Oral and Intestinal Candida Colonization in Patients Undergoing Hematopoietic Stem-Cell Transplantation

Transcription

1 BRIEF REPORT Oral and Intestinal Candida Colonization in Patients Undergoing Hematopoietic Stem-Cell Transplantation Ines Zollner-Schwetz, 1 Holger W. Auner, 2,a Astrid Paulitsch, 3 Walter Buzina, 3 Philipp B. Staber, 2 Petra Ofner-Kopeinig, 4 Emil C. Reisinger, 5 Horst Olschewski, 1 and Robert Krause 1 1 Division of Infectious Diseases, Department of Pulmonology, 2 Department of Hematology, 3 Institute of Hygiene, Microbiology and Environmental Medicine, and 4 Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz, Austria; 5 Division of Infectious Diseases and Tropical Medicine, Department of Internal Medicine, University of Rostock, Rostock, Germany We investigated (1) the prevalence and quantity of, as well as risk factors for, orointestinal Candida colonization in patients undergoing hematopoietic stem-cell transplantation (HSCT) and (2) the genetic relatedness of colonizing C. albicans strains. Mouth-wash and stool samples were collected from 77 patients before they underwent HSCT and on days 1, 8, and 15 and were quantitatively cultured. C. albicans isolates were genotyped by microsatellite-marker analysis. The prevalence and quantity of orointestinal Candida colonization varied over time. In 48% (13/27) of multicolonized patients, the same Candida genotype was present in oral and intestinal samples. Oral colonization and decontamination of the gut by vancomycin and paromomycin were risk factors for intestinal Candida colonization. Invasive fungal infections are serious complications in patients undergoing treatment for cancer. In patients undergoing hematopoietic stem-cell transplantation (HSCT), Candida albicans is Received 10 November 2007; accepted 25 January 2008; electronically published 19 May Potential conflicts of interest: none reported. Presented in part: 46th Interscience Conference on Antimicrobial Agents and Chemotherapy 2006, San Francisco, California, September 2006 (poster M-909). a Current affiliation: Gene Regulation and Chromatin Group, Medical Research Council Clinical Sciences Centre, Imperial College, Hammersmith Campus, Du Cane Road, London W12 ONN, United Kingdom. Reprints or correspondence: Dr. Robert Krause, Div. of Infectious Diseases, Dept. of Pulmonology, Medical University of Graz, Auenbruggerplatz 20, A-8036 Graz, Austria (robert.krause@meduni-graz.at). The Journal of Infectious Diseases 2008; 198: by the Infectious Diseases Society of America. All rights reserved /2008/ $15.00 DOI: / responsible for most mucosal and disseminated Candida infections [1]. Although antifungal prophylaxis with fluconazole has been shown to decrease invasive Candida infections in patients undergoing HSCT [2], these infections remain a major cause of morbidity and mortality in patients with neutropenia [3]. Most invasive Candida infections are considered to originate from endogenous microflora in the host. A comprehensive review of the literature suggested that the gut is the primary source of candidemia [4]. Multicolonization with Candida isolates from the oropharynx and on rectal swabs has been reported to be associated with a significantly higher incidence of invasive candidiasis [5]; however, the study reporting this finding did not investigate the genetic relatedness between orointestinal isolates and invasive Candida isolates. Marr et al. [1] have demonstrated that oral Candida colonization contributes to a 3-fold increase in the risk of development of candidemia in patients undergoing allogeneic HSCT. Concomitant intestinal Candida colonization was not investigated [1]. The results of that study do not clarify whether oral colonization itself was the risk factor for invasive Candida infections in patients