsewhere [20,29,43,44]. Risk factors are reviewed in Table 1. Adapted with permission from [22]. diversity of patient profiles in SICUs, only large multicenter Candida species (particularly C glabrata, C krusei) was noted studies are likely to provide the information needed for during the prospective period of our study (the first 2 years of further progress. Conceptual basis for design of such trials implementation of the preemptive strategy) [21]. Studies are were excellently described in two recent reviews [51,53], still in progress in the SICU in our hospital in order to assess particularly patient selection, study endpoints, and SICU fungal ecology. Data of >5 years of preemptive antifungal definition of infections. strategy are likely to help address this issue. Confirmation of Secondly, one of the major conceptual drawbacks of the these monocentric ecological data in a multicentric assessment preemptive strategy concept is that it could result in an will be needed to provide definitive conclusions. epidemiological shift in SICUs, due to a non-targeted overuse of Thirdly, the economic impact of preemptive strategy antifungal agents. However, no emergence of azole-resistant needs to be assessed. As stressed in our previous study, the
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 47 FRÉDÉRIC GRENOUILLET, LAURENCE MILLON, GILLES BLASCO, AND RENAUD PIARROUX
cost calculation of a care strategy is complex. The estimated therapy?” is still being asked [52]. We hope that the quest cost of one case of proven candidiasis is close to €35 000 for this “holy grail” will be realized in the near future. [7,8]. Candida colonization was asssociated with an additional €8000 in direct hospital costs for each SICU- Acknowledgements colonized patient [54]. Pharmacoeconomic analyses on The authors would like to thank Lois Rose for her assistance preemptive strategy should take into account costs induced in reviewing the manuscript. by fungal cultures and by adminstration of antifungals to highly colonized patients, and should compare both of these Disclosures with savings gained by infections that were prevented. No The authors have no relevant financial interests to disclose. dramatic increase in antifungal SICU expenses was shown in the SICU when a preemptive strategy was applied [21,55]. References For patients with sepsis, decisions concerning initiation of 1. Vincent JL, Bihari DJ, Suter PM et al. The prevalence of nosocomial infection in intensive care units in Europe. Results of the European Prevalence of Infection in Intensive Care antifungal therapy can be modified because colonization (EPIC) Study. EPIC International Advisory Committee. JAMA 1995;274:639–44. 2. Eggimann P, Garbino J, Pittet D. Epidemiology of Candida species infections in critically ill data are available, and this might change the cost non-immunosuppressed patients. Lancet Infect Dis 2003;3:685–702. calculation. It could eliminate the administration of 3. Blumberg HM, Jarvis WR, Soucie JM et al.; National Epidemiology of Mycoses Survey(NEMIS) Study Group. Risk factors for candidal bloodstream infections in surgical antifungal empirical treatment in septic patients when CI is intensive care unit patients: the NEMIS prospective multicenter study. The National <0.5. Precise data on individual yeast epidemiology could Epidemiology of Mycosis Survey. Clin Infect Dis 2001;33:177–86. 4. Jacobs S, Price Evans D, Tariq M et al. Fluconazole improves survival in septic shock: a also favor the use of fluconazole instead of costly broad- randomized double-blind prospective study. Crit Care Med 2003;31:1938–46. spectrum antifungal agents such caspofungin, voriconazole, 5. Falagas ME, Apostolou KE, Pappas VD. Attributable mortality of candidemia: a systematic review of matched cohort and case-control studies. Eur J Clin Microbiol Infect Dis or lipid-associated amphotericin B. 2006;25:419–25. Additional studies are required to answer the following 6. Zaoutis TE, Argon J, Chu J et al. The epidemiology and attributable outcomes of candidemia in adults and children hospitalized in the United States: a propensity analysis. questions: Is mycological monitoring of anatomical sites Clin Infect Dis 2005;41:1232–9. efficient enough to differentiate patients who are at high risk 7. Goff DA, Sierawski SJ, Fass RJ. Cost analysis of Candida infection among surgical invasive care units patients. Clin Drug Invest 1996;12:176–80. of developing invasive candidiasis? Which anatomical sites 8. Rentz AM, Halpern MT, Bowden R. The impact of candidemia on length of hospital stay, need to be investigated – only urine, respiratory tract, and outcome, and overall cost of illness. Clin Infect Dis 1998;27:781–8. 9. Rex JH, Sobel JD. Prophylactic antifungal therapy in the intensive care unit. Clin Infect Dis rectum as suggested by Maggill et al. – or additional sites 2001;32:1191–200. [40]? How should other risk factors be included to optimize 10. Rocco TR, Reinert SE, Simms HH. Effects of fluconazole administration in critically ill patients: analysis of bacterial and fungal resistance. Arch Surg 2000;135:160–5. preemptive strategy as suggested by Leon et al. [13]? While 11. Gleason TG, May AK, Caparelli D et al. Emerging evidence of selection of fluconazole- fluconazole is the drug of choice for preventative strategy tolerant fungi in surgical intensive care units. Arch Surg 1997;132:1197–201; discussion 1202. [51,53], how can patients who are highly colonized by 12. Paphitou NI, Ostrosky-Zeichner L, Rex JH. Rules for identifying patients at increased risk for candidal infections in the surgical intensive care unit: approach to developing fluconazole-resistant strains be better managed? Do they practical criteria for systematic use in antifungal prophylaxis trials. require “only” stringent surveillance and implementation of Med Mycol 2005;43:235–43. 13. Leon C, Ruiz-Santana S, Saavedra P et al. A bedside scoring system (“Candida score”) for early antifungal therapy when infection starts or should early antifungal treatment in nonneutropenic critically ill patients with Candida preemptive treatment based on other antifungals, i.e. colonization. Crit Care Med 2006;34:730–7. 14. Solomkin J, Flohr A, Quie P et al. The role of Candida in intraperitoneal infections. Surgery candins or recently developed azoles such as voriconazole 1980;88:524–30. and posaconazole, be implemented in these patients? Could 15. Vincent JL, Anaissie E, Bruining H et al. Epidemiology, diagnosis and treatment of systemic Candida infection in surgical patients under intensive care. Intensive Care Med serological markers, such as mannans or (1,3)-β-D-glucan, 1998;24:206–16. be used to improve the efficiency of a preemptive strategy 16. Voss A, Hollis RJ, Pfaller MA et al. Investigation of the sequence of colonization and candidemia in nonneutropenic patients. J Clin Microbiol 1994;32:975–80. based on colonization data [56,57]? 17. Pittet D, Monod M, Filthuth I et al. Contour-clamped homogeneous electric field gel electrophoresis as a powerful epidemiologic tool in yeast infections. Am J Med 1991;91:256–63S. Conclusion 18. Stephan F, Bah MS, Desterke C et al. Molecular diversity and routes of colonization of Candida albicans in a surgical intensive care unit, as studied using microsatellite markers. In summary, epidemiological and pathophysiological studies Clin Infect Dis 2002;35:1477–83. have demonstrated the role of colonization in the 19. Marco F, Lockhart SR, Pfaller MA et al. Elucidating the origins of nosocomial infections with Candida albicans by DNA fingerprinting with the complex probe Ca3. development of invasive candidiasis in SICU patients. Recent J Clin Microbiol 1999;37:2817–28. clinical trials argue in favor of early antifungal treatment of 20. Pittet D, Monod M, Suter PM et al. Candida colonization and subsequent infections in critically ill surgical patients. Ann Surg 1994;220:751–8. high-risk patients, if possible, even before the occurrence of 21. Piarroux R, Grenouillet F, Balvay P et al. Assessment of preemptive treatment to prevent signs of infection. However, before the preemptive approach severe candidiasis in critically ill surgical patients. Crit Care Med 2004;32:2443–9. 22. Eggimann P, Garbino J, Pittet D. Management of Candida species infections in critically ill can be widely used, further studies are needed to better non-immunosuppressed patients. Lancet Infect Dis 2003;3:772–85. define indications and settings for this promising therapeutic 23. Ostrosky-Zeichner L. Prophylaxis and treatment of invasive candidiasis in the intensive strategy. Thus, the classic rounds question “should ICU care setting. Eur J Clin Microbiol Infect Dis 2004;23:739–44. 24. Sole A, Morant P, Salavert M et al. Aspergillus infections in lung transplant recipients: risk patients receive antifungal prophylaxis or preemptive factors and outcome. Clin Microbiol Infect 2005;11:359–65.
48 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 PREEMPTIVE ANTIFUNGAL THERAPY IN CRITICALLY ILL SURGICAL PATIENTS
25. Wey SB, Mori M, Pfaller MA et al. Risk factors for hospital-acquired candidemia. A 43. Shorr AF, Chung K, Jackson WL et al. Fluconazole prophylaxis in critically ill surgical matched case–control study. Arch Intern Med 1989;149:2349–53. patients: a meta-analysis. Crit Care Med 2005;33:1928–35; quiz 1936. 26. Bross J, Talbot GH, Maislin G et al. Risk factors for nosocomial candidemia: a case–control 44. Vardakas KZ, Samonis G, Michalopoulos A et al. Antifungal prophylaxis with azoles in study in adults without leukemia. Am J Med 1989;87:614–20. high-risk, surgical intensive care unit patients: a meta-analysis of randomized, placebo- 27. Borzotta AP, Beardsley K. Candida infections in critically ill trauma patients: a retrospective controlled trials. Crit Care Med 2006;34:1216–24. case-control study. Arch Surg 1999;134:657–64; discussion 664–5. 45. Millon L, Manteaux A, Reboux G et al. Fluconazole-resistant recurrent oral candidiasis in 28. Pelz RK, Hendrix CW, Swoboda SM et al. Double-blind placebo-controlled trial of human immunodeficiency virus-positive patients: persistence of Candida albicans strains fluconazole to prevent candidal infections in critically ill surgical patients. Ann Surg with the same genotype. J Clin Microbiol 1994;32:1115–8. 2001;233:542–8. 46. Wingard JR, Merz WG, Rinaldi MG et al. Increase in Candida krusei infection among 29. Eggimann P, Francioli P, Bille J et al. Fluconazole prophylaxis prevents intra-abdominal patients with bone marrow transplantation and neutropenia treated prophylactically with candidiasis in high-risk surgical patients. Crit Care Med 1999;27:1066–72. fluconazole. N Engl J Med 1991;325:1274–7. 30. Ostrosky-Zeichner L, Pappas PG. Invasive candidiasis in the intensive care unit. Crit Care 47. Wingard JR, Merz WG, Rinaldi MG et al. Association of Torulopsis glabrata infections with Med 2006;34:857–63. fluconazole prophylaxis in neutropenic bone marrow transplant patients. Antimicrob 31. Lipsett PA. Surgical critical care: fungal infections in surgical patients. Crit Care Med Agents Chemother 1993;37:1847–9. 2006;34:S215–24. 48. Ostrosky-Zeichner L, Sable C, Sobel J et al. Multicenter retrospective development and 32. Botterel F, Desterke C, Costa C et al. Analysis of microsatellite markers of Candida albicans validation of a clinical prediction rule for invasive candidiasis (IC) in the intensive care setting. used for rapid typing. J Clin Microbiol 2001;39:4076–81. Focus on Fungal Infections 14. New Orleans, USA, 24–26 March 2004. Abstract 51. 33. Dean DA, Burchard KW. Surgical perspective on invasive Candida infections. World J Surg 1998;22:127–34. 49. Bone RC, Balk RA, Cerra FB et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Candida J Am Acad Dermatol 31 34 Odds FC. Pathogenesis of infections. 1994; :S2–5. Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 35. Garbino J, Lew DP, Romand JA et al. Prevention of severe Candida infections in 1992;101:1644–55. nonneutropenic, high-risk, critically ill patients: a randomized, double-blind, 50. Dubau B, Triboulet C, Winnock S. [Use of the colonization index]. Ann Fr Anesth Reanim placebo-controlled trial in patients treated by selective digestive decontamination. Intensive Care Med 2002;28:1708–17. 2001;20:418–20. 36. Pelz RK, Lipsett PA, Swoboda SM et al. The diagnostic value of fungal surveillance cultures 51. Calandra T, Marchetti O. Clinical trials of antifungal prophylaxis among patients in critically ill patients. Surg Infect (Larchmt) 2000;1:273–81. undergoing surgery. Clin Infect Dis 2004;39(Suppl 4):S185–92. 37. Charles PE, Doise JM, Quenot JP et al. Candidemia in critically ill patients: difference of 52. Ostrosky-Zeichner L. Prophylaxis or preemptive therapy of invasive candidiasis in the outcome between medical and surgical patients. Intensive Care Med 2003;29:2162–9. intensive care unit? Crit Care Med 2004;32:2552–3. 38. Eloy O, Marque S, Mourvilliers B et al. Contibution of Pittet’s index, antigen assay, IgM, 53. Lipsett PA. Clinical trials of antifungal prophylaxis among patientss in surgical critical care and total antibodies in the diagnosis of invasive candidiasis in intensive care unit. units: concepts and considerations. Clin Infect Dis 2004;39(Suppl 4):S193–9. J Mycol Med 2006;16:113–8. 54. Olaechea PM, Palomar M, Leon-Gil C et al. Economic impact of Candida colonization 39. Charles PE. Multifocal Candida species colonization as a trigger for early antifungal and Candida infection in the critically ill patient. Eur J Clin Microbiol Infect Dis therapy in critically ill patients: what about other risk factors for fungal infection? 2004;23:323–30. Crit Care Med 2006;34:913–4. 55. Triboulet C, Duquenne G, Breton Y et al. Routine screening for Candida colonization. 40. Magill SS, Swoboda SM, Johnson EA et al. The association between anatomic site of Critical Care 2004;8(Suppl 1):S114. Candida colonization, invasive candidiasis, and mortality in critically ill surgical patients. Diagn Microbiol Infect Dis 2006;55:293–301. 56. Ostrosky-Zeichner L, Alexander B, Kett D et al. Multicenter clinical evaluation of the (1–3) 41. Pappas PG, Rex JH, Sobel JD et al. Guidelines for treatment of candidiasis. Clin Infect Dis beta-D-glucan assay as an aid to diagnosis of fungal infections in humans. Clin Infect Dis 2004;38:161–89. 2005;41:654–9. 42. Cruciani M, de Lalla F, Mengoli C. Prophylaxis of Candida infections in adult trauma and 57. Sendid B, Jouault T, Coudriau R et al. Increased sensitivity of mannanemia detection tests surgical intensive care patients: a systematic review and meta-analysis. Intensive Care Med by joint detection of alpha- and beta-linked oligomannosides during experimental and 2005;31:1479–87. human systemic candidiasis. J Clin Microbiol 2004;42:164–71.
