- Email: [email protected]
Distribution of fluconazole-resistant Candida bloodstream isolates among hospitals and inpatient services in Israel
ORIGINAL ARTICLE
10.1111/1469-0691.12004
Distribution of fluconazole-resistant Candida bloodstream isolates among hospitals and inpatient services in Israel R. Ben-Ami1, G. Rahav2, H. Elinav3, I. Kassis4, I. Shalit5, T. Gottesman6, O. Megged7, M. Weinberger8, P. Ciobotaro9, P. Shitrit10, G. Weber11, A. Paz12, D. Miron13, I. Oren4, J. Bishara5, C. Block3, N. Keller2, D. P. Kontoyiannis14 and M. Giladi1 for the Israeli Candidaemia Study Group* 1) Tel Aviv Sourasky Medical Centre, Tel Aviv, 2) Chaim Sheba Medical Centre, Tel Hashomer, 3) Hadassah Hebrew-University Hospital, Jerusalem, 4) Rambam Medical Centre, Haifa, 5) Rabin Medical Centre and Schneider Children’s Hospital, Petah Tikva, 6) Wolfson Medical Centre, Holon, 7) Shaare Zedek Medical Centre, Jerusalem, 8) Assaf Harofeh Medical Centre, Zerifin, 9) Kaplan Medical Centre, Rehovot, 10) Meir Medical Centre, Kfar Saba, 11) Carmel Medical Centre, Haifa, 12) Bnai Zion Medical Centre, Haifa, 13) Emek Medical Centre, Afula and Ziv Medical Centre, Zefat, Israel and 14) MD Anderson Cancer Centre, Houston, TX, USA
Abstract The emergence of fluconazole-resistant Candida (FRC) is worrisome, but little is known about susceptibility patterns in different nosocomial settings. We prospectively analysed Candida bloodstream isolates in 18 medical centres in Israel (six tertiary-care and 12 community hospitals). The study included 444 episodes of candidaemia (450 patient-specific isolates, 8.5% fluconazole-resistant). Institutional FRC bloodstream infection rates correlated with annual inpatient days, and were strongly associated with the presence and activity of haematology/oncology services. Infection with Candida krusei and fluconazole-resistant Candida glabrata occurred exclusively in hospitals with >600 beds. These findings suggest that empirical antifungal strategies should be tailored to the nosocomial setting. Keywords: Candidaemia, drug resistance, epidemiology, fluconazole Original Submission: 2 May 2012; Revised Submission: 8 August 2012; Accepted: 14 August 2012 Editor: E. Roilides Clin Microbiol Infect
Corresponding author: R. Ben-Ami, Infectious Disease Unit, Tel Aviv Sourasky Medical Centre, 6 Weizman St., Tel Aviv 64239, Israel E-mail: [email protected] *Listed in Appendix I.
Candidaemia is a serious nosocomial infection that carries significant morbidity and mortality [1,2]. Epidemiological surveys have revealed regional variations in candidaemia incidence rates, species distribution, drug-resistance patterns and mortality rates [3–14]. However, even within a defined geographical region, candidaemia may occur in widely disparate settings, from small community hospitals with basic inpatient care facilities to large tertiary centres that provide highly specialized services. Nevertheless, little is known about the characteristics of Candida spp. in different nosocomial settings. We performed a prospective nationwide hospital laboratory-based survey of candidaemia in Israel from November
2005 through to June 2007 [15]. Eighteen medical centres, which together account for 75% of the acute-care hospital beds in Israel, were included: six tertiary-care hospitals (6002 beds), and 12 smaller community hospitals (5678 beds). Hospitals were located in northern (n = 7), central (n = 9) and southern (n = 2) districts of Israel. Ten hospitals were clustered in three urban zones: zone A (greater Tel Aviv area, four hospitals), zone B (Jerusalem, three hospitals) and zone C (Haifa, three hospitals). Demographic and clinical data were prospectively entered into standardized data forms by on-site investigators and reviewed by the principle investigator. Candida isolates were sent to the central study site for species identification using standard microbiology methods. Susceptibility to fluconazole, voriconazole, amphotericin B and caspofungin was determined using the E-test (bioMerieux, Marcy l’Etoile, France) method. Because of trailing growth with the E-test, C. glabrata fluconazole susceptibility was determined using broth microdilution according to CLSI method M27-A3 [16]. We used