undergoing HSCT or whether oral colonization was associated with intestinal colonization by the same Candida strain that subsequently led to invasive candidiasis. The burden of Candida colonization in terms of the number of colony-forming units in each milliliter of saliva ram of feces has also been considered a risk factor for invasive candidiasis [3]. Because oral and intestinal colonization precede invasive candidiasis [1, 4], identification of risk factors for oral and intestinal colonization would be of great interest for the selection of patients at risk for colonization and subsequent invasive infection, as well as for the implementation of preventive strategies. The present study therefore aimed to investigate the prevalence and quantity of, as well as the risk factors for, orointestinal Candida colonization in patients undergoing HSCT. In addition, the genetic relatedness of colonizing C. albicans strains was investigated. Patients and methods. After they had provided informed consent, 77 patients undergoing autologous or allogeneic HSCT as well as nonmyeloablative stem-cell transplantation were enrolled in the present study. Demographic, clinical, and laboratory data were extracted from charts and computerized databases; 34 patients were female, and 43 were male, and their mean SD age was years (range, years). All patients received anti-infectious prophylaxis consisting of trimethoprimsulfamethoxazole (160 mg of trimethoprim, 800 mg of sulfamethoxazole) by mouth twice daily 2 8 days before HSCT and, 150 JID 2008:198 (1 July) BRIEF REPORT

2 from conditioning until discharge, received amphotericin B (200 mg) by mouth 5 times daily. From 8 days before HSCT until neutrophil recovery, 16 of the 77 patients also were treated for decontamination of the gut, by administration of vancomycin (500 mg) twice daily by mouth and of paromomycin (500 mg) thrice daily by mouth, at the clinician s discretion. In addition, the patients undergoing allogeneic HSCT (21 patients) and 1 patient undergoing autologous transplantation (pre-hsct treatment including chemotherapy and/or radiation) received systemic antifungal prophylaxis from the beginning of conditioning until engraftment, defined as an absolute neutrophil count of 1000/ L: 18 patients received fluconazole, 2 received caspofungin, 1 received voriconazole, and 1 received intraconazole. Thirty-one patients received empirical antifungal therapy (liposomal amphotericin B, voriconazole, or caspofungin) after 3 days of either ineffective first-line antibiotic therapy or, when patients had received prophylaxis, after 3 days of ineffective second-line antibiotic therapy. Mouth-wash samples for Candida cultures were collected, by use of 10 ml of sterile saline as described elsewhere [1], at baseline (1 7 days before HSCT), 1 day after HSCT and 8 and 15 days after HSCT. Stool samples were collected at the same time points. Quantitation of Candida cultures from mouth-wash and stool samples was performed as described elsewhere, by use of Candida CHROMagar [1, 6]. Genotyping of C. albicans isolates was performed as described elsewhere, by use of primers for the cell division cycle protein gene, CDC3, and the imidazole glycerol phosphate deshydratase gene, His3 [7]. The numerical index of discriminatory power (DP) of the analysis, based on the test probability that 2 unrelated Candida isolates sampled from the test population would be found to be in different typing groups, was calculated for the combination of both microsatellite markers, on the basis of a previously published formula, with a desired DP of 0.9 [7]. Amphotericin B, fluconazole, voriconazole, posaconazole, and caspofungin were tested against C. albicans and C. glabrata strains, by use of E-test strips used according to the manufacturer s instructions (AB