Infections Post Transplant Candida Glabrata and Candida

Home , Candida glabrata

Bone Marrow Transplantation (2001) 28, 873–878  2001 Nature Publishing Group All rights reserved 0268–3369/01 $15.00 www.nature.com/bmt Infections post transplant Candida glabrata and Candida krusei fungemia after high-risk allogeneic marrow transplantation: no adverse effect of low-dose ﬂuconazole prophylaxis on incidence and outcome

A Safdar1, F van Rhee2, JP Henslee-Downey2, S Singhal2 and J Mehta2

1Division of Infectious Diseases, Department of Medicine, University of South Carolina School of Medicine, Columbia, SC, USA; 2Division of Transplantation, Department of Medicine, University of South Carolina School of Medicine, and Palmetto-Richland Memorial Hospital, Columbia, SC, USA

Summary: Keywords: C. glabrata; C. krusei; BMT; fungemia; fluconazole prophylaxis Candidemia is a serious complication in patients following allogeneic blood, marrow, and organ transplantation. Fourteen patients developed nosocomial fungemia among 204 allogeneic marrow transplants Nosocomial fungemia due to Candida species has steadily performed during 1997–1999. Incidence of hematogen- increased over the past four decades.1–4 In recipients of ous candidiasis was 6.8 per 100 allogeneic BMT. All 14 blood and marrow transplantation, systemic candidiasis is had an indwelling central venous catheter (CVC) and a serious complication associated with high morbidity and fluconazole (100–200 mg daily) was given prophylac- mortality.5,6 Antifungal prophylaxis with fluconazole dur- tically. In 11 (78.5%) neutropenic patients, duration ing the early post-transplantation period has been shown between agranulocytosis and diagnosis of fungemia was to significantly reduce the incidence of candidemia,7 and median, ؎ s.d.) 10 ؎ 8 days. Candida glabrata (53.3%) improve short-term survival.8 Fluconazole prophylaxis for) was the most common yeast species, followed by C. extended periods (Ͼ75 days) in recipients of high-risk krusei (33.3%), and C. parapsilosis (13.3%). Candida marrow transplantation was also beneficial.9 albicans was conspicuously absent. Ten patients (71.4%) Breakthrough infections in patients receiving triazole- had primary transplant-related complication (Ͼ2 days) based antifungal prophylaxis, and changes in the spectrum including hemolytic uremic syndrome/thrombotic of Candida species have been reported sporadically during -severe the last decade.10,11 Increase in the non-C. albicans funge ,(5 ؍ thrombocytopenic purpura (HUS/TTP) (n mia in this setting was probably multifactorial although .(2 ؍ and bacteremia (n ,(3 ؍ hemorrhagic cystitis (n Seven (50.0%) patients expired and in three (21.4%) selection pressure due to fluconazole and itraconazole may deaths were attributed to fungemia. The impact of a have played an important role. As most nosocomial funge- primary transplant-related complication on short-term mia arises from the host’s endogenous flora, transient or permanent changes in the colonizing yeast species in (0.07 ؍ survival in this setting was not significant (P (HUS/TTP (P Ͼ 0.5); neutropenia (P Ͼ 0.5); GVHD (P patients receiving these antimicrobial prophylaxis, was not Removal of CVC did not alter outcome in our unexpected.12,13 However, emergence of de novo resistance .((0.35 ؍ group (P у0.5) although in patients with persistent fun- in fluconazole susceptible C. albicans following prophy- gemia (Ͼ72 h), and those with preceding bacteremia, laxis with triazoles was worrying.14 Conven- At our bone marrow transplantation center, low-dose .(0.002 ؍ mortality was significantly higher (P tional prognosticators of poor outcome did not fluconazole (100–200 mg daily) is routinely included in the adversely effect short-term survival in our transplant early post-transplant prophylaxis regimen. We report, recipients with hematogenous candidiasis. The predomi- characteristics of nosocomial fungemia in patients undergo- nance of C. glabrata and C. krusei breakthrough infec- ing high-risk allogeneic marrow transplantation during tions was similar to what is seen with high-dose flucon- 1997–1999. azole (400 mg) prophylaxis, and no adverse effects of low-dose fluconazole in terms of increased incidence of non-susceptible Candida species was seen. Bone Marrow Patients and methods Transplantation (2001) 28, 873–878. Study design Correspondence: Dr A Safdar, Division of Infectious Diseases, Depart- ment of Medicine, University of South Carolina School of Medicine, Two Retrospective review of Candida bloodstream invasion was Medical Park, Suite-502, Columbia, SC 29203, USA performed in patients following allogeneic marrow trans- Received 20 April 2001; accepted 23 August 2001 plantation at Palmetto-Richland Memorial Hospital Cancer Candidemia in BMT recipients on fluconazole prophylaxis A Safdar et al 874 Center, Columbia, SC, from 1 January 1997 to 31 (50 mg/kg) given daily ϫ two doses; high-dose methypred- December 1999. Patient information, and laboratory results nisolone (1 g/m2) administered every 12 h ϫ four doses. was retrieved from patient charts and computerized hospital Low-dose cyclosporin (3 mg/kg) was started day Ϫ1 and data systems. All blood culture specimens were processed levels were maintained between 100 and 200 as measured at the Microbiology Laboratory, Palmetto-Richland by monoclonal antibody technique and switched to the oral Memorial Hospital, SC, USA. form