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 49 Diagnostic and Therapeutic Approaches to Fungal Infections in Critical Care Settings: Different Options but the Same Strategy
Rafael Zaragoza1 and Javier Pemán2 1Intensive Care Unit, Hospital Universitario Peset, and 2Department of Microbiology, Hospital Universitario La Fe, Valencia, Spain. LEADING ARTICLE
Invasive fungal infections, especially in the critical care setting, have become an excellent target for prophylactic, empiric, and preemptive therapy interventions due to their associated high morbidity, mortality rate, increased incidence, and healthcare costs. For these reasons, new studies and laboratory tests have been developed over the last few years in order to formulate an early therapeutic intervention strategy in an attempt to reduce the high mortality rate associated with these infections. In recent years, there have been important developments in antifungal pharmacotherapy. Evidence-based studies have shown the roles that the new antifungal drugs play in the treatment of invasive mycosis in seriously ill and complex patients, although data from critically ill patients are more limited and must be taken and extrapolated from studies based on the general hospital population. New antifungal agents have been analyzed in different clinical situations in critical care units, and the increasing number of non-Candida albicans species and the high mortality rates in these settings suggest that the application of an early broad spectrum antifungal therapy in critically ill patients with fungal infection should be recommended. J Invasive Fungal Infect 2007;1(2):50–8.
Epidemiology and etiology of invasive Rhizopus microsporus [13] have been described in intensive fungal infections care settings during the last few years, with an elevated Critical care medicine has advanced greatly in the past few morbidity and poor outcome from infection. decades. Patients with complex medical and surgical disorders are surviving longer due to equally complex medical and Mortality and associated risk factors surgical interventions. These often involve the “collateral Candida infections are associated with a significant mortality damage” of circumventing normal body defense mechanisms. rate, especially among critically ill patients [14]. The crude Approximately 10.4% of the episodes of infection in an mortality rate of these infections is high (40–75%), and the intensive care unit (ICU) are related to a Candida species, mortality rate attributable to candidemia has been estimated with the majority being nosocomial [1]. However, this rate to be 25–38% by several authors [3,15–17]. A recent review could be underestimated due to the fact that at least 4% of of matched cohort and case–control studies was undertaken critically ill patients who die in an ICU are found to have an in order to examine the mortality rate that could be linked to unexpected fungal infection during postmortem examination candidemia [18]. This analysis included studies that [2]. Furthermore, ICU admission itself has become an compared the mortality rate of patients with candidemia independent risk factor for the development of a Candida with that of matched patients who did not have candidemia. spp. infection [3,4]. Although less frequent, aspergillosis The data suggest that candidemia is indeed associated with (particularly in patients with chronic obstructive pulmonary a considerable mortality rate that can be attributed, at least disease [5–7]) and other emergent moulds and yeasts such as to some degree, to the infection itself and not only to the Trichosporon asahii [8], Saccharomyces cerevisiae [9,10], presence of comorbidity. Hansenula anomala [11], Dipodascus capitatus [12], or In recent years, the species of Candida that result in candidemia have shifted from a predominance of Candida Address for correspondence: Rafael Zaragoza, Intensive Care Unit, albicans to non-C albicans species (NCA). This increase has Hospital Universitario Dr. Peset, Avenida Gaspar Aguilar 90, 46017 been attributed to the use of fluconazole prophylaxis [19]. Valencia, Spain. Email: [email protected] Approximately half of the reported cases of candidemia are
50 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ANTIFUNGAL STRATEGIES IN CRITICAL CARE SETTINGS
now caused by NCA [3,17,20], and several publications reported the development of a bedside scoring system based have indicated that infections with these species have a on risk factors that allows early antifungal treatment when worse prognosis than those caused by C albicans [21–23]. candidemia is suspected in non-neutropenic ICU patients. This Other adverse outcome predictors described in candidemia “Candida score” was based on the predictive value of episodes are the stay in the ICU itself, renal failure, previously reported risk factors. Using a logistic regression thrombocytopenia, hematological malignancy, and the need analysis adjusting for possible confounding variables, the for mechanical ventilation or inotropic support [17,22]. In authors found that total parenteral nutrition, surgery, a Spanish multicenter study involving patients in ICUs of multifocal yeast colonization, and severe sepsis were 28 hospitals, an Acute Physiology and Chronic Health independently associated with a greater risk for proven Evaluation II (APACHE II) score of >20 at the time of candidal infection. The scores for the individual factors were: candidemia was associated with a higher mortality rate [16], whereas early treatment with antifungal medication and • Parenteral nutrition +0.908. removal of central venous catheters were factors protective • Surgery +0.997. against death [16,17]. Furthermore, it is well known that • Multifocal colonization +1.112. invasive fungal infections can be associated with inadequate • Severe sepsis +2.038. empirical antifungal treatment [24]. Two recent reports have demonstrated a strong association between a delay in the In this large cohort of critically ill patients, the authors start of antifungal therapy and an increase in hospital concluded that a “Candida score” of >2.5 (sensitivity 81%, mortality rates [25,26]; thus, it is necessary to recognize that specificity 74%) could accurately select patients who would time is of the utmost importance when considering the benefit from early antifungal treatment [33]. therapy of invasive fungal infections. Poor outcomes are, in part, associated with the difficulty in establishing the microbiological diagnosis at an early stage Empirical, preemptive therapy, and prophylaxis of the infection. Blood culture results are positive in only The early identification of risk factors for the development of 50% of invasive Candida and Fusarium infections and are very candidemia, such as peritonitis, abdominal surgery, previous rarely positive in cases of invasive aspergillosis. Cultures of administration of broad-spectrum antibiotics, parenteral bronchoalveolar lavagefluid or brushing specimensare positive in nutrition, multiple lumen catheters, prior Candida spp. <50% of subjects with invasive pulmonary aspergillosis. Finally, colonization, renal replacement therapy, and mechanical positive results of cultures of specimens from non-sterile body ventilation [15,27,28] has become the cornerstone of sites may be related to either colonization or disease, and empirical treatment of fungal infections in ICU settings in distinguishing between these can be difficult. Non-culture-based order to reduce the high mortality rate associated with these diagnostic tests may provide a useful adjunct to these more infections [29,30]. traditional approaches. Of these, detection of circulating (1,3)-β- Ostrosky-Zeichner and colleagues created a prediction rule D-glucan (BG), galactomannan, or C albicans germ tube for invasive candidiasis in a retrospective, multicenter study antibodies (CAGTA) have appeared promising and could be setting (12 units in the USA and Brazil). The rule requires that useful as a preemptive therapy guide in these patients [34–37]. a combination of one “major” and two “minor” criteria be Recent multicenter clinical trial results have demonstrated met, with the criteria reflecting common risk factors for that the BG assay can be used to measure serum BG in patients who stay in the ICU for >3 days and are expected to clinical specimens with a high specificity and positive stay for an additional 2 days. This rule can be applied to predictive value for subjects with proven or probable approximately 10% of patients who stay in the ICU for ≥4 invasive fungal infection compared with control subjects. days, and, approximately 10% of patients to whom this rule is This test appeared to be useful both as a single-point assay applied will develop proven or probable invasive candidiasis. for patients hospitalized with suspicion of a fungal infection Using a more broad definition of risk, patients with any or as a part of a surveillance strategy in high-risk individuals. combination of diabetes mellitus, new onset hemodialysis, use Serum BG cutoff values of 60 or 80 pg/mL seemed to be of total parenteral nutrition, or receipt of broad-spectrum optimal for this test. The BG assay had a high positive antibiotics had an invasive candidiasis rate of 16.6% predictive value for subjects infected with Candida (except compared with a rate of 5.1% for patients who were lacking C parapsilosis), Aspergillus, or Fusarium species. Moreover, these characteristics (p=0.001). Fifty-two percent of patients the performance of the assay did not seem to be staying ≥4 days in the ICU met this particular rule and the rule significantly affected by antifungal therapy. However, the captured 78% of the patients who eventually developed assay did not detect elevated levels of BG in subjects invasive candidiasis [31,32]. Recently, a Spanish group infected with Mucor, Rhizopus, or Cryptococcus species.
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 51 RAFAEL ZARAGOZA AND JAVIER PEMÁN
Figure 1. Antigen-antibody reaction (white arrows) enhanced Although previously published data failed to show a by indirect immunofluorescence showing a positive Candida reduction in mortality rates with fluconazole prophylaxis in Germ tube antibody result (400x). surgical critical care patients [40,41], a recent meta-analysis of azole prophylaxis of Candida infections in a heterogeneous population of trauma and surgical intensive care patients did demonstrate reduced rates of candidemia, attributable mortality, and overall mortality rates [42]. Thus, data must be interpreted carefully in the context of local practices prior to the application of universal prophylaxis in this setting.
Available antifungal agents and therapy options The past few years have brought exciting developments in antifungal pharmacotherapy. Evidence-based studies using the new antifungal agents are accumulating and these drugs are assuming important roles in the pharmacotherapy of invasive fungal infections in seriously ill and complex patients. However, data in these patients are more limited and must be
Courtesy of MD Moragues, J Pontón, and G Quindós. recovered from general hospital population studies. Voriconazole, the first available second-generation triazole, has been approved by the US Food and Drug Administration Furthermore, the results in patients undergoing hemodialysis and by the European Medicines Agency for the treatment of must be considered with caution due to the false-positive invasive aspergillosis, serious infections caused by Fusarium results described in this patient population [36]. and Scedosporium apiospermum fluconazole-resistant In 2006, the current authors’ group evaluated an invasive Candida infections (such as C krusei and C glabrata), immunofluorescence assay (IFA) for CAGTA detection and candidemia in non-neutropenic patients. (Candida albicans IFA immunoglobulin G; Vircell, Granada, A landmark trial showed that voriconazole had proven Spain) in a selected population of critically ill patients (Fig. 1) efficacy in the treatment of invasive aspergillosis [43], with [38]. Although there were no differences between CAGTA- the authors of the study concluding that treatment with this positive and -negative patients in terms of age, sex, agent resulted in a better clinical response, improved survival Sequential Organ Failure Assessment score, renal failure, and rate, and fewer serious adverse reactions than treatment hepatic failure, the intra-ICU mortality rate was significantly with amphotericin B. Moreover, voriconazole has been lower in patients who tested positive for CAGTA (25% vs. demonstrated to be an effective and well-tolerated treatment 65.2%; p=0.025). These results suggest that a strategy for refractory or less common invasive fungal infections [44]. based on early determination of CAGTA expression may In spite of these encouraging data on voriconazole, reduce the ICU mortality rate of patients with risk factors for the best first-line treatment for candidemia remains the development of invasive candidiasis. However, more controversial, especially in critically ill patients. Clinical studies are necessary to validate this approach in critical studies have shown that amphotericin B, fluconazole, care settings. caspofungin, and voriconazole have similar efficacy in the Piarroux et al. assessed the efficacy of a preemptive treatment of Candida bloodstream infections [45–48]. In antifungal therapy in preventing proven candidiasis in accordance with recent guidelines [49], many experts favor critically ill surgical patients, using a corrected colonization initial treatment with amphotericin B in severely ill or index (ratio of highly positive samples to the total numbers clinically unstable patients, but its renal toxicity could of samples cultured) to measure the intensity of Candida present a serious problem in these individuals, which may mucosal colonization [39]. In patients with a corrected preclude its use as first-line therapy [50,51]. While the colonization index of ≥0.4, early preemptive antifungal triazole, fluconazole, may be selected on the basis of its therapy with fluconazole caused a reduction in the incidence efficacy and safety [45,46], the increasing frequency of of ICU-acquired proven candidiasis (from 2.2% to 0%). patients infected with Candida strains that are resistant to However, it is possible that the widespread use of this this drug highlight the need for initial treatment with a approach could be limited by the capacity of the broader-spectrum agent – at least until the Candida species microbiology laboratory to analyze multiple samples. is identified – in order to avoid inadequate empirical
52 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ANTIFUNGAL STRATEGIES IN CRITICAL CARE SETTINGS