Biotest). The association between patients characteristics and treatment and the prevalence of orointestinal Candida colonization was analyzed by 2 tests. The following variables were examined as potential risk factors: age, sex, conditioning regimen, irradiation, transplantation type, antifungal prophylaxis, oral decontamination of the gut, and oral Candida colonization (the latter only for intestinal Candida colonization). The variables that were significantly associated with orointestinal Candida colonization were included in a binary logistic-regression model. Friedman s test was used to compare Candida colony counts on the specified investigation days. Statistical analysis was performed by SPSS (version 14.0); for all results, P.05 was considered to be significant. Results. Of the 77 patients, 16 were treated for acute leukemia, 39 were treated for multiple myeloma, and 22 were treated for either non-hodgkin or Hodgkin disease; 56 patients underwent autologous HSCT, 5 underwent allogeneic HSCT, and 16 underwent nonmyeloablative stem-cell transplantation. A total of 8 patients received total-body irradiation, 28 received fractional irradiation and chemotherapy, and 41 received chemotherapy only. Risk factors for orointestinal colonization with Candida at baseline are summarized in table 1. For oral Candida colonization, no risk factor could be detected. The risk factors associated with intestinal colonization, both at baseline and at days 1, 8, and 15, were decontamination of the gut by vancomycin and paromomycin (P.05 for all days) and oral Candida colonization (P.05 at baseline and at days 8 and 15). Antifungal prophylaxis had no effect on Candida-colonization rates or quantitative cultures at baseline or at days 1 and 8. At day 15, antifungal prophylaxis was associated with decreased oral and intestinal colonization rates (P.01) and with decreased quantities of intestinal Candida (P.01). There was no case of proven invasive candidiasis in the present study. The prevalence, quantity, and time course of orointestinal colonization with Candida are presented in table 2. A total of 51 (66%) of the 77 patients were colonized at least once with C. albicans, in either the oral cavity or the intestinal tract. Of these 51 patients, 12 (24%) also were colonized with at least 1 nonalbicans species. A total of 11 (14%) of the 77 patients were colonized only with non-albicans species. Overall, C. glabrata was the most frequently isolated non-albicans species (30 isolates), followed by C. tropicalis (5 isolates) and C. krusei (2 isolates). The remaining 15 (19%) of the 77 patients were not colonized with Candida at any time point. A total of 174 C. albicans isolates cultured from 50 patients were genotyped, and 65 different microsatellite profiles were identified. The discriminatory power of our analysis was Of these 50 patients, 27 (54%) were multicolonized with C. albicans, defined as the isolation of C. albicans from at least 1 mouthwash and at least 1 stool sample [5]. In 13 (48%) of the 27 multicolonized patients, the same C. albicans genotype was cultured from oral and intestinal samples, whereas 5 (19%) of the 27 multicolonized patients had different genotypes. In 9 (33%) of the 27 multicolonized patients, identical C. albicans genotypes in combination with nonidentical C. albicans genotypes were cultured from oral and intestinal samples (table A1, which is available only in the electronic edition of the Journal). Susceptibility tests were performed with 177 C. albicans and 38 C. glabrata strains: for C. albicans, the MIC 90 values for fluconazole, voriconazole, posaconazole, caspofungin, and amphotericin B were 1.5 mg/l, mg/l, mg/l, mg/l, and 0.19 mg/l, respectively; for C. glabrata, the MIC 90 values for fluconazole, voriconazole, posaconazole, caspofun- BRIEF REPORT JID 2008:198 (1 July) 151