after day ϩ21 and weaned gradually through the first year post-BMT; moderate-dose methylprednisolone (30 mg/m2) administered before antithymocyte globulin (ATG); Definitions ATG (10 mg/kg) was given daily ϫ 12 doses on day ϩ5 to ϩ16, and steroid dose was tapered 10% weekly and Candidemia was defined as a clinical illness associated with switched to prednisone orally after day ϩ21. Patients were the presence of Candida in patients bloodstream. Presence housed in a positive-pressure high efficiency particulate air of Candida in blood culture was considered clinically sig- (HEPA)-filtered BMT unit. All patients received granulo- nificant even in the absence of clinical signs or symptoms, cyte colony-stimulating factor (G-CSF) from day ϩ1 until especially in individuals at risk for invasive candidiasis the white blood cell (WBC) count exceeded 5000/␮l. Flu- such as agranulocytosis, uremia, hyperalimentation, pro- conazole 100 to 200 mg daily was given prophylactically longed exposure to multiple antibiotics, persistent hypergly- either orally or intravenously in patients intolerant of oral cemia, uremia, severe mucositis involving oro-pharyngeal medication. Broad-spectrum antibiotics, intravenous Ig, mucosa, and those receiving adrenal glucocorticoids.15,16 At inhaled pentamidine or trimethoprim-sulfamethoxazole and least one positive blood culture for a single Candida species antiviral agents were used as indicated.20 during the same hospitalization was regarded as an episode. Two patients underwent MSD marrow graft transplan- All consecutive blood cultures with Candida during the tation. Conditioning consisted of TBI 1200 cGy, etoposide same hospitalization obtained within 72 h were considered (50 mg/kg) ϫ 1 day, and cyclophosphamide (50 mg/kg) as a single episode. If more than one Candida species given on day Ϫ3 and Ϫ2. Graft-versus-host disease prophy- was isolated, infection due to each was considered as a laxis was initiated on day Ϫ1, cyclosporin (3 mg/kg) to separate episode. maintain serum level between 300 and 400 ng/ml; metho- Breakthrough fungemia was defined as the presence of trexate on day ϩ1 (15 mg/m2), and then reduced to 10 Candida in a patient’s bloodstream while receiving prophy- mg/m2 on days ϩ3, ϩ6 and ϩ11. lactic fluconazole (100 to 200 mg daily) for greater than 48 h.17 Persistent infection was defined as presence of Can- dida in blood culture specimens for greater than or equal Microbiology to 72 h while receiving appropriate antifungal therapy. All patients with persistent fungemia were evaluated for sec- Presumptive identification of C. albicans was performed by ondary ocular, endovascular, and joint infections. Neutro- germ-tube test, organisms that failed to form a germ tube penia was defined as absolute neutrophil count (ANC) following 3 h of incubation at 37°C in sterile horse serum Ͻ500 cells/mm3 and ANC of Ͻ100 cells/mm3 was were further tested by an auxanographic plate method (BBL regarded as profound granulocytopenia. Agranulocytosis Microbiology Systems, Cockeysville, MD, USA). The Can- was a non-detectable neutrophil count in peripheral circu- dida species not identified by the above methods were sub- lation. The onset of granulocytopenia and detection of Can- jected to additional standard testing including YBC Yeast dida species in blood culture specimens is reported as the Identification Card (BioMerieux Vitek, Hazelwood, MO, duration of neutropenia and expressed in days.18 USA) and APIC Biochemical Strip (BioMerieux Vitek). Infection is interchangeably used for hematogenous can- Mycology laboratories at the Department of Health and didiasis, and granulocytopenia for neutropenia throughout Environmental Control (DEHC, Columbia, SC, USA) pro- the text of this article. Mortality was defined as death vided extra-institutional reference for yeast not identified attributable to fungemia either directly or indirectly and in by the above methods. the absence of other known terminal events such as intra- cranial hemorrhage, or septic shock resulting from infections other than invasive candidiasis.19 Severe hemorrhagic Antifungal therapy cystitis requiring transfusion of blood and blood products, All clinically ill patients with candidemia received treat- clinically stable HUS/TTP, fungal pneumonia, and rectal ment with amphotericin based preparations including Abel- abscesses were not considered terminal events. cet (3–5 mg/kg daily) (The Lipsome Company, Princeton, NJ, USA) and Ambisome (3–5 mg/kg daily) (Fujisawa Conditioning regimen and GVHD prophylaxis Healthcare, Deerfield, IL, USA). Upon initial identification of Candida species, intravenous fluconazole (100 to 200 Twelve of 14 patients who lacked a matched sibling donor mg) was given through all three lumens of the central (MSD) underwent partially matched related donor (PMRD) venous catheter during the first 24 h provided the patient marrow graft transplantation. In order to improve was not ill. During this period a decision was made regard- engraftment, total body irradiation (TBI) was increased ing possible colonization of the central venous catheter or from 1332 cGy to 1500 cGy. Etoposide (20 mg/kg) was serious systemic candidiasis. In patients that fell into the administered once; cytosine arabinoside (3 g/m2) admini- latter category, treatment with an amphotericin-based agent stered twice daily ϫ six doses; cyclophosphamide was started promptly.