antifungal treatment and an associated increased mortality caspofungin associated with cyclosporin, the high cost of rate [52]. Thus, “de-escalation” therapy in these cases may these drugs, the difficulty of studying their action in vitro in be considered. De-escalation of antifungal therapy, a clinical laboratory (especially caspofungin), and the according to the definition proposed by Marin Kollef, can be worrying number of breakthrough candidemias caused by thought of as a strategy to balance the need to provide C parapsilosis (in the caspofungin-treated group in the study adequate initial antifungal treatment of high-risk patients of Mora Duarte et al. [47]). Furthermore, isolates from many with the avoidance of unnecessary antimicrobial utilization of the echinocandin-treatment failures published in the (which promotes resistance) using an early broad-spectrum literature had high minimum inhibitory concentrations, antifungal agent and subsequent switching to a narrower- especially against C parapsilosis. Conversely, due to their spectrum drug when mycological identification and safety profile and activity on biofilms, especially in renal susceptibility studies are provided [53]. dysfunction, all echinocandins have become first-line agents Another triazole, itraconazole, has recently been in patients with renal insufficiency admitted to an ICU or developed as an intravenous formulation that has a wide when artificial devices are related to the infection [59]. spectrum of activity. However, the absence of clinical trials in Successful therapy with compassionate use of Candida infections (particularly in ICU settings), issues of voriconazole for the treatment of candidemia and invasive drug interactions, and the limited data regarding the use of candidiasis in patients intolerant or refractory to other cyclodextrin in its formulation are factors that may restrict its antifungal agents has been reported [60]. This study showed use as a first-line therapy in critically ill patients. In addition, that voriconazole may be a suitable agent for the salvage 46–53% of C glabrata and 31% of C krusei isolates are treatment of invasive candidiasis, even in the setting of resistant to itraconazole [49]. previous azole exposure and C krusei infection. These A novel class of antifungal compounds that target the findings have been confirmed by two Spanish groups glucan synthesis enzyme complex (echinocandins) has [61,62]. Another observational Spanish multicenter study been developed in the last few years. Currently, three has been performed to assess the clinical use and tolerability echinocandins are available: caspofungin, micafungin, and of voriconazole in daily practice in the ICU setting for the anidulafungin. Published reports suggest that caspofungin is treatment of fungal infection in critically ill patients [63]. The equivalent to standard therapy (amphotericin B) in terms of key contribution of this study is the description of the efficacy in the treatment of invasive Candida infections [47]. scenario in which voriconazole is used in daily practice in the Anidulafungin is the only antifungal compound that has ICU. Notably, voriconazole was frequently used for the been able to demonstrate superiority over a comparator salvage treatment of invasive fungal infections in subjects (fluconazole) in a recent invasive candidiasis clinical trial, with previous azole exposure. The typical patient was a although these results must be considered with caution due middle-aged man with an underlying medical disease, in to the fact that the study had been powered a priori for particular, active malignancy or chronic bronchitis, who had equivalency and has not yet been published. However, the received treatment with antibiotics, corticosteroids, or efficacy and safety profile of anidulafungin indicate that it chemotherapeutic drugs prior to admission to hospital, and should be readily considered as a first-line option for the with a high APACHE II score on admission to the ICU. The treatment of invasive candidiasis [54,55]. prescription of voriconazole was based on the presence of Micafungin has become the second available agent of documented invasive fungal infection previously treated the echinocandin class that is approved for use in clinical with other antifungal drugs. C albicans and Aspergillus practice (currently, only in Japan and the US for treatment of fumigatus were the most common pathogens and the use of esophageal candidiasis and prophylaxis in subjects with voriconazole was effective in 50% of cases. The drug was neutropenia). This agent shares an identical spectrum of in well tolerated and there were no treatment discontinuations vitro activity against C albicans, non-albicans species of due to adverse events. Candida, and Aspergillus species with caspofungin [56]. Results from the first randomized, prospective, multicenter Although there are descriptions of the use of micafungin study in non-neutropenic patients with candidemia who were in candidemia [57,58], to date, these studies have not treated with voriconazole or amphotericin B deoxycholate been published. followed by fluconazole demonstrated that the drugs were It should be noted that the administration of equivalent with regard to efficacy and mortality rates [48]. echinocandins could have several limitations in ICU settings. There was a high proportion of infection due to NCA species Factors that urge caution include the lack of data regarding (55%), with a similar distribution of the species between the the number of patients admitted in the ICU in the study of two treatment groups. Successful response rates were similar in Mora-Duarte et al. [47], the potential hepatotoxic effect of the voriconazole and amphotericin B/fluconazole arms;
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 53 RAFAEL ZARAGOZA AND JAVIER PEMÁN
however, for C tropicalis infections, the proportion of patients identification of an increasing number of NCA isolates [19]. responding to treatment was substantially higher in the group Consequently, the application of an early tailored therapy in treated with voriconazole. This is in spite of the fact that these critically ill patients with fungal infection is recommended strains were susceptible in vitro to amphotericin B. Although [68]. For this reason, voriconazole (due its broad spectrum renal dysfunction was significantly lower in the voriconazole- of action and good safety profile in ICU settings), treated group, the incidence of visual disturbances was slightly caspofungin, and anidulafungin (particularly in renal higher. These minor side effects are usually transient and dysfunction) could be attractive options in the application of resolve after the patient has become tolerant to the drug or this strategy in critically ill patients. Finally, the antifungal treatment is discontinued. However, in critically ill patients, it is drug selection must be based on the individual difficult to detect this complication because most are sedated characteristics of the patient, and should particularly focus and mechanically ventilated. The results of this study can be on the presence of renal or hepatic failure and possible easily applied to critically ill patients as approximately half of interactions with other drugs. A strategy for de-escalation the patients included were admitted to an ICU. The only (tailoring) of antifungal drugs, and the diagnosis limitation to the use of intravenous administration of status–treatment protocol are proposed in Figs. 2 and 3. voriconazole in these patients could be the accumulation and toxicity of cyclodextrin in severe renal dysfunction, although Combination therapy there are no data addressing this possibility in patients The availability of new antifungal agents with single undergoing renal replacement therapy. There is no requirement mechanisms of action and improved tolerability has widened for the adjustment of oral voriconazole in patients with renal the possibilities for the use of combination antifungal therapy dysfunction, although severe hepatic dysfunction could limit its (combination of two antifungal drugs) for difficult-to-treat, use. Potential drug interactions should also be kept in mind, opportunistic mycoses. Furthermore, additive in vitro although to date, no dose-dependent relationship for interactions of voriconazole and echinocandins have been voriconazole has been observed; however, a wide inter-patient observed. Few randomized clinical trials have examined the variability exists with contrasting predictive intra-individual role of this type of therapy for invasive mycoses, and no kinetics. Nevertheless, voriconazole must be discontinued in prospective, randomized trial of antifungal agent combinations cases of interactions with rifampicin or sirolimus [64]. has been completed for invasive mould infections. The results Encouraging clinical experience suggests that of in vitro studies and animal models suggest that combination voriconazole may be a new therapeutic alternative in therapy with azoles and echinocandins may have additive critically ill patients, not only as salvage treatment but also activity against Aspergillus species and suggest a great as an additional first-line option in suspected or proven potential for combination therapy, confirming the need for Candida, and as a first choice in Aspergillus infection. further investigation [69]. However, the possible benefit of Posaconazole, a new extended-spectrum triazole, has combination therapy with voriconazole for disseminated in vitro activity against a variety of pathogenic fungi, including cryptococcosis and invasive candidiasis (or other emerging Aspergillus, Candida, Cryptococcus, and Histoplasma. This new yeasts) needs to be elucidated. At present, a combination of agent has been studied extensively in a variety of animal amphotericin B plus flucytosine, and monotherapy with models of mycoses and appears to be a useful oral agent in fluconazole, an echinocandin, voriconazole, or amphotericin B open trials for mycoses, including refractory fungal infections may be more desirable in these settings. [65,66]. In patients undergoing chemotherapy for acute The use of other classes of drugs with antifungals – as a myelogenous leukemia or myelodysplastic syndrome, combination therapy – should also be considered. For posaconazole prevented invasive fungal infections more instance, a human recombinant monoclonal antibody effectively than either fluconazole or itraconazole and improved against heat shock protein 90, Mycograb® (NeuTec Pharma, the overall survival rate [67]. However, due to the lack of Manchester, UK), has been shown to act synergistically with availability of an intravenously administered formulation, it is amphotericin B against a broad spectrum of Candida species very difficult to consider this drug as a therapeutic option in in vitro [70]. Subsequently, a double-blind, randomized invasive candidiasis in critically ill patients at present. study was conducted to determine whether lipid-associated The principal advantages and disadvantages of systemic amphotericin B plus Mycograb was superior to lipid- antifungal agents in the ICU setting are summarized in Table 1. associated amphotericin B plus placebo in patients with culture-confirmed invasive candidiasis [71]. Patients received A similar and uniform strategy a lipid-associated formulation of amphotericin B plus a 5-day A recent publication shows a shift toward the use of course of Mycograb or placebo, having been stratified on antifungal drugs other than fluconazole due to the the basis of Candida species (C albicans versus NCA
54 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ANTIFUNGAL STRATEGIES IN CRITICAL CARE SETTINGS
Table 1. Antifungal therapy options in the intensive care unit setting. Antifungal Drug Potential advantages Potential disadvantages class
Polyenes Amphotericin B Clinical efficacy [45,46], Limited use in renal failure [50], deoxycholate broad-spectrum activity [49]. other adverse events [49]. Amphotericin B Broad-spectrum activity [49], Only assayed in one large lipid complex possibility of combination with other retrospective study [72], limited use antifungal drugs and Mycograb® [71], in renal failure, other adverse less nephrotoxicity than amphotericin B events [49]. deoxycholate [49]. Liposomal Clinical efficacy, broad-spectrum activity, Limited use in renal failure, other amphotericin B possibility of combination with other adverse events [49]. antifungal drugs and Mycograb® [71], less nephrotoxicity than amphotericin B, deoxycholate, and amphotericin B lipid complex [49]. Echinocandins Anidulafungin Superiority in clinical trials [54], Pending publication, broad-spectrum activity, good safety profile, C parapsilosis breakthrough use in renal failure, possibility of combination candidemia [55]. with other antifungal drugs [54,55]. Caspofungin Clinical efficacy, broad-spectrum activity, C parapsilosis breakthrough good safety profile, use in renal failure [47], candidemia, No data in intensive care possibility of combination with other unit patients, pending in vitro antifungal drugs [68]. susceptibility study methodology [68]. Micafungin Broad-spectrum activity, good safety profile, C parapsilosis breakthrough use in renal failure, possibility of combination candidemia [56]. To date, candidemia with other antifungal drugs [56]. studies have been published only in abstract form [57,58]. Not licensed in Europe. Azoles Fluconazole Clinical efficacy, good safety profile [45,46], Poor activity against C glabrata low cost, possibility of combination with and no activity against C krusei other antifungal drugs [73]. [49]. Itraconazole Clinical efficacy, good safety profile [49], Absence of clinical trials in low cost. Candida infections, drug–drug interactions, limited data regarding the use of cyclodextrin in its formulation, poor activity against C glabrata and C krusei [49,68]. Posaconazole Clinical efficacy, broad-spectrum activity, Only oral formulation available, good safety profile, possibility of efficacy demonstrated in one open combination with other antifungal drugs, trial [66]. activity against Mucor spp. [65,66]. Voriconazole Clinical efficacy [48], broad-spectrum activity, Intravenous formulation should good safety profile, possibility of combination be restricted in severe renal with other antifungal drugs, Aspergillus first failure, drug–drug interactions [64]. option treatment, activity against emergent fungi [64]. species). A favorable overall response, defined as a complete combination therapy group (odds ratio [OR] 5.8, 95% clinical and mycological response, with resolution of all signs confidence interval [CI] 2.41–13.79; p<0.001). A greater and symptoms of candidiasis and culture-confirmed proportion of the combination therapy group, compared eradication of the pathogen, by day 10 was obtained for 29 with the amphotericin B alone group, met individual efficacy of 61 (48%) patients in the amphotericin B alone group, criteria including clinical response (86% vs. 52%, OR 5.4, compared with 47 of 56 (84%) patients in the Mycograb 95% CI 2.21–13.39; p<0.001), mycological response (89%
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 55 RAFAEL ZARAGOZA AND JAVIER PEMÁN
Figure 2. Hospital algorithm for antifungal drug use based on a de-escalation strategy for suspected or proven invasive candidiasis.
Previous azole prophylaxis, and/or previous documented C krusei or C glabrata infection, and/or hemodynamic instability, and/or multiple colonization by non-C albicans species and/or ICU admission
Yes No
Consider broad-spectrum activity coverage
• Amphotericin B-lipid complex: 3–5 mg/kg/day • Fluconazole: 400–800 mg/day intravenously. • Liposomal amphotericin B: 3 mg/kg/day Consider sequential therapy to oral • Caspofungin: 70 mg/24 h (day 1) + 50 mg/day route (400 mg/day) • Anidulafungin: 200 mg/24 h (day 1) + 100 mg/day • Voriconazole inteavenously: 6 mg/kg/12 h (day 1) + 4 mg/kg/12 h. Consider sequential therapy to oral route (200 mg twice daily)
Consider de-escalation if possible
Figure 3. Proposed antifungal treatment of invasive candidiasis based on diagnostic stage in critical care settings.
Early diagnostic test Risk factors Diagnosis
Treatment Prophylaxis Preemptive Empirical based on treatment treatment positive culture results