3 Table 1. Analysis of risk factors for Candida species colonization at baseline. Oral colonization Intestinal colonization Colonized/not Colonized/not Risk factor colonized, no. RR (95% CI) colonized, no. RR (95% CI) Age ( ) ( ) 49 years 38/15 21/24 49 years 9/10 4/15 Sex ( ) ( ) Female 18/13 7/19 Male 29/12 18/20 Transplantation type Autologous 36/15 Reference 19/25 Reference NST 8/ ( ) 5/ ( ) Allogeneic 3/ ( ) 1/ ( ) Radiation status ( ) ( ) Irradiated 19/7 12/14 Not irradiated 28/18 13/25 Antifungal prophylaxis ( ) ( ) Present 12/10 6/14 Absent 35/15 19/25 Contamination status ( ) a ( ) Decontaminated 13/3 10/6 Not decontaminated 34/22 15/33 Oral colonization b ( ) Present /2 Absent /19 NOTE. CI, confidence interval; NST, nonmyeloablative stem-cell transplantation; Reference, autologous stem-cell transplantation (which was used as the reference to compare the Candida-colonization risk of different types of transplantation); RR, relative risk. a P.031 in univariate analysis. b P.002 in univariate analysis; P.007 (95% CI, ) in multivariate analysis. gin, and amphotericin B were 64 mg/l, 4 mg/l, 1 mg/l, 1.5 mg/l and 0.38 mg/l, respectively. Discussion. In the present study, we investigated the prevalence and quantity of orointestinal Candida colonization in patients undergoing HSCT. By analyzing prospectively collected mouth-wash and stool samples, we demonstrated that the rate of Table 2. Description of time course of Candida colonization. oral and intestinal colonization varied over time, in the study population overall and in individual patients. For example, several patients had positive cultures at baseline and negative cultures at days 1 and 8, followed by positive cultures, with genetically identical Candida strains, at day 15. Similarly, in individual patients the quantity of Candida colonization was found to fluc- Colonization type, parameter Baseline Day 1 Day 8 Day 15 Oral Prevalence, no. a (%) 46/72 (63.9) 31/45 (68.9) b 25/63 (39.7) b 26/54 (48.1) b Median (range) of quantity, 120 ( ) 75 ( ) 25 ( ) 505 ( ) c Intestinal Prevalence, no. a (%) 25/64 (39.1) 15/48 (31.2) d 15/57 (26.3) 16/52 (30.8) Median (range) of quantity, ( ) ( ) ( ) ( ) Oral plus intestinal Same C. albicans genotype 12/50 6/50 6/50 6/50 a The denominators in the fractions are 77 because no samples were obtained from some of the 77 patients. b P.001 vs. baseline. c P.05 vs. day 1; P.01 vs. day 8. d P.05 vs. baseline C. albicans isolates from 50 patients were genotyped. 152 JID 2008:198 (1 July) BRIEF REPORT

4 tuate widely over time (table A1, which is available only in the electronic edition of the Journal). Comparable results were obtained in a recent study in which oral Candida colonization was also unstable [8]. Overall, orointestinal Candida colonization appears to be a dynamic process. Candida colonization of the gastrointestinal tract is considered to be a major risk factor for invasive candidiasis in patients with neutropenia [4]. It has been speculated that oral colonization indicates intestinal colonization with the same Candida strain [1, 9]. Although some studies have investigated the genetic relatedness of colonizing and infecting Candida strains [4, 10], the genetic relatedness of Candida strains colonizing the oral cavity and the intestinal tract has not been investigated in patients undergoing HSCT. In the present study, only 13 (48%) of the 27 patients who were multicolonized with C. albicans had the same C. albicans genotype in oral and intestinal samples. Therefore, in clinical practice it is not advisable to consider oral Candida colonization as being a marker for intestinal Candida colonization. In the present study, oral Candida colonization was a risk factor for intestinal colonization; however, oral