Bone Marrow Transplantation Candidemia in BMT recipients on fluconazole prophylaxis A Safdar et al 875 Removal of intravascular catheter In four patients acute graft-versus host disease (GVHD) was diagnosed prior to the onset of infection. Chronic The indwelling central venous catheter was removed GVHD and veno-occlusive disease (VOD) was present in immediately in severely ill patients, or those receiving high- one each. The median duration from marrow transplan- dose corticosteroids for GVHD or with ATG prophylaxis. tation to the diagnosis of fungemia was 12 Ϯ 288 days, In CVCs not removed, antifungal treatment was given via including patient 13 (Table 2) who developed fungemia 6 all three lumens. All intravascular lines (central and weeks after receiving chemotherapy for relapsed Hodgkin’s peripheral) were removed in patients with candidemia for disease. An indwelling central vascular catheter (CVC) was longer than 48 h. present in all patients for longer than 7 days. During these 36 months, 15 episodes of candidemia Statistical analysis including two episodes in patient 10 were observed (Table The chi-square test and Fisher’s exact test were used to 2). In six patients (42.9%) fungemia persisted for greater determine categorical predictors of nosocomial fungemia. than or equal to 72 h. Eight episodes of C. glabrata (53.3%) Differences between the categorical variables were con- were most common, followed by five C. krusei (33.3%) and sidered significant if the two-tail P value was less than 0.05. two C. parapsilosis (13.3%). A primary transplant-related complication was identified in 10 individuals (Table 2). Five developed acute hemolytic uremic syndrome/ Results thrombotic thrombocytopenic purpura (HUS/TTP), three had severe hemorrhagic cystitis requiring blood trans- Two hundred and four allogeneic marrow transplants were fusions, and two were bacteremic. Candida esophagitis performed from 1 January 1997 to 31 December 1999. (patient 6) and relapsing rectal abscess requiring multiple Fourteen patients developed nosocomial candidemia while surgical debridement preceded fungemia in patient 10 receiving fluconazole (150 mg median daily dose) for (Table 2). greater than 7 days. The incidence of hematogenous can- Seven (50%) responded to therapy. Others expired 3 to didiasis was 6.8 per 100 BMT; 8.4 per 100 mismatched- 24 days from the last positive blood culture for a yeast. related, and 6.2 per 100 matched-related allogeneic marrow All patients in the latter group had an underlying primary graft transplants. In 30 matched-unrelated marrow graft transplant-related complication (P ϭ 0.07) (Table 2). Both recipients, and 63 patients that underwent autologous per- patients with bacteremia that subsequently became fung- ipheral stem cell reinfusion, fungemia was not observed. emic expired (P ϭ 0.002). Underlying agranulocytosis (P Eleven (78.6%) were profoundly neutropenic (ANC у0.5), acute GVHD (P ϭ 0.07), HUS/TTP (P у0.5), or Ͻ 3 100/mm ) and the interval between onset of neutropenia hemorrhagic cystitis (P Ͼ0.5) did not confer poor outcome. and isolation of Candida from blood cultures was (median Death was attributed directly or indirectly to hematogenous Ϯ Ϯ s.d.) 10 8 days. Twelve received partial HLA-mis- candidiasis in three patients and the duration was 3 to 7 matched marrow grafts from related donors (PMRD). Two days. Three patients (75%) expired among four that underwent matched-related donor (MRD) graft transplan- underwent removal of an indwelling central intravascular tation. The underlying cancer and patient demographic catheter. In our group intravascular catheter exchange was information is summarized in Table 1. not associated with improved survival (P у0.5).