Start fluconazole in Start early treatment with voriconazole, Apply de-escalation strategy surgical high-risk patients amphotericin B, or echinocandins and if possible, in accordance with if risk factors are present maintain until resolution of signs susceptibility test results and symptoms (at least 14 days from the start ot treatment)
vs. 54%, OR 7.1, 95% CI 2.64–18.94; p<0.001), Candida- Nevertheless, further clinical trials are required to clarify this attributable mortality rate (4% vs. 18%, OR 0.2, 95% CI issue in intensive care settings. 0.04–0.80; p=0.025), and rate of culture-confirmed clearance of the infection (hazard ratio 2.3, 95% CI Concluding remarks 1.4–3.8; p=0.001). These results underscore the potential Invasive fungal infections, especially in the critical care for the use of combination therapy in critically ill patients. setting, have become an excellent target for prophylactic,
56 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ANTIFUNGAL STRATEGIES IN CRITICAL CARE SETTINGS
empiric, and preemptive therapy interventions due to the high 23. Morgan J, Meltzer MI, Plikaytis BD et al. Excess mortality, hospital stay, and cost due to candidemia: a case-control study using data from population-based candidemia morbidity and mortality rate associated with these infections, surveillance. Infect Control Hosp Epidemiol 2005;26:540–7. 24. Garnacho-Montero J, Garcia-Garmendia JL, Barrero-Almodovar A et al. Impact of their increased incidence, and their attendant healthcare costs. adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive Early diagnosis and treatment are associated with a better care unit with sepsis. Crit Care Med 2003;31:2742–51. 25. Morrell M, Fraser VJ, Kollef MH. Delaying the empiric treatment of candida bloodstream prognosis in invasive candidiasis. Fluconazole, amphotericin B, infection until positive blood culture results are obtained: a potential risk factor for hospital echinocandins, and voriconazole are the antifungal agents mortality. Antimicrob Agents Chemother 2005;49:3640–5. 26. Garey KW, Rege M, Pai MP et al. Time to initiation of fluconazole therapy impacts currently available to treat invasive candidiasis in critically ill mortality in patients with candidemia: a multi-institutional study. Clin Infect Dis patients. The antifungal drug selection must be based on the 2006;43:25–31. 27. Alvarez-Lerma F, Palomar M, Leon C et al. Colonización y/o infección fúngica en unidades individual characteristics of the patient. A tailored therapy de cuidados intensivos. Estudio multicéntrico de 1.562 pacientes. Med Clin (Barc) (de-escalation) must also be considered in ICU settings. 2003;121:161–6. 28. Blumberg HM, Jarvis WR, Soucie JM et al. Risk factors for candidal bloodstream infections in surgical intensive care unit patients: the NEMIS prospective multicenter study. The National Epidemiology of Mycosis Survey. Clin Infect Dis 2001;33:177–86. Disclosures 29. Ibanez-Nolla J, Nolla-Salas M, Leon MA et al. Early diagnosis of candidiasis in non- The authors have no relevant financial interests to disclose. neutropenic critically ill patients. J Infect 2004;48:181–92. 30. Garnacho-Montero J, León C, Almirante B et al. Recomendaciones terapéuticas para infecciones fúngicas en el paciente crítico no neutropénico. Conferencia de consenso. References Conclusiones. Med Intensiva 2005;Suppl 3:43–52. 31. Paphitou NI, Ostrosky-Zeichner L, Rex JH. Rules for identifying patients at increased risk 1. Alberti C, Brun-Buisson C, Burchardi H et al. Epidemiology of sepsis and infection in ICU for candidal infections in the surgical intensive care unit: approach to developing practical patients from an international multicentre cohort study. Intensive Care Med Med Mycol 43 2002;28:108–21. criteria for systematic use in antifungal prophylaxis trials. 2005; :235–43. 2. Dimopoulos G, Piagnerelli M, Berre J et al. Post mortem examination in the intensive care 32. Ostrosky-Zeichner L. Prophylaxis and treatment of invasive candidiasis in the intensive unit: still useful? Intensive Care Med 2004;30:2080–5. care setting. Eur J Clin Microbiol Infect Dis 2004;23:739–44. 3. Tortorano AM, Peman J, Bernhardt H et al. Epidemiology of candidaemia in Europe: results 33. Leon C, Ruiz-Santana S, Saavedra P et al. A bedside scoring system (“Candida score”) of 28-month European Confederation of Medical Mycology (ECMM) hospital-based for early antifungal treatment in nonneutropenic critically ill patients with Candida surveillance study. Eur J Clin Microbiol Infect Dis 2004;23:317–22. colonization. Crit Care Med 2006;34:730–7. 4. Puzniak L, Teutsch S, Powderly W et al. Has the epidemiology of nosocomial candidemia 34. Hope WW, Walsh TJ, Denning DW. Laboratory diagnosis of invasive aspergillosis. Lancet changed? Infect Control Hosp Epidemiol 2004;25:628–33. Infect Dis 2005;5:609–22. 5. Janssen JJ, Strack van Schijndel RJ, van der Poest Clement EH et al. Outcome of ICU 35. Quindos G, Moragues MD, Ponton J. Is there a role for antibody testing in the diagnosis treatment in invasive aspergillosis. Intensive Care Med 1996;22:1315–22. of invasive candidiasis? Rev Iberoam Micol 2004;21:10–4. 6. Meersseman W, Vandecasteele SJ, Wilmer A et al. Invasive aspergillosis in critically ill 36. Ostrosky-Zeichner L, Alexander BD, Kett DH et al. Multicenter clinical evaluation of the patients without malignancy. Am J Respir Crit Care Med 2004;170:621–5. (1–>3) beta-D-glucan assay as an aid to diagnosis of fungal infections in humans. Clin Infect Dis 2005;41:654–9. 7. Garnacho-Montero J, Amaya-Villar R, Ortiz-Leyba C et al. Isolation of Aspergillus spp. from the respiratory tract in critically ill patients: risk factors, clinical presentation and 37. Moragues MD, Ortiz N, Iruretagoyena JR et al. Evaluation of a new commercial test outcome. Crit Care 2005;9:R191–9. (Candida albicans IFA IgG) for the serodiagnosis of invasive candidiasis. Enferm Infecc Microbiol Clin 2004;22:83–8. 8. Wolf DG, Falk R, Hacham M et al. Multidrug-resistant Trichosporon asahii infection of nongranulocytopenic patients in three intensive care units. J Clin Microbiol 38. Zaragoza R, Peman J, Ramirez P et al. Clinical and management utility of Candida germ 2001;39:4420–5. tube antibody detection in critically ill patients. Preliminary data. 46th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco Abstract 9. Lherm T, Monet C, Nougiere B et al. Seven cases of fungemia with Saccharomyces M-1618. 2006. boulardii in critically ill patients. Intensive Care Med 2002;28:797–801. 39. Piarroux R, Grenouillet F, Balvay P et al. Assessment of preemptive treatment to prevent 10. Munoz P, Bouza E, Cuenca-Estrella M et al. Saccharomyces cerevisiae fungemia: an severe candidiasis in critically ill surgical patients. Crit Care Med 2004;32:2443–9. emerging infectious disease. Clin Infect Dis 2005;40:1625–34. 40. Eggimann P, Francioli P, Bille J et al. Fluconazole prophylaxis prevents intra-abdominal 11. Kalenic S, Jandrlic M, Vegar V et al. Hansenula anomala outbreak at a surgical intensive candidiasis in high-risk surgical patients. Crit Care Med 1999;27:1066–72. care unit: a search for risk factors. Eur J Epidemiol 2001;17:491–6. 41. Sandven P, Qvist H, Skovlund E et al. Significance of Candida recovered from 12. Ersoz G, Otag F, Erturan Z et al. An outbreak of Dipodascus capitatus infection in the ICU: intraoperative specimens in patients with intra-abdominal perforations. Crit Care Med three case reports and review of the literature. Jpn J Infect Dis 2004;57:248–52. 2002;30:541–7. 13. Maravi-Poma E, Rodriguez-Tudela JL, de Jalon JG et al. Outbreak of gastric mucormycosis 42. Cruciani M, de Lalla F, Mengoli C. Prophylaxis of Candida infections in adult trauma and associated with the use of wooden tongue depressors in critically ill patients. Intensive surgical intensive care patients: a systematic review and meta-analysis. Intensive Care Med Care Med 2004;30:724–8. 2005;31:1479–87. 14. Leleu G, Aegerter P, Guidet B. Systemic candidiasis in intensive care units: a multicenter, 43. Herbrecht R, Denning DW, Patterson TF et al. Voriconazole versus amphotericin B for matched-cohort study. J Crit Care 2002;17:168–75. primary therapy of invasive aspergillosis. N Engl J Med 2002;347:408–15. 15. Petri MG, Konig J, Moecke HP et al. Epidemiology of invasive mycosis in ICU patients: a 44. Perfect JR, Marr KA, Walsh TJ et al. Voriconazole treatment for less-common, emerging, or prospective multicenter study in 435 non-neutropenic patients. Paul-Ehrlich Society for refractory fungal infections. Clin Infect Dis 2003;36:1122–31. Chemotherapy, Divisions of Mycology and Pneumonia Research. Intensive Care Med 1997;23:317–25. 45. Rex JH, Bennett JE, Sugar AM et al. A randomized trial comparing fluconazole with amphotericin B for the treatment of candidemia in patients without neutropenia. 16. Nolla-Salas J, Sitges-Serra A, Leon-Gil C et al. Candidemia in non-neutropenic critically ill Candidemia Study Group and the National Institute. N Engl J Med 1994;331:1325–30. patients: analysis of prognostic factors and assessment of systemic antifungal therapy. Study Group of Fungal Infection in the ICU. Intensive Care Med 1997;23:23–30. 46. Phillips P, Shafran S, Garber G et al. Multicenter randomized trial of fluconazole versus amphotericin B for treatment of candidemia in non-neutropenic patients. Canadian 17. Almirante B, Rodriguez D, Park BJ et al. Epidemiology and predictors of mortality in cases Candidemia Study Group. Eur J Clin Microbiol Infect Dis 1997;16:337–45. of Candida bloodstream infection: results from population-based surveillance, Barcelona, Spain, from 2002 to 2003. J Clin Microbiol 2005;43:1829–35. 47. Mora-Duarte J, Betts R, Rotstein C et al. Comparison of caspofungin and amphotericin B N Engl J Med 347 18. Falagas ME, Apostolou KE, Pappas VD. Attributable mortality of candidemia: a systematic for invasive candidiasis. 2002; :2020–9. review of matched cohort and case-control studies. Eur J Clin Microbiol Infect Dis 48. Kullberg BJ, Sobel JD, Ruhnke M et al. Voriconazole versus a regimen of amphotericin B 2006;25:419–25. followed by fluconazole for candidaemia in non-neutropenic patients: a randomised non- 19. Sendid B, Cotteau A, Francois N et al. Candidaemia and antifungal therapy in a French inferiority trial. Lancet 2005;366:1435–42. University Hospital: rough trends over a decade and possible links. BMC Infect Dis 49. Pappas PG, Rex JH, Sobel JD et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2006;6:80. 2004;38:161–89. 20. Pfaller MA, Jones RN, Doern GV et al. Bloodstream infections due to Candida species: 50. Bates DW, Su L, Yu DT et al. Mortality and costs of acute renal failure associated with SENTRY antimicrobial surveillance program in North America and Latin America, amphotericin B therapy. Clin Infect Dis 2001;32:686–93. 1997–1998. Antimicrob Agents Chemother 2000;44:747–51. 51. Blot SI, Vandewoude KH, Hoste EA, et al. Effects of nosocomial candidemia on outcomes 21. Klingspor L, Tornqvist E, Johansson A et al. A prospective epidemiological survey of of critically ill patients. Am J Med 2002;113:480–5. candidaemia in Sweden. Scand J Infect Dis 2004;36:52–5. 52. Ibrahim EH, Sherman G, Ward S et al. The influence of inadequate antimicrobial treatment 22. Ben Abraham R, Keller N, Teodorovitch N et al. Predictors of adverse outcome from of bloodstream infections on patient outcomes in the ICU setting. candidal infection in a tertiary care hospital. J Infect 2004;49:317–23. Chest 2000;118:146–55.
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 57 RAFAEL ZARAGOZA AND JAVIER PEMÁN
53. Kollef MH. Optimizing antibiotic therapy in the intensive care unit setting. Crit Care 63. Alvarez-Lerma F, Nicolas-Arfelis J, Rodriguez-Borregan J, et al. Clinical use and tolerability 2001;5:189–95. of voriconazole in the treatment of fungal infections in critically ill patients. J Chemother 54. Reboli A, Rotstein C, Pappas P et al. Anidulafungin vs. fluconazole for treatment of 2005;17:417–27. candidemia and invasive candidiasis. 45th Interscience Conference on Antimicrobial 64. Peman J, Salavert M, Canton E et al. Voriconazole in the management of nosocomial Agents and Chemotherapy, Washington, DC Abstract 1040. 2005. invasive fungal infections. Ther Clin Risk Management 2006;2:1–30. 55. Vazquez JA, Sobel JD. Anidulafungin: a novel echinocandin. Clin Infect Dis 65. Carrillo-Munoz AJ, Giusiano G, Ezkurra PA et al. Antifungal agents: mode of action in 2006;43:215–22. yeast cells. Rev Esp Quimioter 2006;19:130–9. 56. Chandrasekar PH, Sobel JD. Micafungin: a new echinocandin. Clin Infect Dis 66. Walsh TJ, Raad I, Patterson TF et al. Treatment of invasive aspergillosis with posaconazole in patients who are refractory to or intolerant of conventional therapy: an externally 2006;42:1171–8. controlled trial. Clin Infect Dis 2007;44:2–12. 57. Ruhnke M et al. Comparison of micafungin ans liposomoal amphoterecin B for invasive 67. Cornely OA, Maertens J, Winston DJ et al. Posaconazole vs. fluconazole or itraconazole candidiasis. Program and abstracts of the 45th Interscience Conference on Antimicrobial prophylaxis in patients with neutropenia. N Engl J Med 2007;356:348–59. Agents and Chemotherapy; September 16–19, 2005; Washington, DC. Abstract M-7222c. 68. Zaragoza R, Peman J. Invasive fungal infections in critically ill patients: different 58. Betts RF, Rotstein C, Talwar D et al. Comparison of micafungin and caspofungin for therapeutic options and a uniform strategy. Rev Iberoam Micol 2006;23:59–63. candidemia or invasive candidiasis (IC). Program and abstracts of the 46th Interscience 69. Marr K. Combination antifungal therapy: where are we now, and where are we going? Conference on Antimicrobial Agents and Chemotherapy; September 27–30, 2006; Oncology (Williston Park) 2004;18:24–9. San Francisco, CA. Abstract M-1308a. 70. Matthews RC, Rigg G, Hodgetts S et al. Preclinical assessment of the efficacy of 59. Kuhn DM, George T, Chandra J et al. Antifungal susceptibility of Candida biofilms: unique mycograb, a human recombinant antibody against fungal HSP90. Antimicrob Agents efficacy of amphotericin B lipid formulations and echinocandins. Antimicrob Agents Chemother 2003;47:2208–16. Chemother 2002;46:1773–80. 71. Pachl J, Svoboda P, Jacobs F et al. A randomized, blinded, multicenter trial of lipid- 60. Ostrosky-Zeichner L, Oude Lashof AM, Kullberg BJ et al. Voriconazole salvage treatment associated amphotericin B alone versus in combination with an antibody-based inhibitor of of invasive candidiasis. Eur J Clin Microbiol Infect Dis 2003;22:651–5. heat shock protein 90 in patients with invasive candidiasis. Clin Infect Dis 61. Munoz P, Sanchez-Somolinos M, Alcala L et al. Candida krusei fungaemia: antifungal 2006;42:1404–13. susceptibility and clinical presentation of an uncommon entity during 15 years in a single 72. Ito JI, Hooshmand-Rad R. Treatment of Candida infections with amphotericin B lipid general hospital. J Antimicrob Chemother 2005;55:188–93. complex. Clin Infect Dis 2005;40:S384–91. 62. Peman J, Canton E, Gobernado M. Epidemiology and antifungal susceptibility of Candida 73. Rex JH, Pappas PG, Karchmer AW et al. A randomized and blinded multicenter trial of species isolated from blood: results of a 2-year multicentre study in Spain. Eur J Clin high-dose fluconazole plus placebo versus fluconazole plus amphotericin B as therapy for Microbiol Infect Dis 2005;24:23–30. candidemia and its consequences in nonneutropenic subjects. Clin Infect Dis 2003;36:1221–8.