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases
2
Clinical Microbiology and Infection
CMI
the recently updated CLSI species-specific antifungal drug susceptibility thresholds [17–19]. The study was approved by the ethics committees of each participating centre. Candidaemia incidence rates were calculated using data obtained from the Israeli Ministry of Health (http:// www.health.gov.il). We analysed the correlation between fluconazole-resistant Candida (FRC) bloodstream infection and hospital volume of activity and the availability and activity of specialized services. Hospital and patient variables that were found to interact significantly with fluconazole-resistance in bivariable analyses were entered into a multivariable logistic regression model in a stepwise fashion, and those variables with a significant effect on the b coefficient
(p <0.05) were retained in the model. The time-at-risk (number of calendar days from the date of hospital admission to the date of candidaemia) was forced into the multivariable model. Chi-squared test or Fisher’s exact test and Student’s t test were used to assess the distribution of categorical and continuous variables, respectively. A type I error of <0.05 was considered statistically significant. Calculations were performed in STATA 11.1 (StataCorp, College Station, TX, USA). The study included 444 patients and 450 patient-specific Candida isolates (Table 1). The adjusted incidence rates of candidaemia were 11/100 000 inpatient-days, 5.57/100 000 population/year. The most frequent species were C. albicans, (n = 200, 44.4%), C. parapsilosis (n = 75, 16.6%), C. tropicalis
TABLE 1. Demographic and clinical characteristics of patients with candidaemia by hospital size Hospital size Characteristic Medical centres Tertiary-care centre Urban zone A B C Other Patient days/year, mean (SD) Intensive care Oncology HSCT Total Patients with candidaemia Gender Male Female Age, median (IQR), years Functional disabilitya Days in hospital, median (IQR)b Inpatient service Medicine Surgery Intensive care Haematology/oncology Charlson score, median (IQR) Malignancy Leukaemia Lymphoma Solid tumour Diabetes mellitus Renal failure Antimicrobial exposurec Azole exposurec Central vascular catheter Parenteral nutrition Mechanical ventilationc Surgeryc Abdominal surgery Chest surgery Cytotoxic chemotherapyc Neutropeniac Renal replacement therapy Prematurity Stem cell transplant Solid organ transplant
<600 beds
>600 beds
Odds ratio (95% CI)
p
0.009
9 (50) 0 (0)
9 (50) 6 (67)
NA
0 2 2 5
4 1 1 3
(44) (11) (11) (33)
NA 0.4 0.4 0.4
(3658) (7573) (2992) (82 755)
NA NA NA NA
2772 0 0 130 733
(0) (22) (22) (55) (2629) (0) (0) (48 302)
6017 11 147 2394 319 446
Total 18 (100) 6 (33) 4 3 3 8
0.20 0.04 0.002 0.058 <0.0001
94 (21.1)
356 (80.1)
49 44 70 39 14
(53) (47) (49–79) (51) (6–27)
189 162 62 138 16
(54) (46) (32–76) (47) (6–30)
NA 0.8 (0.4–1.4) NA
35 21 52 4 3
(37.6) (22.5) (55.9) (4.3) (1–6)
110 76 145 49 3
(31.3) (21.6) (41.3) (13.9) (2–6)
0.7 0.9 0.5 3.6 NA
0 5 16 28 10 83 11 66 32 52 39 23 5 6 3 9 10 0 1
(0) (5.3) (17.2) (30.1) (10.7) (89.2) (11.8) (70.9) (34.4) (55.9) (41.9) (24.7) (5.3) (6.4) (3.2) (9.6) (10.7) (0) (1.0)
34 18 97 80 50 327 48 265 115 147 136 64 21 76 51 30 19 22 3