Table 1 Characteristics of 14 high-risk allogeneic marrow transplantation recipients with fungemia Discussion n 14 Sex (%) Bloodstream infections in patients with profound neutro- Male 8 (57.1) penia are associated with prolonged hospitalization, and Female 6 (42.9) poor short-term survival.21,22 Systemic fungal infections are Ϯ Ϯ Age (mean s.d. years) 31.9 14.2 serious complications as they carry higher morbidity and WBC: absolute neutrophil count Ͻ100 cells/mm3 (%) 11 (78.6) mortality compared with frequently isolated nosocomial Underlying cancer (%) 14 (100) Acute myeloid leukemia 7 (50) blood borne pathogens such as Staphylococcus aureus, and 23,24 Acute lymphoblastic leukemia 4 (28.6) Pseudomonas aeruginosa, etc. Chronic myeloid leukemia 2 (14.3) The initial strategies in preventing systemic yeast infec- Multiple myeloma 1 (7.1) tions with nystatin and ketoconzole in high-risk patients Marrow grafts (%) were promising.25 During the early 1990s prophylaxis with Partially mismatched, related donor (PMRD) 12 (85.7) fluconazole became increasingly common due to improved 2 HLA-mismatched grafts 9 (64.3) 26,27 3 HLA-mismatched grafts 3 (21.4) bioavailability and preferred toxicity profile. A decline Matched related donor (MRD) 2 (14.3) in systemic candidiasis (approximately seven cases per 100 T cell depletion (%) 13 (92.9) allogeneic transplants) was observed in patients treated 8 Systemic corticosteroids (%) 14 (100) with fluconazole during high-risk periods. This compared favorably with the incidence of candidemia in our patients Graft-versus-host disease (%) 5 (35.7) Acute GVHD 4 (28.6) (6.8/100 allogeneic BMT) in whom low-dose fluconazole Chronic GVHD 1 (7.1) (100–200 mg daily) was routinely given prophylactically Veno-occlusive disease (VOD) (%) 1 (7.1) during the early post-transplantation period. This decline in the hematogenous candidiasis was