58 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 CLINICAL REVIEWS Commentary and Analysis on Recent Key Papers
Clinical reviews were prepared by Zeina Kanafani
ASPERGILLOSIS recently published trial compared posaconazole with fluconazole prophylaxis in patients with graft-versus-host disease following allogeneic stem cell transplantation [2]. Treatment of invasive aspergillosis with posaconazole Posaconazole was superior to fluconazole in preventing in patients who are refractory to or intolerant of aspergillosis (2.3% vs. 7.0%; p=0.006). Additionally, fewer conventional therapy: an externally controlled trial breakthrough episodes were observed in the posaconazole Walsh TJ, Raad I, Patterson TF et al. arm (1.0% vs. 5.9%; p=0.001). In another trial, Clin Infect Dis 2007;44:2–12. posaconazole improved survival (p=0.04) and was superior to both fluconazole and itraconazole in preventing invasive This study from Walsh and colleagues suggests that aspergillosis in patients with prolonged neutropenia (1% vs. posaconazole is a viable option in the treatment of 7%; p<0.001) [3]. 1. Raad II, Hanna H, Boktour M et al. Posaconazole compared to high-dose lipid patients with invasive aspergillosis who are refractory formulations of amphotericin B as salvage therapy for invasive aspergillosis in patients with to or intolerant of conventional therapy. hematologic malignancy; Abstract M-874. 46th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco, CA; 2006. 2. Ullmann AJ, Lipton JH, Vesole DH et al. Posaconazole or fluconazole for prophylaxis in This was an open-label, prospective trial in which patients severe graft-versus-host disease. N Engl J Med 2007;356:335–47. 3. Cornely OA, Maertens J, Winston DJ et al. Posaconazole vs. fluconazole or itraconazole refractory to or intolerant of conventional therapy received prophylaxis in patients with neutropenia. N Engl J Med 2007;356:348–59. posaconazole for the treatment of invasive aspergillosis. Address for reprints: TJ Walsh, Immunocompromised Host These subjects were compared with an external, Section, National Cancer Institute, Bethesda, MD 20892, USA. retrospectively collected control group. The success rate at Email: [email protected] the end of therapy was 42% in the posaconazole group and 26% in the control group (p=0.006). Micafungin (FK463), alone or in combination with Additional data recently presented at the 46th Interscience other systemic antifungal agents, for the treatment Conference on Antimicrobial Agents and Chemotherapy of acute invasive aspergillosis (ICAAC) confirmed the efficacy of posaconazole in invasive Denning DW, Marr KA, Lau WM et al. aspergillosis. In a comparative retrospective cohort study, J Infect 2006;53:337–49. Raad et al. compared posaconazole to lipid formulations of amphotericin B as salvage therapy for invasive aspergillosis in This study was a prospective non-comparative trial patients with hematological malignancy [1]. A total of 51 that examined the efficacy of micafungin, alone or in patients received posaconazole and 49 patients received high- combination with other antifungal agents, in the dose lipid formulations of amphotericin B (≥7.5 mg/kg/day). treatment of adult and pediatric patients with various The response rate was significantly better with posaconazole underlying diseases and proven or probable invasive than with lipid formulations of amphotericin B (39% vs. 8%; aspergillosis. Clinical response was achieved in 36% of p≤0.001). In addition, the mortality rate attributable to patients in the overall population (50% of patients who aspergillosis was lower in the posaconazole group (39% vs. received micafungin alone). Of the patients who = 63%; p 0.02). On multivariable analysis, posaconazole completed the 6-week follow-up period, 32% were treatment was the only factor significantly associated with deemed to have a complete or partial response. improved outcome (odds ratio 8.8; 95% confidence interval 2.6–29.4). In this non-comparative, prospective trial the authors found Furthermore, posaconazole appears to be useful in that micafungin is efficacious as part of the therapeutic preventing invasive aspergillosis in high-risk patients. A regimen of patients with invasive aspergillosis. The sole
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 59 CLINICAL REVIEWS
other published trial on micafungin in the treatment of Impact of the intensity of the pretransplantation aspergillosis was conducted by Kohno and colleagues in conditioning regimen in patients with prior Japan [1]. In that study, patients had a variety of underlying invasive aspergillosis undergoing allogeneic diseases and did not receive any other antifungal therapy hematopoietic stem cell transplantation: a while on micafungin. At the end of therapy, six of the retrospective survey of the Infectious Diseases 10 patients with invasive pulmonary aspergillosis had a Working Party of the European Group for Blood favorable clinical response. and Marrow Transplantation In addition, two studies recently presented at the Martino R, Parody R, Fukuda T et al. 46th Interscience Conference on Antimicrobial Agents and Blood 2006;108:2928–36. Chemotherapy (ICAAC) evaluated the use of micafungin alone or in combination with other antifungal agents in the The Infectious Diseases Working Party of the European treatment of invasive aspergillosis. The first study was Group for Blood and Marrow Transplantation undertook conducted in adult bone marrow transplant recipients this retrospective study in order to determine outcomes = (n 98). Eight patients were treated with micafungin alone of patients with invasive aspergillosis (IA) who underwent while the remaining 90 patients received micafungin in allogeneic hematopoietic stem cell transplantation combination with other antifungal therapy [2]. Twenty-two (HSCT) with a conventional or reduced-intensity patients on combination therapy (24%) and three patients conditioning regimen. Overall, IA progressed in 27/129 on micafungin alone (38%) had a partial or complete (21%) patients after allogeneic HSCT. Early and late risk response to therapy. The second study was conducted in factors for progression of IA were identified, and the 58 pediatric patients with invasive aspergillosis [3]. Forty-seven authors generated a risk model for this progression. patients received micafungin in combination with liposomal amphotericin B. Only two patients received micafungin alone. Patients with hematological malignancies who undergo Nine patients exhibited complete response to therapy (16%) chemotherapy are often referred for subsequent allogeneic and another 17 patients had a partial response (29%). In both hematopoietic stem cell transplantation (HSCT). However, studies, micafungin was well tolerated. invasive aspergillosis (IA) is a common complication of Evidence from other clinical trials suggests that chemotherapy, and until recently was considered to be an micafungin is also efficacious in the treatment of absolute contraindication to allogeneic HSCT due to the mucocutaneous and invasive candidiasis, as well as in the potential for relapse of IA after transplantation, and an elevated treatment of patients with hematological malignancies and nonrelapse mortality rate. Following advances in antifungal febrile neutropenia. drugs and development of reduced-intensity chemotherapeutic The present authors conducted a prospective non- conditioning regimens, successful outcomes of allogeneic HSCT comparative trial that examined the efficacy of micafungin have been reported in patients with prior IA. alone or in combination with other antifungal agents in the The aims of this study were to determine outcomes in treatment of 225 adult and pediatric patients with various patients who had a recent history of proven or probable IA underlying diseases and proven or probable invasive undergoing allogeneic HSCT with a conventional aspergillosis. Clinical response was achieved in 36% of myeloablative or a reduced-intensity conditioning regimen, patients, half of whom had received micafungin alone. In all, and to identify risk factors for progression of IA after 183 of the 331 enrolled patients died during treatment. Of allogeneic HSCT. those who completed the 6-week follow-up period, 32% A total of 129 patients (median age 42 years) from were deemed to have a complete or partial response. 44 centers across 12 countries who received an allogeneic 1. Kohno S, Masaoka T, Yamaguchi H et al. A multicenter, open-label clinical study of HSCT were studied. IA progressed in 27 patients (21%) after micafungin (FK463) in the treatment of deep-seated mycosis in Japan. Scand J Infect Dis 2004;36:372–9. transplantation, and the incidence of progression was similar in 2. Kontoyiannis DP, Ratanatharathorn V, Van Burik J et al. Micafungin alone or in patients receiving either conditioning regimen; however, there combination with other licensed antifungal therapy in bone marrow transplant recipients with invasive aspergillosis; Abstract M-878. 46th Annual Interscience Conference on was a trend toward a higher IA-related mortality in the Antimicrobial Agents and Chemotherapy. San Francisco, CA; 2006. conventionally treated group. Patients in partial remission or 3. Flynn PM, Seibel N, Arrieta AC et al. Treatment of invasive aspergillosis in pediatric patients with micafungin alone or in combination with other systemic antifungal agents; complete remission from infection at transplantation were less Abstract M-891. 46th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco, CA; 2006. likely to progress than those who were not (17% vs. 32%; p=0.012). Due to the wide range of antifungal regimens Address for reprints: DW Denning, Education and Research Centre, employed, it was not possible to identify significant differences Wythenshawe Hospital and University of Manchester, Academic Department of Medicine and Surgery, Southmoor Road, Manchester between these in terms of IA progression in each the two M23 9LT, UK. Email: [email protected] conditioning groups, although voriconazole monotherapy as
60 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 COCCIDIOIDOMYCOSIS
second-line prophylaxis in the immediate post-HSCT phase autopsy diagnosis, and there was a very high overall showed a trend in reducing progression of IA (p=0.15). mortality rate (92.7%) in these subjects. The authors suggest In a multivariate analysis, early progression of IA (<30 days that this may be due to delayed diagnosis as a result of the post-transplantation) was more common in patients who infrequent isolation of the fungus from blood cultures – in received a conventional conditioning regimen (15% vs. 7%; most of the cases, surgical intervention was vital in obtaining p=0.054). Late progression of IA (>30 days post- a definitive diagnosis. Furthermore, Aspergillus aortitis can transplantation) was more common in those with often go undetected by trans-thoracic and trans-esophageal cytomegalovirus, bone marrow or cord blood as a source of echocardiography, but is more readily identified by computed stem cells, and in those with grades II–IV acute graft-versus- tomography, magnetic resonance imaging, and angiography. host disease. Other variables that increased the risk of Although there have not been any studies of the optimal progression during the entire transplantation period were treatment for infection to date, the findings of this literature a longer duration of neutropenia (>20 days post- review suggest that the infected tissue should be excised transplantation), advanced status of the underlying disease, and a systemic antifungal agent should be given, with and a short time-period between the start of antifungal results indicating that this should be continued for 3 months therapy and the allogeneic HSCT (<6 weeks). From these after the last evidence of active disease. data, the authors generated a risk model for progression of IA. The authors conclude that the development of better The authors conclude that a history of IA is not an diagnostic methods is essential if outcomes are to be absolute contraindication for allogeneic HSCT, particularly if improved. Moreover, given that in the majority of cases the patients have a low-risk profile. Furthermore, they suggest presented in this review the Aspergillus spores originated from that high-risk subjects could be prospectively assessed in the air in the operating theater, special attention should be studies that aim to reduce the high risk of progression of IA. paid to the maintenance of ventilation ducts in these areas.
Address for reprints: R Martino, Hematology Department of Address for reprints: DW Denning, Education and Research Centre, Sant Pau Hospital, Autonomous University of Barcelona, Barcelona, Wythenshawe Hospital, Southmoor Road, Manchester M23 9LT, UK. Spain. Email: [email protected] Email: [email protected]
Post-operative aspergillosis Pasqualotto AC, Denning DW. COCCIDIOIDOMYCOSIS Clin Microbiol Infect 2006;12:1060–76.
The authors of this literature review of postoperative Coccidioidomycosis in patients with aspergillosis report that successful treatment requires diabetes mellitus prompt diagnosis, and surgical excision of the affected Santelli AC, Blair JE, Roust LR. tissue together with antifungal therapy. Regular Am J Med 2006;119:964–9. maintenance of operating theater ventilation systems should be undertaken to minimize the risk of infection. In this retrospective review of 329 immunocompetent patients with coccidioidomycosis, Santelli et al. demonstrate The authors of this review performed a literature search that diabetics are more likely than non-diabetics to have to investigate cases of surgical site infection with Aspergillus cavitary lung disease and relapsed infection. In addition, in spp. in immunocompetent individuals. Keyword searches of diabetic patients with higher serum glucose concentrations, Medline, LILACS (Latin American and Caribbean Health the risk for dissemination and treatment failure increased Science Literature Database), and EMBASE (Excerpta Medica significantly. Glycemic control seems to have an important Database) databases identified more than 500 cases. In their impact on the outcome of coccidioidomycosis. review, the authors discuss the presentations and treatment of infection following different types of surgery. Cardiac In this paper Santelli and colleagues describe the effect of surgery was the procedure that was most commonly glycemic control on the clinical characteristics of diabetic associated with infection (n=188), followed by dental patients with coccidioidomycosis. Although most infected surgery (>100), and ophthalmological surgery (>90). patients are otherwise healthy individuals, Coccidioides Among those with infection following cardiac surgery, the immitis is a well known opportunistic organism targeting mean time to diagnosis was 2.7 months, but this varied from patients with depressed cellular immunity, such as patients <1 to >12 months, and the diagnosis was often made with HIV infection, bone marrow transplant recipients, and following autopsy. Indeed, only 43.5% of cases had a pre- patients receiving immunosuppressive therapy.
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 61 CLINICAL REVIEWS
There is strong evidence to suggest that diabetes mellitus It has been suggested that the diminished cell-mediated is an important determinant of the manifestations and immunity in pregnancy contributes to the establishment outcomes of various infectious processes, including fungal and dissemination of fungal infections. During pregnancy, infections. In one study, oral colonization and infection with there is decline of CD4+ T cell function and augmentation Candida occurred more frequently in 405 patients with of suppressor T lymphocyte function. In addition, insulin-dependent diabetes mellitus than in 268 non-diabetic interleukin production by the placenta causes down- controls (15.1% vs. 3.0%) [1]. Upon multivariate analysis, regulation of type 1 T helper cells and upregulation of poor glycemic control (as evidenced by an elevated type 2 T helper cells. The resulting immune modulation glycosylated hemoglobin level) was an independent facilitates the dissemination of endemic fungal infections predictor of Candida colonization and infection (odds ratio that are usually kept in check by the host’s cell- 1:9). In another similar study, fungal colonization in the mediated immunity [1]. vagina and rectum of pregnant women was more common Infections with other dimorphic fungi have been rarely in diabetic patients with poor glycemic control than non- reported in pregnancy. Whitt et al. described three cases of diabetic women [2]. The majority of isolated fungal strains disseminated histoplasmosis in pregnant women, two of were Candida species (89.6%). which (one with diabetes mellitus and the second with HIV Other fungal infections that have been strongly infection) resulted in a fatal outcome [2]. The patient associated with diabetes mellitus include zygomycosis. who survived was previously healthy. Transplacental However, unlike the case with coccidioimycosis and Candida transmission of Histoplasma capsulatum was confirmed in infections, glycemic control does not seem to be the one fetus. principal factor predisposing to zygomycosis, as affected Blastomycosis is slightly more frequently encountered in patients tend to have similar hemoglobin A1c levels to non- pregnancy than histoplasmosis. A review of the medical affected patients [3]. literature by Lemos et al. yielded 16 cases of pregnancy- Of the other dimorphic fungi, Blastomyces dermatitidis associated blastomycosis [3]. The authors were able to can also act as an opportunistic pathogen in diabetic identify three additional cases at the University of Mississippi patients. In fact, 22% of all patients with blastomycosis have Medical Center (Jackson, MI, USA). The majority of cases an underlying diagnosis of diabetes mellitus [4]. consisted of pulmonary blastomycosis (14 cases) and skin 1. Guggenheimer J, Moore PA, Rossie K et al. Insulin-dependent diabetes mellitus and oral involvement was present in nine cases. Antifungal therapy soft tissue pathologies: II. Prevalence and characteristics of Candida and Candidal lesions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000;89:570–6. was administered in 16 patients (amphotericin B in 2. Nowakowska D, Kurnatowska A, Stray-Pedersen B et al. Species distribution and influence 13 cases). There were no fatalities and complete cure was of glycemic control on fungal infections in pregnant women with diabetes. J Infect 2004;48:339–46. achieved in 14 cases. Four additional women experienced 3. Moye J, Rosenbloom AL, Silverstein J. Clinical predictors of mucormycosis in children with type 1 diabetes mellitus. J Pediatr Endocrinol Metab 2002;15:1001–4. good responses to treatment. Transplacental infection was 4. Lemos LB, Baliga M, Guo M. Blastomycosis: the great pretender can also be an documented in two cases resulting in the deaths of opportunist. Initial clinical diagnosis and underlying diseases in 123 patients. Ann Diagn Pathol 2002;6:194–203. both fetuses. The present report provides a comprehensive review of Address for reprints: JE Blair, Division of Infectious Diseases, Mayo the literature on coccidioidomycosis in pregnancy. Although Clinic, 13400 East Shea Boulevard, Scottsdale, AZ 85259, USA. infections with dimorphic fungi are uncommon in pregnancy, they are associated with significant morbidity. In Coccidioidomycosis in pregnancy: case the 80 cases identified, there was a higher likelihood of report and review of the literature dissemination of the infection during the third trimester Crum NF, Ballon-Landa G. (96%) compared with the first (50%) and second trimesters Am J Med 2006;119:993.e11–7. (62%), with an overall mortality rate of 36%. The authors report that early recognition of disease and treatment are The authors of this report identified 80 cases of associated with an improved outcome. coccidioidomycosis in pregnant women in the literature. 1. Formby B. Immunologic response in pregnancy: its role in endocrine disorders of pregnancy and influence on the course of maternal autoimmune diseases. Endocrinol There was a higher likelihood of dissemination of the Metab Clin North Am 1995;24:187–205. 2. Whitt SP, Koch GA, Fender B et al. Histoplasmosis in pregnancy: case series and report of infection during the third trimester (96%) compared transplacental transmission. Arch Intern Med 2004;164:454–8. with the first (50%) and second trimesters (62%). The 3. Lemos LB, Soofi M, Amir E. Blastomycosis and pregnancy. Ann Diagn Pathol overall mortality rate was 36%. Early diagnosis and 2002;6:211–5.