colonization apparently indicates the general susceptibility to intestinal Candida colonization, rather than to colonization of the gut by the same Candida genotype. In the present study, we demonstrated that oral Candida colonization and decontamination of the gut by vancomycin and paromomycin are risk factors for intestinal Candida colonization. Normal intestinal bacterial microflora is important for maintenance of a balanced intestinal microecology with low Candida counts; because of its disruption of the indigenous intestinal flora, antimicrobial therapy (e.g., metronidazole and ciprofloxacin) leads to increased Candida positivity in stool cultures and to increased intestinal Candida counts [6, 11]. In accordance with these findings, the present study found that oral treatment with vancomycin and paromomycin was associated with increased Candida colonization. Because intestinal Candida colonization is considered to be a major risk factor for invasive candidiasis in patients with neutropenia, oral decontamination of the gut by these agents therefore should be avoided. In our patients, C. glabrata accounted for 30 colonizing isolates (MIC 90 for fluconazole, 64 mg/l). Marr et al. have demonstrated that fluconazole prophylaxis selects for fluconazoleresistant Candida species, possibly leading to fungemia acquired endogenously via the gastrointestinal tract [1]. We did not detect invasive candidiasis due to C. glabrata. In addition, we found no fluconazole-resistant C. albicans strains. We observed no case of proven invasive candidiasis. Similarly, a recent study found a very low (0.9%) incidence of invasive candidiasis in a large population of patients undergoing HSCT [12]. In summary, the findings of the present study suggest that oral Candida colonization cannot be thought of as being a marker for intestinal colonization by Candida strains. Acknowledgments We thank Christina Strempfl and Waltraut Eberharth for their expert technical assistance. References 1. Marr KA, Seidel K, White TC, Bowden RA. Candidemia in allogeneic blood and marrow transplant recipients: evolution of risk factors after the adoption of prophylactic fluconazole. J Infect Dis 2000; 181: Goodman JL, Winston DJ, Greenfield RA, et al. A controlled trial of fluconazole to prevent fungal infections in patients undergoing bone marrow transplantation. N Engl J Med 1992; 326: Blijlevens NM, Donnelly JP, de Pauw BE. Impaired gut function as risk factor for invasive candidiasis in neutropenic patients. Br J Haematol 2002; 117: Nucci M, Anaissie E. Revisiting the source of candidemia: skin ut? Clin Infect Dis 2001; 33: Martino P, Girmenia C, Micozzi A, De Bernardis F, Boccanera M, Cassone A. Prospective study of Candida colonization, use of empiric amphotericin B and development of invasive mycosis in neutropenic patients. Eur J Clin Microbiol Infect Dis 1994; 13: Krause R, Schwab E, Bachhiesl D, et al. Role of Candida in antibioticassociated diarrhea. J Infect Dis 2001; 184: Botterel F, Desterke C, Costa C, Bretagne S. Analysis of microsatellite markers of Candida albicans used for rapid typing. J Clin Microbiol 2001; 39: Tollemar J, Gross N, Dolgiras N, Jarstrand C, Ringden O, Hammarstrom L. Fungal prophylaxis by reduction of fungal colonization by oral administration of bovine anti-candida antibodies in bone marrow transplant recipients. Bone Marrow Transplant 1999; 23: Wingard JR. Importance of Candida species other than C. albicans as pathogens in oncology patients. Clin Infect Dis 1995; 20: Reagan DR, Pfaller MA, Hollis RJ, Wenzel RP. Characterization of the sequence of colonization and nosocomial candidemia using DNA fingerprinting and a DNA probe. J Clin Microbiol 1990; 28: Trenschel R, Peceny R, Runde V, et al. Fungal colonization and invasive fungal infections following allogeneic BMT using metronidazole, ciprofloxacin and fluconazole or ciprofloxacin and fluconazole as intestinal decontamination. Bone Marrow Transplant 2000; 26: Pagano L, Caira M, Nosari A, et al. Fungal infections in recipients of hematopoietic stem cell transplants: results of the SEIFEM B-2004 study Sorveglianza Epidemiologica Infezioni Fungine Nelle Emopatie Maligne. Clin Infect Dis 2007; 45: BRIEF REPORT JID 2008:198 (1 July) 153