Bone Marrow Transplantation Candidemia in BMT recipients on ﬂuconazole prophylaxis A Safdar et al 876 Table 2 Characteristics of breakthrough fungemia in patients receiving low-dose ﬂuconazole prophylaxis

Age/Sex Cancer-BMT ANC Preceding Candida spp. Post-BMT Persistence GVHD Outcomea Cause of death Ͻ100/mm3 complications days (h) acute

1 30 F AML-PMRD ϩ Staphylococcus aureus C. krusei 272— Expired, day 16 Bacterial sepsis ϩ Bacteroides spp. sepsis 2 23 F ALL-PMRD ϩ Fungal pneumonia C. glabrata 11 120 — Expired, day 10 Relapsed ALL 3 52 M CML-PMRD ϩ HUS/TTP C. krusei 6 —— Alive 4 23 M ALL-PMRD Ϫ HUS/TTP ϩ cystitis C. glabrata 30 — ϩ Expired, day 3 Candidemia 5 44 F MM-PMRD Ϫ HUS/TTP C. glabrata 42 72 ϩ Expired, day 7 Candidemia 6 11 M ALL-MRD ϩ Yeast oesophagitis C. 216 72 ϩ Expired, day 22 Relapsed ALL parapsilosis 7 19 M AML-PMRD ϩ HUS/TTP ϩ CNS C. glabrata 11 120 ϩ Expired, day 3 Candidemia 8 49 M AML-PMRD ϩ Cystitis C. glabrata 59 —— Alive 9 33 M CML-PMRD Ϫ HUS/TTP ϩ cystitis C. glabrata 85 —— Alive 10 31 F AML-PMRD ϩ Rectal abscesses C. krusei 11 168 — Expired, day 24 Intra-cranial ϩ C. glabrata hemorrhage 11 46 M AML-PMRD Ϫ None C. glabrata 13 —— Alive 12 11 F ALL-MRD ϩ None C. krusei 6 —— Alive 13 25 M AML-PMRD ϩ None C. krusei 3 years —— Alive 14 50 F AML-PMRD ϩ None C. 3 —— Alive parapsilosis

Patients age is given in years. AML ϭ acute myelogenous leukemia; ALL ϭ acute lymphocytic leukemia; CML ϭ chronic lymphocytic leukemia; MM ϭ multiple myeloma; PMRD ϭ partially matched related donor graft; MRD ϭ matched related donor graft; CNS ϭ coagulase-negative staphylococcus. Cystitis ϭ severe hemorrhagic cystitis due to adenovirus or polyoma virus. aOutcome – death (in days) from the last positive blood culture for Candida; Patient 13 developed candidemia 6 weeks after receiving chemotherapy for relapsed Hodgkin’s disease and 11 days later presented with bilateral C. krusei septic arthritis of knee joints.