treatment were associated with improved maternal and Address for reprints: NF Crum, Infectious Diseases Division, fetal outcome. Naval Medical Center, San Diego, CA 92134-1006, USA. Email: [email protected]
62 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 CANDIDIASIS
CANDIDIASIS Uniform distribution of three Candida albicans microsatellite markers in two French ICU populations supports a lack of nosocomial Candida biofilms on implanted biomaterials: cross-contamination a clinically significant problem Eloy O, Marque S, Botterel F et al. Ramage G, Martinez JP, Lopez-Ribot JL. BMC Infect Dis 2006;6:162. FEMS Yeast Res 2006;6:979–86. This small epidemiological study, based at two intensive Over the past 10–15 years, an increased understanding of care units in France, found no evidence to suggest that the significance of microbial biofilms has developed. The nosocomial transmission of Candida albicans occurs present minireview by Ramage et al. describes the unique between patients. characteristics of biofilms in general and the clinical impact of Candida biofilms on implanted medical devices in particular. Candida albicans accounts for 55% of all yeast bloodstream infections in intensive care units (ICUs) [1], most of which Biofilms are structured, three-dimensional microbial are nosocomial, and the organism represents a growing communities attached to a surface and encapsulated within concern. Genotyping is one method of determining the a protective extracellular matrix. This unique environment is route of contamination. Using three highly specific distinct from that of planktonic (free-floating) cells, and polymorphic microsatellite markers (PMMs), the authors provides survival advantages to the microbe such as performed a follow-up to a previous single-center study protection from host immune defenses and increased showing that cross-contamination between patients was resistance to antimicrobial agents. A number of studies have unlikely [2], with the aim of extending their findings to a shown dramatically reduced sensitivity of Candida biofilms more generalized setting. to flucanozole and amphotericin B [1–3]. Novel antifungals Data from the earlier study showed that, between such as caspofungin, a member of the echinocandins, look November 1999 and October 2000, 43 of 94 patients (46%) more promising as potent activity against C albicans within had at least one positive culture for C albicans, 36 of whom the approved therapeutic range has been demonstrated [3]. had positive samples after 72 h of hospitalization. In the second Use of artificial materials in implanted medical devices hospital ICU, between February 2000 and January 2002, 36 of allows microbes such as Candida to colonize and proliferate in 60 patients (60%) had at least one positive culture for the body, and the infection can, in many cases, only be C albicans, all of which were positive after 72 h. Of these, 26 eradicated by removing the device. This can be difficult or even had a specific multilocus genotype, two had a common impossible, resulting in a reservoir for persistent infections multilocus genotype, and eight had the most common being established inside the patient. Biofilm formation often genotype in the general population. Of the patients who had leads to infections associated with high rates of morbidity and common genotypes, the time between their periods of stay in mortality, and can also cause device malfunction. The problem hospital differed by 13–78 days, making any direct cross- is widespread as different types of catheters and endotracheal contamination unlikely. By comparing the three PMMs between tubes are used in millions of patients every year, some of which the two ICUs the authors found no evidence of a difference carry a risk of Candida infection of up to 30%. Other common between the rate of nosocomial C albicans colonizations, devices frequently affected by Candida biofilms include despite some potentially important methodological differences. prosthetic joints, neurosurgical shunts, heart valves, voice Thus, these data support the findings of the earlier study that prostheses, and dentures. ICU patients tend to harbor their own C albicans isolates and The authors conclude that the current mortality rates that cross-contamination is unlikely. They also extend the associated with biofilm-related infections are unacceptably findings of the previous study to suggest that no genotype is high, and that novel drugs and strategies are required to specific to a particular hospital. The authors suggest that efforts better control this problem. should be focused on treating the risk factors for infection rather 1. Hawser SP, Douglas LJ. Resistance of Candida albicans biofilms to antifungal agents in than preventing cross-contamination, as this rarely occurs. vitro. Antimicrob Agents Chemother 1995;39:2128–31. 2. Baillie GS, Douglas LJ. Candida biofilms and their susceptibility to antifungal agents. 1. Tortorano AM, Peman J, Bernhardt H et al. Epidemiology of candidaemia in Europe: results Methods Enzymol 1999;310:644–56. of 28-month European Confederation of Medical Mycology (ECMM) hospital-based surveillance study. Eur J Clin Microbiol Infect Dis 2004;23:317–22. 3. Ramage G, VandeWalle K, Bachmann SP et al. In vitro pharmacodynamic properties of three antifungal agents against preformed Candida albicans biofilms determined by time- 2. Stephan F, Bah MS, Desterke C et al. Molecular diversity and routes of colonization of kill studies. Antimicrob Agents Chemother 2002;46:3634–6. Candida albicans in a surgical intensive care unit, as studied using microsatellite markers. Clin Infect Dis 2002;35:1477–83. Address for reprints: JL López-Ribot, Department of Biology, The University of Texas at San Antonio, 6900 North Loop 1604 West, Address for reprints: S Bretagne, Unité Mixte de Recherche BIPAR 956, San Antonio, TX 78249, USA. Email: [email protected] Université Paris 12, Créteil, France. Email: [email protected]
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 63 CLINICAL REVIEWS
Caspofungin for invasive candidiasis at even in patients who were critically ill, is one of the reasons a tertiary care medical center caspofungin is an attractive alternative treatment. Zaas AK, Dodds Ashley ES, Alexander BD et al. Overall, the authors conclude that caspofungin therapy is a Am J Med 2006;119:993.e1–6. successful treatment for invasive candidiasis (similar to previous clinical trials) and should, therefore, be considered by clinicians Caspofungin has been reported to be a successful therapy as a primary therapy, particularly for non-albicans species. 1. Centers for Disease Control and Prevention – in patients who have invasive candidiasis infection but are http://www.cdc.gov/ncidod/dbmd/diseaseinfo/candidiasis_inv_g.htm resistant to existing treatments such as amphotericin B or azoles. In this article, the authors describe a single-center Address for reprints: AK Zaas, Box 3355 Duke South Blue Zone, Duke University Medical Center, Durham, NC 27710, USA. study of 99 patients with 104 episodes of caspofungin Email: [email protected] treatment, and recommend that caspofungin should be used as first-line therapy, particularly in non-albicans Candida infections. Oral nystatin prophylaxis to prevent invasive candidiasis in neonatal intensive care unit Invasive candidiasis is one of the most common fungal Ozturk MA, Gunes T, Koklu E et al. infections and is usually treated with amphotericin B or with Mycoses 2006;49:484–92. azole drugs. Non-albicans Candida spp. (e.g. C glabrata) are more likely to be resistant to existing treatments, and In this study, the authors investigate the usefulness of comprise >50% of the isolates that cause invasive oral nystatin prophylaxis for the prevention of invasive candidiasis. However, a few study groups have reported in fungal infections in infants admitted to the neonatal favor of caspofungin, a new antifugal agent, in patients who intensive care unit. They found that infants who received were not responding to amphotericin B or to fluconazole nystatin were less likely to develop invasive candidiasis treatment. In order to study the effects of caspofungin for than infants who did not receive such prophylaxis. the treatment of invasive candidiasis, the authors assessed a single-center experience of 104 consecutive courses of Invasive candidiasis in neonates is associated with significant caspofungin in 99 patients. morbidity and mortality rates. By 6 weeks of life, 62% of From the Duke University Medical Center pharmacy extremely low-birth weight infants (ELBW; <1000 g) are antimicrobial indications database, the authors identified colonized with Candida species. In this study the authors 241 patients who were receiving caspofungin (319 courses in investigate the value of oral nystatin in the prevention of total) from 2001–2003. Of these, only 99 patients (104 courses invasive candidiasis in infants admitted to the neonatal of caspofungin) met the criteria for inclusion in the study. The intensive care unit. They conclude that oral nystatin was most common sites of infection were the bloodstream and the effective in preventing invasive candidiasis in ELBW and very abdomen (66% and 25%, respectively). Non-albicans Candida low-birth weight infants. infections comprised the majority (77%) of infections treated Three trials have studied the use of oral antifungal with caspofungin, of which C glabrata was the most common prophylaxis in preterm infants, with inconsistent results. In the isolate and C parapsilosis the least common. first study, miconazole oral gel was compared with placebo in a Following a median follow-up of 84 days, 81 patients sample of 757 infants [1]. Although rectal colonization was (82%) remained free of candidiasis. Eighty-three percent of significantly lower in the miconazole group compared with the bloodstream and 84% of abdominal infections were cured placebo group (19.5% vs. 36.2%; p<0.0001), there was no after caspofungin treatment. On the whole, C parapsilosis reduction in the incidence of systemic fungal infection with had the highest clinical cure success rate (100%) while miconazole prophylaxis (2.0% vs. 2.6%). The second trial C albicans had the lowest (77%). Of the 104 “episodes”, compared the outcome of 33 preterm infants who received 20% were administered caspofungin as primary therapy oral nystatin prophylaxis with that of a control group of (the overall success rate was 76%) and the remaining 34 preterm infants [2]. Oral nystatin was successful in reducing 80% received various other antifungal treatments prior to fungal colonization, systemic fungal infection, and mortality caspofungin therapy. Although the trend was not significant, rate in this cohort of patients. The third study compared oral there was a decrease in the use of other antifungal treatments nystatin with oral fluconazole and found no significant prior to caspofungin therapy between years 2001–2003. differences in mortality rate or systemic fungal infections [3]. Treatment with caspofungin resulted in a few rare adverse Due to these conflicting data, additional outcome studies events, which included elevated liver function tests, wheezing, are required before oral fungal prophylaxis can be widely and infusion-related hypotension. The lack of adverse events, recommended in preterm infants.
64 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 CRYPTOCOCCOSIS
1. Wainer S, Cooper PA, Funk E et al. Prophylactic miconazole oral gel for the prevention of neonatal fungal rectal colonization and systemic infection. Pediatr Infect Dis J respectively. These results were confirmed in another study 1992;11:713–6. where anidulafungin was successful in eradicating 84% of 2. Sims ME, Yoo Y, You H et al. Prophylactic oral nystatin and fungal infections in very-low- birthweight infants. Am J Perinatol 1988;5:33–6. Candida species from the bloodstream [4]. The highest 3. Violaris K, Doraiswamy B, Olawepo O et al. Fluconazole vs. nystatin prophylaxis for fungal infection in very low birth weight (VLBW) infants. Ped Res 1998;43:254A. eradication rate was for C glabrata (95%). The results of double-blind, randomized, controlled trial of anidulafungin Address for reprints: T Gunes, Department of Paediatrics, Division of compared with fluconazole in patients with invasive Neonatology, School Medicine, Erciyes University, 38039 Kayseri, Turkey. Email: [email protected] candidiasis were recently presented [5]. The success rate at 6-weeks follow-up was 56% with anidulafungin and 44% with fluconazole. In vitro pharmacodynamics of anidulafungin 1. Krause DS, Simjee AE, van Rensburg C et al. A randomized, double-blind trial of anidulafungin versus fluconazole for the treatment of esophageal candidiasis. Clin Infect and caspofungin against Candida glabrata Dis 2004;39:770–5. isolates, including strains with decreased 2. Vazquez J, Schranz JA, Krause D et al. Efficacy from a phase 2/3 study of anidulafungin (ANID) in patients with azole-refractory mucosal candidiasis; Abstract P1040. 44th caspofungin susceptibility Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington, DC; Cota J, Carden M, Graybill JR et al. 2004. 3. Krause DS, Goldstein BP, Wible M et al. A Phase 2, randomized, dose-ranging study Antimicrob Agents Chemother 2006;50:3926–8. evaluating the safety and efficacy of anidulafungin in invasive candidiasis and candidemia. Antimicrob Agents Chemother 2004;48:2021–4. 4. Pfaller MA, Diekema DJ, Boyken L et al. Effectiveness of anidulafungin in eradicating Anidulafungin was recently approved by the US Food and Candida species in invasive candidiasis. Antimicrob Agents Chemother 2005;49:4795–7. 5. Reboli A, Rotstein C, Pappas P et al. Anidulafungin vs. fluconazole for treatment of Drug Administration for the treatment of esophageal candidemia and invasive candidiasis; Abstract 1040. 45th Interscience Conference on candidiasis, candidemia, and deep-tissue candidiasis. The Antimicrobial Agents and Chemotherapy. Washington, DC; 2005. authors of this report discuss the in vitro pharmacodynamics Address for reprints: NP Wiederhold, University of Texas at Austin College of anidulafungin compared with caspofungin against strains of Pharmacy, UTHSCSA, Clinical Pharmacy, MSC 6220, 7703 Floyd Curl of Candida glabrata. Drive, San Antonio, TX 78229, USA. Email: [email protected]
In this in vitro study, anidulafungin was compared with caspofungin for activity against 18 isolates of Candida CRYPTOCOCCOSIS glabrata. These isolates included three with elevated caspofungin minimum inhibitory concentrations (MICs). Overall, anidulafungin MICs were at least two dilutions Symptomatic relapse of HIV-associated lower than those of caspofungin. All three isolates with cryptococcal meningitis after initial fluconazole reduced susceptibility to caspofungin were fully susceptible monotherapy: the role of fluconazole resistance to anidulafungin. and immune reconstitution In a previous randomized, double-blind, controlled trial, Bicanic T, Harrison T, Niepieklo A et al. the activity of anidulafungin was compared with that of Clin Infect Dis 2006;43:1069–73. fluconazole in patients with endoscopically proven esophageal candidiasis [1]. A total of 300 patients were The authors of this study prospectively enrolled HIV randomized to the anidulafungin arm and 301 to the patients with relapsed cryptococcal meningitis who had fluconazole arm. Endoscopic response rates were 97% in the received initial fluconazole monotherapy. In 76% of anidulafungin group and 99% in the fluconazole group. relapses in which cerebrospinal fluid culture tested However, relapse rates at the 2-week follow-up were positive for Cryptococcus, the isolate had reduced greater in patients receiving anidulafungin (36% vs. 10%). susceptibility to fluconazole. In addition, more than Anidulafungin has also been shown to be effective in the one-third of the relapses were attributed to immune treatment of oropharyngeal candidiasis in patients unresponsive reconstitution inflammatory syndrome. to treatment with fluconazole [2]. A total of 18 patients with azole-refractory oropharyngeal candidiasis were enrolled in an In this study, Bicanic and colleagues found that fluconazole open-label study, of whom 17 had AIDS. The response rate to resistance and immune reconstitution inflammatory syndrome anidulafungin was very encouraging (17 of 18 patients). predicted clinical relapses of cryptococcal meningitis in HIV In addition, anidulafungin was evaluated in three patients who received initial fluconazole monotherapy. different dose regimens (50, 75, and 100 mg/day) in To date, three clinical trials have investigated the efficacy patients with invasive candidiasis [3]. The evaluable of fluconazole alone or in combination with other antifungal population consisted of 68 patients. The success rates at agents as initial therapy in cryptococcal meningitis, although 2 weeks of follow-up were 72%, 85%, and 83%, none had a randomized, controlled design.