5 Table A1. Quantity and microsatellite genotypes of 174 oral and intestinal Candida albicans isolates from 50 patients. Baseline a 1 day after HSCT 8 days after HSCT 15 days after HSCT Patient Colonization type 1 Oral 2 GT GT 1 10 GT 2 Neg Intestinal Neg Neg Neg Neg 2 Oral 10 3 GT 3 10 GT 20 Neg Neg Intestinal NS Neg Neg Neg 3 Oral 25 GT GT 7 10 GT 6 4 GT 5 Intestinal NS Neg Neg Neg 4 Oral 10 6 GT GT GT 4 NS Intestinal 10 4 GT GT GT47 NS 5 Oral 10 6 GT GT GT 8 NS Intestinal NS 10 7 GT GT35 NS 6 Oral 10 3 GT 9 NS NS NS Intestinal GT 10 NS NS NS 7 Oral 10 2 ; 10 GT 10, GT 13 7 GT 37 Neg 10 6 GT37 Intestinal Neg GT 22 Neg NS 8 Oral 7 GT GT 10 Neg NS Intestinal Neg 10 4 GT 10 Neg NS 9 Oral 1 GT 10 1 GT 45 Neg NS Intestinal GT 45 Neg Neg NS 10 Oral 10 GT 10 Neg NS Neg Intestinal Neg Neg NS Neg 11 Oral GT GT GT GT55 Intestinal 10 8 GT GT 55 NS GT11 12 Oral 40 GT 15 NS 25 GT GT15 13 Oral Neg 40 GT 45 Neg NS Intestinal Neg Neg Neg NS 14 Oral GT GT 17 NS Neg Intestinal GT 61 Neg NS Neg 15 Oral GT 63 NS NS GT51 Intestinal GT 12 Neg NS GT18 16 Oral 80 GT 19 Neg Neg Neg Intestinal 10 6 GT GT 19 Neg Neg 17 Oral GT GT GT57 Neg Intestinal Neg Neg Neg Neg 18 Oral Neg NS Neg Neg Intestinal Neg NS GT23 NS 19 Oral 10 3 GT 41 NS 80 GT41 NS Intestinal 10 GT 24 NS Neg NS 20 Oral 1 GT GT 24 Neg Neg Intestinal Neg 10 2 Neg Neg 21 Oral 6 GT 25 NS NS 20 GT25 Intestinal NS Neg NS Neg 22 Oral 4 GT 25 Neg Neg 10 6 GT25 Intestinal Neg Neg Neg GT25 23 Oral 2 GT 29 10; GT 29, GT GT GT29 Intestinal Neg GT GT GT29 24 Oral GT 34 Neg Neg 10 7 GT33 Intestinal GT 33 Neg NS GT28 25 Oral GT 30 5 GT 30 Neg NS Intestinal NS Neg Neg NS (continued)

6 Table A1. (Continued.) Baseline a 1 day after HSCT 8 days after HSCT 15 days after HSCT Patient Colonization type 26 Oral NS 2 GT 31 2 GT GT31 Intestinal NS Neg GT GT31 27 Oral 14 GT 32 NS 12 GT GT32 Intestinal GT 32 NS 10 2 GT32 NS 28 Oral 10 6 GT GT GT34 NS Intestinal GT GT 34 Neg NS 29 Oral Neg Neg 1 GT36 NS Intestinal GT 36 Neg Neg NS 30 Oral GT GT GT37 NS Intestinal Neg Neg Neg NS 31 Oral NS Neg 10 GT GT37 Intestinal NS Neg Neg 10 5 GT37 32 Oral 10 GT 39 2 GT 37 Neg NS Intestinal Neg Neg NS NS 33 Oral 18 GT 38 Neg NS Neg Intestinal GT 38 NS NS Neg 34 Oral 15 GT GT 40 NS Neg Intestinal Neg Neg NS Neg 35 Oral GT 41 NS Neg NS 36 Oral GT 42 NS Neg Neg Intestinal Neg NS NS Neg 37 Oral 50 GT 49 NS NS 10 4 GT54 Intestinal Neg NS NS Neg 38 Oral GT 45 NS 1 GT45 1 NI 39 Oral GT 45 NS Neg NS Intestinal Neg NS Neg NS 40 Oral GT 45 1 GT 48 NS Neg Intestinal Neg NS NS Neg 41 Oral GT 45 Neg 72 GT GT45 Intestinal GT GT GT45 Neg 42 Oral 10 3 GT 58 NS 10 6 GT GT46 Intestinal GT 58 NS 10 2 GT58 NS 43 Oral GT 52 NS Neg 10 4 GT52 Intestinal GT 52 Neg Neg GT52 44 Oral GT GT 52 1 GT52 10 GT52 Intestinal Neg Neg Neg Neg 45 Oral GT GT 53 NS 10 5 GT53 Intestinal Neg Neg NS NS 46 Oral 80 GT GT 57 Neg NS Intestinal GT GT GT57 NS 47 Oral 10 4 GT GT GT59 50 GT59 Intestinal 10 5 GT 59 Neg Neg 10 2 GT60 48 Oral 80 GT 60 NS 30 GT60 NS Intestinal Neg NS Neg NS 49 Oral GT 64 NS 7 GT GT64 Intestinal GT 64 NS 2, GT GT64 50 Oral Neg NS 50 GT65 Neg NOTE. HSCT, hematopoietic stem-cell transplantation; Neg, negative culture/no growth; NI, not investigated; NS, no sample obtained. a 1 7 days before HSCT. b Representing consecutively numbered microsatellite genotypes; for example, GT (base pairs; CDC3 marker allele 1 and 2 and HIS 3 maker allele 1 and 2), GT , etc.