accompanied by reports of a rise in ‘difficult to treat’ break- proportion of fungemia due to C. parapsilosis was less than through systemic fungal infections.6,9–11 A change in the 15% (Table 2). This was in contrast to a sharp rise in C. spectrum of infections due to Candida species with either parapsilosis bloodstream infections reported in patients at low-triazole susceptibility profile or yeast with intrinsic cancer centers from the USA, Latin America, and Eur- resistance to fluconazole was concerning.28–30 The disap- ope.34–37 The reason for this increase in C. parapsilosis pearance of C. albicans and predominance of C. glabrata infections among non-surgical, non-pediatric, high-risk and C. krusei systemic infections in our hospitalized mar- patients with an underlying malignancy remains row transplantation recipients during recent years was not uncertain.34 unexpected. This shift in Candida species towards virulent During the past 3 years, half of our fungemic transplant yeast strains with restricted triazole susceptibility profiles recipients expired. In three, death was attributed due to fun- was linked with prior exposure to fluconazole used for gemia (Table 2). The overall response to antifungal therapy prophylaxis or prolonged pre-emptive therapy.30–32 in our group was good. This was in contrast to higher mor- The near exponential increase in hematogenous candidi- tality reported in patients with C. glabrata and C. krusei asis due to the non-albicans Candida species may be in systemic infections at another center.2,6,10,11,38 part due to a shift in the spectrum of yeast colonization in During this retrospective study, 71% had a preceding pri- these high-risk individuals. Recently, we reported a marked mary marrow transplant-related complication (Table 2). increase in colonization due to C. glabrata in patients Survival in patients with an ongoing complication such as undergoing treatment for cancer at a comprehensive cancer acute GVHD (P ϭ 0.07), severe hemorrhagic cystitis center in New York.30 In patients with hematologic mali- requiring blood transfusions (P Ͼ0.5), HUS/TTP (P Ͼ0.5), gnancy and those undergoing blood and marrow trans- and profound granulocytopenia (P у0.5) was not signifi- plantation, short-term survival was severely reduced in the cantly altered. Surprisingly, removal of indwelling intravas- presence of clinically significant colonization due to C. gla- cular catheters did not significantly change outcome in our brata and C. krusei.33 A shift in the spectrum of species patients either (P у0.5). However, in patients with Candida associated with systemic candidiasis appears to directly cor- isolation for longer than 72 h following initial diagnosis relate with patients endogenous yeast reservoir.12,13 We sus- (persistent infection), short-term survival was markedly pect that disappearance of bloodstream infections due to C. compromised (P ϭ 0.002). Similarly, patients with preced- albicans in our highly susceptible patient population in part, ing bacteremia had a higher mortality in this setting (P reflect a change in endogenous Candida reservoirs, and also ϭ 0.002). a selective protection provided by low-dose fluconazole In conclusion, an overall benefit of low-dose fluconazole against systemic disease due to C. albicans, C. tropicalis, prophylaxis appears to be comparable to experience and other Candida species.30–33 reported with higher doses at other centers.8,9 We did not Breakthrough infections due to C. parapsilosis in adults observe an unfavorable outcome in patients with systemic are infrequent and during this study period we observed the candidiasis due to highly virulent yeasts or Candida species