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 65 CLINICAL REVIEWS
The first of these trials was a small, open, non- Scedosporium is an opportunistic dematiaceous fungus comparative study of intravenous fluconazole followed by that causes serious invasive infections in patients oral fluconazole in AIDS patients with acute cryptococcal with altered immune defenses. In addition to patients meningitis [1]. Thirteen patients received fluconazole with malignancies, Scedosporium apiospermum and 400 mg/day intravenously for 12–16 days followed by Scedosporium prolificans have been increasingly reported fluconazole 400 mg/day orally to complete therapy. The in transplant recipients. fluconazole dose was decreased to 200 mg/day if follow-up The authors of this report present a review of 25 cases of cerebrospinal fluid (CSF) cultures were negative at 32 weeks. Scedosporium infections in patients admitted to the MD Treatment was successful in six patients (46%); one patient Anderson Cancer Center (Houston, TX, USA) between 1989 died secondary to cryptococcal meningitis. and 2006. Most infections (n=21) were caused by In another trial, 14 AIDS patients with crypto-coccal S apiospermum. Interestingly, cases of S proliticans were not meningitis received fluconazole 800–1000 mg intra- identified until the year 2000. Sixteen patients had venously [2]. Clinical response at the end of therapy was disseminated disease, of whom 12 had fungemia. The achieved in eight of 11 (73%) patients. The time to first disease was associated with a high mortality rate (12/18 for negative CSF culture was longer in patients with initial CSF S apiospermum and 4/4 for S prolificans). cryptococcal antigen titers of ≥1:1024 and when the In a recent review, Husain and colleagues identified minimum inhibitory concentration of fluconazole was 80 transplant patients with Scedosporium infections [1]. The 4 μg/mL. comparator group consisted of 190 patients who were also Finally, a more recent study evaluated the efficacy of diagnosed with Scedosporium infections but were not different antifungal combinations for the treatment of HIV- transplant recipients. In the transplant group, 23 patients associated cryptococcal meningitis [3]. A total of 64 patients (29%) were hematopoeitic stem cell transplant (HSCT) were randomized to receive either amphotericin B recipients while the remaining 57 patients had received solid at 0.7 mg/kg/day, amphotericin B plus flucytosine organ transplants (SOTs). S apiospermum was more 100 mg/kg/day, amphotericin B plus fluconazole prevalent than S prolificans (76% vs. 24%). Scedosporiosis 400 mg/day, or triple therapy with amphotericin B, developed most commonly within the first 6 months after flucytosine, and fluconazole. Amphotericin B plus flucytosine transplantation (64%). The vast majority of patients (96%) was found to be the most rapidly fungicidal combination. were receiving corticosteroids at the time of infection. 1. Stern JJ, Pietroski NA, Buckley RM et al. Parenteral and oral fluconazole for acute Just one-third were receiving antifungal prophylaxis. The cryptococcal meningitis in AIDS: experience with thirteen patients. Ann Pharmacother 1992;26:876–82. disease frequently presented as a disseminated infection 2. Menichetti F, Fiorio M, Tosti A et al. High-dose fluconazole therapy for cryptococcal (54%), with fungemia occurring in 17% of cases. When meningitis in patients with AIDS. Clin Infect Dis 1996;22:838–40. 3. Brouwer AE, Rajanuwong A, Chierakul W et al. Combination antifungal therapies for HIV- comparing transplant with non-transplant recipients, associated cryptococcal meningitis: a randomised trial. Lancet 2004;363:1764–7. patients with hematological malignancies were at highest Address for reprints: T Bicanic, Desmond Tutu HIV Centre, Wehner and risk of developing disseminated infection (86%), followed Beit Building North, Institute of Infectious Disease and Molecular by HSCT recipients (69%), HIV patients (57%), and Medicine, University of Cape Town Faculty of Health Sciences, Anzio Road, Observatory 7925, Cape Town, South Africa. SOT recipients (55%). The same pattern was seen for Email: [email protected] fungemia (66% with hematological malignancies, 25% with HSCT, 23% with HIV, and 16% with SOT). The mortality rate in the transplant group was 58%. OTHER MYCOSES Disseminated infection, fungemia, central nervous system involvement, and renal failure were associated with higher mortality rates. Scedosporium infection in a tertiary care cancer Taken together, these data demonstrate that center: a review of 25 cases from 1989–2006 Scedosporium infections in transplant patients or in those Lamaris GA, Chamilos G, Lewis RE et al. with hematological malignancies are associated with a high Clin Infect Dis 2006;43:1580–4. rate of dissemination, and a poor outcome. 1. Husain S, Munoz P, Forrest G et al. Infections due to Scedosporium apiospermum and Scedosporium prolificans in transplant recipients: clinical characteristics and impact of This paper presents a review of the clinical manifestations antifungal agent therapy on outcome. Clin Infect Dis 2005;40:89–99. and outcome of 25 cancer patients with Scedosporium Address for reprints: DP Kontoyiannis, Deptartment of Infectious infections. The infection was associated with a high Diseases, Infection Control and Employee Health, Unit 402, mortality rate. The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA. Email: [email protected]
66 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 THERAPEUTICS
1. Madan V, Bisset D, Harris P et al. Phaeohyphomycosis caused by Exophiala salmonis. Br J Catheter-associated fungemia due to Exophiala Dermatol 2006;155:1082–4. oligosperma in a leukemic child and review of 2. Lecciones JA, Lee JW, Navarro EE et al. Vascular catheter-associated fungemia in patients fungemia cases caused by Exophiala species with cancer: analysis of 155 episodes. Clin Infect Dis 1992;14:875–83. 3. Lunardi LW, Aquino VR, Zimerman RA et al. Epidemiology and outcome of Rhodotorula Al-Obaid I, Ahmad S, Khan ZU et al. fungemia in a tertiary care hospital. Clin Infect Dis 2006;43:e60–3. Eur J Clin Microbiol Infect Dis 2006;25:729–32. 4. Kontoyiannis DP, Torres HA, Chagua M et al. Trichosporonosis in a tertiary care cancer center: risk factors, changing spectrum and determinants of outcome. Scand J Infect Dis 2004;36:564–9.
In this report, the authors describe a case of catheter- Address for reprints: ZU Khan, Department of Microbiology, Faculty of associated Exophiala oligosperma fungemia and provide Medicine, Kuwait University, Safat, 13110, PO Box 24923, Kuwait. a review of the literature. A total of 29 cases of Exophiala Email: [email protected] infection were identified, 14 of which had a diagnosis of malignancy. Of 23 patients who had a catheter in place, its removal was required in 20 cases. Two cases proved to THERAPEUTICS be fatal; in both of these, the catheter was retained.
Exophiala (Wangiella) is a black dematiaceous fungus of Issues related to the design and interpretation the phylum Ascomycota order Chaetothyriales and is of clinical trials of salvage therapy for invasive among the fungi that cause phaeohyphomycosis [1]. In mold infection addition to subcutaneous infections or mycetoma, Almyroudis NG, Kontoyiannis DP, Sepkowitz KA et al. Exophiala can cause disseminated infections such as Clin Infect Dis 2006;43:1449–55. fungemia and endocarditis. The authors of the current report describe the case of a pediatric patient with The authors of this article review the issues related to leukemia who developed catheter-associated fungemia the design and interpretation of clinical trials of salvage caused by the demateaceous fungus Exophiala antifungal treatments for invasive mold infections. oligosperma. Furthermore, they review the literature, Patients who are refractory to standard antifungal identifying 29 cases of Exophiala infection. Fourteen of treatments usually have a worse prognosis than those these patients had a diagnosis of malignancy. Catheter who have become intolerant; thus, the authors propose removal was required in 20 of 23 patients who had a that these two groups of patients should be analyzed catheter in place. In two cases that had a fatal outcome, separately in salvage therapy studies. Furthermore, the the catheter was retained. authors suggest a composite outcome assessment to The vast majority of catheter-related fungemia episodes define refractory infection. Other factors, such as occurring in cancer patients are due to Candida species [2]. heterogeneity of patients and pathogenic risk factors, However, a variety of other less common fungi have and the initial antifungal treatment are also discussed. also emerged as causes of bloodstream infections in cancer patients. Patients who are refractory to or have developed intolerance Rhodotorula is a yeast of the family Cryptococcaceae to standard antifungal therapy are generally considered for that has been strongly associated with catheter-related inclusion in trials of salvage antifungal therapy. The authors infections in patients with altered immune defenses, with of this article review issues related to the design and Rhodotorula mucilaginosa being responsible for the interpretation of such studies. majority of cases. In a recent review of the literature, 66 One of the common problems of salvage study design is cases of Rhodotorula fungemia were identified [3]. The the inclusion of standard therapy-refractory and -intolerant most common underlying disease was malignancy in 44 patients in the same group. Frequently, those who patients (67%). Catheter removal was required in 77% of experience toxicity to initial therapy are classed as cases and the mortality rate was 18%. “intolerant”, and often respond well to a subsequent agent. Trichosporon is another opportunistic yeast that causes Those who are refractory to initial therapy have a poorer bloodstream infection in susceptible populations [4]. Risk outcome; thus, the authors suggest that these two groups factors include leukemia, neutropenia, and high-dose should be analysed separately. Following on from this, corticosteroid use. Trichosporonosis carries a high mortality problems associated with variable definitions of refractory rate exceeding 50% in cancer patients. Several other fungi mold infections are discussed. The authors suggest that well- have been occasionally reported to cause fungemia in defined clinical, radiological, and laboratory criteria are cancer patients, including Saccharomyces, Acremonium, employed, and that the definition of refractory disease could Phialemonium, Fusarium, and Scedosporium. be limited to those with evidence of disease progression in
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 67 CLINICAL REVIEWS
response to standard therapy. Furthermore, they propose a was not more effective than placebo in improving clinical signs composite outcome assessment for refractory disease, in and symptoms in this group of patients. which there is a worsening at least two of three criteria These results confirm findings in other studies. For (clinical, radiological, and mycological), with patients instance, in a 3-month, prospective, open trial conducted in assessed for ≥7 days. Switzerland, nasally applied amphotericin B spray was A prolonged neutropenia is the major risk factor for ineffective in controlling the growth of nasal polyps [1]. invasive mold infection; however, many other risk factors Another recent study described the effect of topical may exist, and patients with a range of diseases (e.g. antifungal therapy on nasal cell activation markers recipients of allogeneic hematopoietic stem cell transplants, (eosinophil cationic protein and tryptase) in chronic advanced AIDS patients) are at risk of infection. Thus, a rhinosinusitis [2]. Nasal amphotericin B treatment did not heterogeneous group of patients is often included in have an effect on eosinophil cationic protein or tryptase salvage studies. The authors propose that randomized levels. Furthermore, eradication of fungal colonization did studies, with specified subgroup analyses should be not correlate with the levels of these activation markers of undertaken. Such studies would also allow subgroup nasal inflammatory cells. analysis on specific pathogens to be performed. However, However, these emerging data contradict earlier studies randomized studies are generally thought to be impractical showing that amphotericin B nasal irrigation may influence due to the requirement for a large number of patients the inflammatory process of chronic rhinosinusitis by for adequate subgroup analysis. Inclusion of a decreasing the production of interleukin-5 (IL-5) from nasal contemporaneous group of control subjects (rather than polyps compared with normal inferior turbinates [3]. Since a historical comparator group) should be considered as high IL-5 levels have been linked to the eosinophilic a useful alternative. inflammation that characterizes chronic rhinosinusitis with nasal polyps, there has been interest in investigating the Address for reprints: NG Almyroudis, Division of Infectious Diseases, Roswell Park Cancer Institute, Elm and Carlton Sts., Buffalo, NY 14263, effect of anti-IL-5 therapy on the growth of nasal polyps. In USA. Email: [email protected] a recent randomized controlled trial, Gevaert and colleagues found that a single intravenous infusion of reslizumab Amphotericin B nasal lavages: not a solution (humanized anti-human IL-5 monoclonal antibody) was for patients with chronic rhinosinusitis effective in reducing the size of nasal polyps in 50% of Ebbens FA, Scadding GK, Badia L et al. patients who received the drug [4]. J Allergy Clin Immunol 2006;118:1149–56. It can be concluded that specific anti-IL-5 therapy could be more effective in treating chronic rhinosinusitis than the non- The authors conducted a randomized, double-blind, specific antifungal therapy with intranasal amphotericin B. placebo-controlled trial to investigate the efficacy of 1. Helbling A, Baumann A, Hanni C et al. Amphotericin B nasal spray has no effect on nasal polyps. J Laryngol Otol 2006;120:1023–5. amphotericin B nasal lavages in the treatment of chronic 2. Weschta M, Rimek D, Formanek M et al. Effect of nasal antifungal therapy on nasal cell rhinosinusitis. They found that amphotericin B nasal activation markers in chronic rhinosinusitis. Arch Otolaryngol Head Neck Surg 2006;132:743–7. lavage was not effective in the treatment of chronic 3. Shin SH, Ye MK. Effects of topical amphotericin B on expression of cytokines in nasal rhinosinusitis with or without nasal polyps. polyps. Acta Otolaryngol 2004;124:1174–7. 4. Gevaert P, Lang-Loidolt D, Lackner A et al. Nasal IL-5 levels determine the response to anti-IL-5 treatment in patients with nasal polyps. J Allergy Clin Immunol In this randomized, double-blind, placebo-controlled trial 2006;118:1133–41. patients with chronic rhinosinusitis were randomized to receive μ Address for reprints: FA Ebbens, Department of Otorhinolaryngology, 25 mL of amphotericin B (100 g/mL) or placebo in each nostril Head and Neck Surgery, Academic Medical Center, Amsterdam, twice daily for 3 months. At the end of the study, amphotericin B The Netherlands. Email: [email protected]
68 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 The American Society of Hematology MEETING REPORT 48th Annual Meeting and Exposition
Orlando, FL, USA, December 9–12, 2006
Ana Berceanu and Raoul Herbrecht Department of Hematology and Oncology, Hôpitaux Universitaires de Strasbourg, Strasbourg, France
Several presentations at the American Society of Hematology Cryptococcus and Zygomycetes. Senn et al. monitored 1,3-β- annual meeting held in December 9–12, 2006 in Orlando, D-glucan levels in 189 episodes of neutropenia following FL, USA, were devoted to invasive fungal infections in induction or consolidation chemotherapy for acute leukemia patients with hematological malignancies or hematopoietic [2]. From a total of 320 febrile episodes, 31 invasive fungal stem cell transplantation (HSCT). Here, the most relevant of infections were diagnosed according to the European these abstracts are reported, with a focus on aspergillosis. Organization for Research and Treatment of Cancer/Mycoses All abstracts appeared in Blood 2006;108(16). Study Group (EORTC/MSG) criteria. A median of 15 samples was analyzed per episode of neutropenia. In patients with Host defense invasive candidiasis, the median time between onset of fever Neutrophils and monocytes are key cells in the host defense – as the first sign of invasive fungal infection – and 1,3-β-D- against Aspergillus infection. Their quantitative deficiency is glucan positivity was 0 days (range –3 to +17 days); this one of the best-known risk factors for developing invasive contrasted with 16 days (range 0 to +51 days) until diagnosis filamentous fungal infection. It is also accepted that of infection using conventional microbiological and imaging functional defects of phagocytic cells, such as in chronic techniques. In patients with invasive aspergillosis, the median granulomatous diseases or in alveolar proteinosis, also time to 1,3-β-D-glucan positivity was 3 days (range –9 to expose patients to opportunistic infection. Loeffler et al. +14 days) compared with a diagnosis at a median of 7 days demonstrate that the activation of neutrophil oxidative burst (range +1 to +21 days) using conventional methods. Overall, induced by Aspergillus is regulated by Toll-like receptor 2 the sensitivity and specificity of the test were 76% and 89%, (TLR2) signaling [1]. The blockade of TLR2 by an antibody respectively, using a 5-pg/mL cut-off value for 1,3-β-D- leads to a significant reduction in oxygen intermediates glucan, and 38% and 99%, respectively, using a cut-off of released by neutrophils after induction through exposure to 11 pg/mL. This study demonstrates, in a large number of zymosan or to Aspergillus. patients at high risk for fungal infection, the clinical utility of a 1,3-β-D-glucan detection test as an additional diagnostic Diagnostics tool in febrile neutropenia episodes. Mycological confirmation of an invasive fungal infection is often lacking in neutropenic patients despite the availability Breakthrough fungal infections of an Aspergillus galactomannan detection test in most Trifilio et al. reported their experience with the prophylactic centers and a recent update of diagnosis criteria. Detection of use of voriconazole in 71 allogeneic HSCT recipients [3]. The a 1,3-β-D-glucan antigen in serum has been used for many patients received voriconazole prophylaxis as they were at years in Japan, but few centers in Western countries have very high risk of invasive aspergillosis infection (history of a experience of using this test. 1,3-β-D-glucan is a cell wall previous episode or steroid therapy due to graft-versus-host component of most fungal pathogens, with the exception of disease [GVHD]). The length of voriconazole therapy was 6–956 days (median 133) with a total number of patient- Address for correspondence: Raoul Herbrecht, Department of days on voriconazole of 13 805. A total of 10 breakthrough Hematology and Oncology, Hôpital de Hautepierre, Hopitaux invasive fungal infections were observed. Candida glabrata Universitaires de Strasbourg, 67098 Strasbourg, France. was the most common pathogen (n=5) followed by Rhizopus Email: [email protected] spp. (n=2), C krusei, Mucor sp. and Cunninghamella sp. (one