Bone Marrow Transplantation Candidemia in BMT recipients on fluconazole prophylaxis A Safdar et al 877 with restricted antifungal susceptibility profiles. In patients 13 Chandrasekar PH, Gatny CM. The effect of fluconazole with primary transplant-related complication such as acute prophylaxis on fungal colonization in neutropenic cancer GVHD, HUS/TTP, neutropenia, and severe hemorrhagic patients. Bone Marrow Transplantation Team. J Antimicrob cystitis, short-term survival was not significantly altered. Chemother 1994; 33: 309–318. However, survival was markedly compromised in the set- 14 Marr KA, White TC, van Burik JA et al. Development of fluconazole resistance in Candida albicans causing dissemi- ting of concurrent bacteremia and those with persistent nated infection in a patient undergoing marrow transplan- hematogenous candidiasis following high-risk allogeneic tation. Clin Infect Dis 1997; 25: 908–910. marrow transplantation. 15 Safdar A, Armstrong D. Infections in patients with neoplastic diseases. In: Shoemaker WC, Grenvik M, Ayers SM, Hol- brook PR (eds). Textbook of Critical Care, 4th edn. Saunders: Acknowledgements Philadelphia, 2000, pp 715–726. 16 Bodey GP, Anaissie EJ, Edwards JE. Definitions of Candida The results of this study were presented in part at the Eleventh infections. In: Bodey GP (ed.). Candidiasis: Pathogenesis, International Symposium on Infections in the Immunocompro- Diagnosis, and Treatment. Raven Press: New York, 1993, mised Host of The International Immunocompromised Host pp 407–408. Society (ICHS), Halifax, Nova Scotia, June 2000. 17 Nguyen MH, Peacock JE Jr, Morris AJ et al. The changing face of candidemia: emergence of non-Candida albicans species and antifungal resistance. Am J Med 1996; 100: 617–623. 18 Abi-Said D, Anaissie E, Uzun O et al. The epidemiology of References hematogenous candidiasis caused by different Candida species. Clin Infect Dis 1997; 24: 1122–1128. 1 Louria DB, Blevins A, Armstrong D et al. Fungemia caused 19 Abbas J, Bodey GP, Hanna HA et al. Candida krusei funge- by ‘nonpathogenic’ yeasts. Arch Intern Med 1967; 119: mia: an escalating serious infection in immunocompromised 247–252. patients. Arch Intern Med 2000; 160: 2659–2664. 2 Meunier-Carpentier F, Kiehn TE, Armstrong D. Fungemia in 20 Henslee-Downey JP, Abhyankar SH, Parrish RS et al. Use of the immunocompromised host. Changing patterns, antigene- partially mismatched related donors extends access to allo- mia, high mortality. Am J Med 1981; 71: 363–370. geneic marrow transplant. Blood 1997; 89: 3864–3872. 3 Viscoli C, Girmenia C, Marinus A et al. Candidemia in cancer 21 Wey SB, Mori M, Pfaller MA et al. Hospital-acquired candid- patients: a prospective, multicenter surveillance study by the emia. The attributable mortality and excess length of stay. Invasive Fungal Infection Group (IFIG) of the European Arch Intern Med 1988; 148: 2642–2645. Organization for Research and Treatment of Cancer (EORTC). 22 Pittet D, Wenzel RP. Nosocomial bloodstream infections. Clin Infect Dis 1999; 28: 1071–1079. Secular trends in rates, mortality, and contribution to total hos- 4 Beck-Sague C, Jarvis WR. Secular trends in the epidemiology pital deaths. Arch Intern Med 1995; 155: 1177–1184. of nosocomial fungal infections in the United States, 1980– 23 Wenzel RP. Nosocomial candidemia: risk factors and attribu- 1990. National Nosocomial Infections Surveillance System. J table mortality. Clin Infect Dis 1995; 20: 1531–1534. Infect Dis 1993; 167: 1247–1251. 24 Pittet D, Li N, Woolson RF et al. Microbiological factors 5 Verfaillie C, Weisdorf D, Haake R et al. Candida infections influencing the outcome of nosocomial bloodstream infec- in bone marrow transplant recipients. Bone Marrow Trans- tions: a 6-year validated, population-based model. Clin Infect plant 1991; 8: 177–184. Dis 1997; 24: 1068–1078. 6 Uzun O, Ascioglu S, Anaissie EJ, Rex JH. Risk factors and 25 Vu Van H, Piens MA, Archimbaud E et al. Serum levels of predictors of outcome in patients with cancer and break- ketoconazole in bone marrow transplanted patients. Nouv Rev through candidemia. Clin Infect Dis 2001; 32: 1713–1717. Franc Hematol 1983; 25: 241–244. 7 Alangaden G, Chandrasekar PH, Bailey E et al. Antifungal 26 Winston DJ, Chandrasekar PH, Lazarus HM et al. Fluconazol prophylaxis with low-dose fluconazole during bone marrow prophylaxis of fungal infection in patients with acute leuke- transplantation. The Bone Marrow Transplantation Team. mia: results of randomized placebo-controlled, double-blind, Bone Marrow Transplant 1994; 14: 919–924. multicenter trial. Ann Intern Med 1993; 118: 495–503. 8 Slavin MA, Osborne B, Adams R et al.Efficacy and safety 27 Goodman JL, Winston DJ, Greenfield RA et al. A controlled of fluconazole prophylaxis for fungal infections after marrow trial of fluconazole to prevent fungal infections in patients transplantation – a prospective, randomized, double-blind undergoing bone marrow transplantation. New Engl J Med study. J Infect Dis 1995; 171: 1545–1552. 1992; 326: 845–851. 9 Marr KA, Seidel K, Slavin MA et al. Prolonged fluconazole 28 Wingard JR. Importance of Candida species other than C. prophylaxis is associated with persistent protection against albicans as pathogens in oncology patients. Clin Infect Dis candidiasis-related death in allogeneic marrow transplant 1995; 20: 115–125. recipients: long-term follow-up of a randomized, placebo- 29 Viscoli C, Castagnola E. Emerging fungal pathogens, drug controlled trial. Blood 2000; 96: 2055–2061. resistance and the role of lipid formulations of amphotericin 10 Wingard JR, Merz WG, Rinaldi MG et al. Increase in Candida B in the treatment of fungal infections in cancer patients: a krusei infection among patients with bone marrow transplan- review. Int J Infect Dis 1998; 3: 109–118. tation and neutropenia treated prophylactically with flucona- 30 Safdar A, Chaturvedi V, Cross EW et al. Prospective study of zole. New Engl J Med 1991; 325: 1274–1277. Candida species in patients at a comprehensive cancer center. 11 Wingard JR, Merz WG, Rinaldi MG et al. Association of Antimicrob Agents Chemother 2001; 45: 2129–2133. Torulopsis glabrata infections with fluconazole prophylaxis in 31 Wingard JR. Importance of Candida species other than C. neutropenic bone marrow transplant patients. Antimicrob albicans as pathogens in oncology patients. Clin Infect Dis Agents Chemother 1993; 37: 1847–1849. 1995; 20: 115–125. 12 Hoppe JE, Klausner M, Klingebiel T et al. Retrospective 32 Alexander BD, Perfect JR. Antifungal resistance trends analysis of yeast colonization and infections in paediatric bone towards the year 2000. Implications for therapy and new marrow transplant recipients. Mycosis 1997; 40:47–54. approaches. Drugs 1997; 54: 657–678.

Bone Marrow Transplantation Candidemia in BMT recipients on ﬂuconazole prophylaxis A Safdar et al 878 33 Safdar A, Perline D, Armstrong D. Impact of Candida species of Candida parapsilosis fungemia in patients with hemato- colonization on mortality in patients with cancer. Clin logic malignancies: clinical aspects, predisposing factors, and Microbiol Infect 2001; 7 (Suppl.1): 336 (Abstr. 1555). differential pathogenicity of the causitive strains. Clin Infect 34 Safdar A, Perlin DS, Armstrong D. Emergence of Candida Dis 1996; 23: 506–514. parapsilosis fungemia as the dominant non-albicans Candida 37 Lopez-Jimenez J, Duarte-Palomino R, Cabezudo E et al. Can- species bloodstream infection at a cancer center in New York. dida parapsiolosis fungemia in bone marrow transplantation Clin Microbiol Infect 2001; 7 (Suppl. 1): 336 (Abstr. 1554). recipients: implications for azole prophylactic therapy. Blood 35 Krcmery V Jr, Sejnova D, Pichnova E. Breakthrough Candida 1997; 89: 3491–3492. tropicalis fungemia during ketoconazole prophylaxis in cancer 38 Goodman M, Pottage JC, Weaver DC. Candida krusei funge- patients. Acta Oncologica 1999; 38: 663–665. mia: report of four cases and review of the literature. Medicine 36 Girmenia C, Martino P, De Bernardis F et al. Rising incidence 1993; 72: 143–150.

Bone Marrow Transplantation