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 69 ANA BERCEANU AND RAOUL HERBRECHT
of each). The 1-year probability of breakthrough invasive chemotherapy than during consolidation treatment in the fungal infection was estimated to be 18%. No Aspergillus preemptive therapy arm (16.4% and 1.4%, respectively) infections were observed. All patients who developed a while the rates were similarly low in the empiric therapy arm Candida breakthrough infection had steady-state trough (3.9% and 1.4%, respectively). Therefore, it can be serum levels of voriconazole of <2 μg/mL. The authors concluded from these two presentations that there is no conclude that: overall difference between both strategies in terms of survival rates, but also that, unexpectedly, there is no • Voriconazole shows excellent efficacy in the prevention decrease in medication costs using the preemptive treatment of aspergillosis. strategy compared with the empiric strategy. • Zygomycetes and C Glabatra breakthrough infections Recently published trial results demonstrated the are a concern in voriconazole-treated patients, although superiority of posaconazole over fluconazole in the the incidence is low. prophylaxis of proven or probable invasive aspergillosis in • Serum levels of voriconazole should be monitored in patients with GVHD, and in preventing invasive fungal order to reduce the possibility of breakthrough infections infections and reducing overall mortality rate in high-risk with fungi that would otherwise be susceptible neutropenic patients [6,7]. Thus, O’Sullivan et al. conducted to voriconazole. a separate cost-effectiveness analysis of antifungal prophylaxis (from a US perspective) in high-risk neutropenic Treatment strategies patients [8]. Posaconazole was found to be associated with Empiric antifungal therapy is a well-accepted strategy in slightly lower costs (US$4887 vs. US$5070) per patient neutropenic patients who have persistent fever. Recent compared with standard azole therapy over a lifetime improvements in the diagnosis and treatment of invasive perspective. Further sensitivity analyses demonstrated that fungal infections suggest that preemptive therapy may be a there was a 75% probability of posaconazole being cost more cost-effective strategy, with fewer patients receiving saving compared with standard azole therapy. The authors excessive treatment and no decrease in efficacy. Cordonnier concluded that posaconazole is likely to be more cost- et al. compared the two therapeutic strategies in patients at effective than standard azole therapy in the prophylaxis of high risk for fungal infection [4]. A total of 293 adult invasive fungal infections in high-risk neutropenic patients. patients with an expected prolonged severe neutropenia Barron et al. retrospectively assessed compliance with were randomized to receive empiric therapy (neutropenia published guidelines for prevention of fungal infection in 242 and persistent fever) or preemptive therapy (onset of HSCT recipients hospitalized in 13 US hospitals [9]. Fifty- antifungal therapy only in the case of pneumonia, severe seven percent of the patients were classified as high risk for mucositis, septic shock, sinusitis, skin lesions suggesting invasive fungal infection. Overall, 85% of the patients filamentous fungal infection, Aspergillus colonization, or a received antifungal prophylaxis, with no difference between positive galactomannan test). The primary endpoint of the proportions of those at high risk and low risk of invasive survival was not inferior in the preemptive therapy arm fungal infections receiving prophylaxis (84% and 86%, (136/143) compared with the empiric therapy arm respectively). The majority of patients received fluconazole (147/150). Although more fungal infections were diagnosed (84%). Compliance with national guidelines for antifungal in the preemptive therapy arm (13/143 vs. 4/150; p<0.02) prophylaxis was low, at only 54%. Prophylaxis failed in 29% the fungal infection-related mortality rate was similar in both of the high-risk patients and in 6% of the low-risk patients. arms (three in the preemptive therapy arm and none in the This study demonstrated a poor compliance with guidelines empiric therapy arm). An economic analysis was performed and the authors stated that prophylaxis was both under- and in the same set of patients by Schwarzinger et al. [5]. Total over-utilized. The reasons for poor adherence to guidelines medication costs were computed from individual patient were not identified in this study. records during their hospital stay. Overall, medication costs were not different in both arms. However, an analysis Monitoring of plasma levels of according to the type of chemotherapy (induction versus antifungal agents consolidation) showed a significant decrease in cost of Plasma levels of newer azoles may differ significantly preemptive therapy during consolidation chemotherapy between patients. Explanations for these differences include (€1387 vs. €2610 for empiric therapy; p<0.02) but not variability in oral bioavailability, genetic polymorphisms in during induction chemotherapy (€5714 vs. €4793; cytochrome isoenzymes, and drug–drug interactions. Data non significant). The difference could be explained by a on plasma levels of antifungals are scarce in clinical practice. higher invasive fungal infection rate during induction Trifilio et al. monitored voriconazole levels in patients with
70 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 ASH 2006
hematological malignancies, mostly allogeneic HSCT patients had a non-malignant disease. Overall, the 12-week recipients, receiving 400–800 mg/day (2.0–16.3 mg/kg) oral survival rate was 52.3%. Nine factors were found to be voriconazole [10]. Steady-state plasma levels of voriconazole associated with a poor overall survival: allogeneic HSC or were measured after ≥5 days of therapy. Median plasma solid organ transplantation, progression of underlying cancer, levels were 1.2 μg/mL (range undetectable to 12.5 μg/mL). prior noninfectious respiratory disease, steroid therapy (doses Fifteen percent of the patients had undetectable levels of of ≥0.2 mg/kg), renal impairment, a low monocyte count, voriconazole in plasma and an additional 12% had levels pleural effusion, dissemination of the aspergillosis, and between 0.2 and 0.5 μg/mL. According to published data probable or proven aspergillosis (compared with possible suggesting a higher efficacy in patients achieving plasma cases). This is the largest study on prognosis factors that has levels >0.5 μg/mL [11], 27% of the patients treated orally been performed in a nonselected patient population to date. with voriconazole did not achieve optimal levels. Nonlinear It underscores the importance of some factors not previously pharmacokinetics does not allow prediction of plasma levels associated with a poor outcome in invasive aspergillosis, such from the daily dose. In another presentation (discussed as steroid therapy, renal impairment, or the presence of earlier), the same group related low plasma levels of noninfectious respiratory disease. voriconazole to breakthrough Candida infections [3]. Although there was no correlation between plasma levels Disclosures and efficacy or adverse events in this presentation, this study Professor Herbrecht has recieved grants/honoraria from Astellas, Gilead, strongly suggests that monitoring the plasma levels of Merck Sharp & Dohne, Pfizer, Schering-Plough, and Zeneus Pharma. voriconazole to avoid subtherapeutic levels in a substantial Dr Berceanu has no relevant financial interests to disclose. proportion of patients at high risk of severe invasive fungal infections be made. References 1. Loeffler J, Mezger M, Wozniok I et al. Toll-like receptor 2 signalling regulates oxidative burst of human neutrophils, but not cytokine response of dendritic cells after contact with Outcome Aspergillus fumigatus. Blood (ASH Annual Meeting Abstracts) 2006;108:3854. Woods et al. proposed the Aspergillus galactomannan 2. Senn L, Robinson JO, Schmidt S et al. Monitoring of 1,3-beta-d-glucan (BGL) antigenemia in neutropenic patients with acute leukemia at high risk for invasive fungal infections (IFI). detection test as a surrogate marker of outcome in invasive Blood (ASH Annual Meeting Abstracts) 2006;108:2924. aspergillosis [12]. Thirty patients, mostly myeloma patients, 3. Trifilio S, Singhal S, Williams S et al. Breakthrough fungal infections after allogeneic hematopoietic stem cell transplantation in patients on prophylactic voriconazole. with respiratory tract aspergillosis tested positive for the Blood (ASH Annual Meeting Abstracts) 2006;108:2849. Aspergillus-specific polysaccharide galactomannan. Among 4. Cordonnier C, Pautas C, Maury S et al. Empirical versus pre-emptive antifungal therapy in high-risk febrile neutropenic patients: a prospective randomized study. Blood (ASH Annual 25 neutropenic patients, the persistence of a positive Meeting Abstracts) 2006;108:2019. galactomannan test was associated with death (five deaths 5. Schwarzinger M, Beauchamp C, Maury S et al. Empirical versus pre-emptive antifungal therapy in high-risk febrile neutropenic patients: an economic analysis. Blood (ASH Annual out of five cases), while a return to a negative value Meeting Abstracts) 2006;108:2021. predicted survival (20 patients surviving out of 20 cases). 6. Ullmann AJ, Lipton JH, Vesole DH et al. Posaconazole or fluconazole for prophylaxis in severe graft-versus-host disease. N Engl J Med 2007;356:335–47. Similarly, a correlation was found in non-neutropenic 7. Cornely OA, Maertens J, Winston DJ et al. Posaconazole vs. fluconazole or itraconazole patients (one death with persistent positive galactomannan prophylaxis in patients with neutropenia. N Engl J Med 2007;356:348–59. 8. O’Sullivan AK, Pandya A, Thompson D et al. Cost-Effectiveness of Posaconazole Versus test and no deaths in four patients whose galactomannan Standard Azole Therapy in the Prevention of Invasive Fungal Infections among High-Risk test became negative). This study confirms the previously Neutropenic Patients in the U.S. Blood (ASH Annual Meeting Abstracts) 2006;108:3311. 9. Barron MA, Fishman N, Baillie GM et al. Multicenter evaluation of fungal prophylaxis in reported correlation between a persistent positive hematopoietic stem cell transplantation. Blood (ASH Annual Meeting Abstracts) galactomannan and a poor outcome [13]; however, it does 2006;108:5293. 10. Trifilio S, Pennick G, Pi J et al. Monitoring plasma voriconazole levels may be essential to not address the issue of lack of sensitivity of galactomannan avoid subtherapeutic levels. Blood (ASH Annual Meeting Abstracts) 2006;108:594. testing at baseline, and thus, it cannot be considered as a 11. http://www.fda.gov/ohrms/dockets/ac/01/briefing/3792b2_02_FDA-voriconazole.htm unique outcome assessment of invasive aspergillosis. 12. Woods G, Miceli M, Grazziutti M et al. Validation of serum Aspergillus galactomannan index as a surrogate endpoint for outcome of invasive aspergillosis: clinical and research Nivoix et al. presented a multivariate analysis of implications. Blood (ASH Annual Meeting Abstracts) 2006;108:2861. pretreatment prognosis factors for survival in 303 13. Herbrecht R, Johnson EM, Barette M et al. Longitudinal assessment of galactomannan antigenemia and relationship with outcome in invasive aspergillosis. Proceeding of the 1st oncohematological patients with possible, probable, or Conference on Advances Against Aspergillosis, 9–11 September 2004, San Francisco, CA, USA. definite invasive aspergillosis [14]. Most patients had acute 14. Nivoix Y, Velten M, Letscher-Bru V et al. Prognostic factors for survival in invasive aspergillosis in adult oncohematological patients. Blood (ASH Annual Meeting Abstracts) leukemia, HSCT, or hematological malignancy. Twenty-one 2006;108:2852.
THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 71 Forthcoming International Events 2007
MAY 25–27 SEPTEMBER 2007 Annual Conference on 30 Antimicrobial Resistance 17–20 Nordic Society for Medical Mycology Bethesda, MD, USA 47th Annual Interscience Conference 4th Scientific Meeting Contact: Sharon Cooper-Kerr on Antimicrobial Agents Helsinki, Finland T +301 656 0003 and Chemotherapy (ICAAC) Contact: Malcom Richardson F +301 907 0878 Chicago, IL, USA T +358 9 191 26894 E [email protected] Contact: American Society F +358 9 191 26382 W www.nfid.org for Microbiology E [email protected] T +202 942 9248 W www.nsmm.nu 27–29 E [email protected] Critical Care 2007 W www.icaac.org JUNE London, UK Contact: Amy Tranter 29–30 4–7 T +20 75 016 711 3rd International Transplant 2nd Mediterranean Congress on E [email protected] Infectious Disease Conference Critical Care Medicine Prague, Czech Republic Brijuni Islands, Croatia JULY Contact: International Medical Press Contact: Vladimir Gasparovic T +44 207 398 0700 T +38 512 388 257 9–13 F +44 207 398 0701 E [email protected] Australian Society for Microbiology 2007 Annual Scientific Meeting OCTOBER 17–20 Adelaide, SA, Australia Canadian Society of Microbiologists Contact: Janette Sofronidis 28–31 CSM 57th Annual Meeting T +61 3 9867 8699 3rd Trends in Medical Mycology Quebec City, QB, Canada E [email protected] (TIMM-3) Contact: Dr. Hani Antoun W www.asm2007.org Torino, Italy E [email protected] Contact: Rob Zikkenheimer W www.csm-2007.ulaval.ca/csm.html AUGUST T +31 73 690 1415 F +31 73 690 1417 24–28 24 E [email protected] APIC's 34th Annual Conference 12th Annual Challenges W www.timm2007.org & International Meeting in Critical Care San Jose, CA, USA Hershey, PA, USA Contact: Gloria Whitaker Contact: Penn State College T +202 4542638 of Medicine F +202 789 1899 T 717 531 6483 E [email protected] F 717 531 5604 W www.apic.org E [email protected]
If you would like your meeting listed here, please contact the Publisher (for details see Contents page).
72 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS VOL 1 NO 2 2007 THE JOURNAL OF INVASIVE FUNGAL INFECTIONS
Reader Survey – Let Us Know What You Think! Please take a few moments to complete this survey. We would value your opinion.
Please photocopy this page, complete the survey below, and fax it back to Remedica on +44 (0)20 7759 2951. Or you can visit THE JOURNAL OF INVASIVE FUNGAL INFECTIONS website and complete the survey online (registration online is FREE): www.invasivefungalinfections.com
1. We are aiming to provide practical information for specialists in fungal infections, and allied healthcare professionals. How would you rate the information presented in this issue?
Strongly agree Strongly disagree
a) The technical quality of information included in THE JOURNAL OF INVASIVE FUNGAL INFECTIONS was acceptable: 1 2 3 4 5
b) The information was relevant to my practice: 1 2 3 4 5
c) The information was presented clearly: 1 2 3 4 5
d) The leading articles provided new information regarding the understanding and treatment of invasive fungal infections: 1 2 3 4 5
e) The clinical review section was helpful and I would like to see analyses in future issues: 1 2 3 4 5
2. Did you learn anything through the CME/CPE activity THE JOURNAL OF INVASIVE FUNGAL INFECTIONS that will change the way you practice medicine? I Yes I No
If so, what? ......
3. Is there anything you learned from the CME/CPE activity THE JOURNAL OF INVASIVE FUNGAL INFECTIONS that prompts you to seek further information that may influence the way you practice medicine in the future? I Yes I No
If so, what? ......
4. Would you like to recommend THE JOURNAL OF INVASIVE FUNGAL INFECTIONS to a colleague? I Yes I No
My colleague’s email is: ......
5. What specific topics do you think should be covered in future issues?
......
Name ...... Job title ......
Institution ......
Address ......
Country ...... Post/zip code ......
Email ......