(9.6) (5.1) (27.6) (22.7) (14.2) (93.1) (13.6) (75.5) (32.7) (41.8) (38.7) (18.2) (5.9) (21.6) (14.5) (8.5) (5.4) (6.2) (0.8)
2.5c 0.9 1.8 0.6 1.3 1.6 1.1 1.2 0.9 0.5 0.8 0.6 1.1 4.0 5.1 0.8 0.4 1.6d 0.7
79 100 19 4 051
(22) (17) (17) (44)
096 (3512) 323 (7738) 150 613 (117 249) 444 (100)
1.0 (0.6–1.6)
(0.4–1.2) (0.5–1.7) (0.3–0.9) (1.2–14.1)
(0.3–3.3) (1.0–3.5) (0.4–1.1) (0.6–3.1) (0.6–3.7) (0.5–2.6) (0.7–2.1) (0.5–1.5) (0.3–0.9) (0.5–1.4) (0.3–1.2) (0.3–3.9) (1.6–11.6) (1.5–26.0) (0.3–2.1) (0.2–1.1) (0.06–42.0)
0.9 0.9 <0.0001 0.6 0.3
238 206 65 177 15
(53.6) (46.4) (33–77) (48.1) (6–30)
0.2 0.8 0.01 0.01 0.8
145 74 197 53 3
(32.6) (16.6) (44.4) (11.9) (2–6)
0.001 1.0 0.04 0.1 0.5 0.2 0.7 0.4 0.8 0.02 0.6 0.1 1.0 0.0008 0.003 0.6 0.09 0.01 1.0
34 23 113 108 60 410 59 331 147 199 175 87 26 82 54 39 29 22 4
(7.6) (5.1) (25.4) (24.3) (13.5) (92.3) (13.2) (74.5) (33.1) (44.8) (39.4) (19.6) (5.8) (18.5) (12.2) (8.7) (6.5) (5.0) (0.9)
HSCT, haematopoietic stem-cell transplant; IQR, interquartile range; NA, not applicable. All values denote number of patients (percentage), except where specified otherwise. p values are for Fisher’s exact test for categorical variables and Student’s t test for scalar variables. Values in bold represent variables that were significantly differently distributed between small (<600 beds) and large (>600 beds) hospitals. a Defined as partial or complete dependence on others for activities of daily living. b Time (days) from date of hospital admission to date of first blood culture positive for Candida spp. c Within 30 days before the onset of candidaemia. d Odds ratio calculated with the Cornfield method. Confidence interval cannot be detrmined because there were no cases in patients from hospitals with <600 beds.
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI
Ben-Ami et al. Fluconazole-resistant Candida
CMI
(n = 77, 17.1%), C. glabrata (n = 69, 15.3%), and C. krusei (n = 14, 3.1%). Six patients had dual-species candidaemia. Fifty-four candidaemia episodes (12.1%) were caused by isolates non-susceptible to fluconazole: 16 (3.6%) susceptible dose-dependent and 38 (8.5%) FRC. Seven centres (38.8%) had oncology services and five (27.7%) had stem-cell transplantation services (Table 1). FRC were found in ten hospitals (rate of resistance in those institutions, 7.1–13.6%). Hospitals where FRC were detected had significantly more total acute beds than hospitals with no FRC—median (interquartile ranges), 752 (657–970) beds versus 405 (310–465) beds, respectively; p 0.009). FRC were detected in 2/9 hospitals (22%) with <600 beds but in 8/9
3
hospitals (88%) with >600 beds (Fig. 1). The pooled incidence of candidaemia was marginally greater in larger hospitals than in smaller ones (11.9 versus 7.3/100 000 patientdays, p 0.050). However, the incidence rate of FRC bloodstream infection was nearly ten-fold higher in larger hospitals (1.2 versus 0.14/100 000 patient-days, p 0.0004). Moreover, there was a significant linear correlation between hospital volume (annual inpatient-days) and the incidence rate of FRC bloodstream infection (Pearson’s coefficient, 0.73, p 0.0005). Fluconazole-resistant C. glabrata and C. krusei occurred exclusively in hospitals with >600 beds: 0.23 versus 0/100 000 patient-days for C. glabrata (p 0.02), and 0.44 versus 0/ 100 000 patient-days for C. krusei (p 0.009; Fig. 1). Resistance
(a)
(b)
(c)
(d)
FIG. 1. Incidence rates of candidaemia by hospital size. Each marker represents a single medical centre. (a) Incidence rates of all candidaemia episodes; (b) incidence rates of fluconazole-resistant Candida bloodstream infection; (c) incidence rates of Candida krusei bloodstream infection; (d) incidence rates of fluconazole-resistant Candida glabrata bloodstream infection. Vertical dashed line marks the median hospital size of 600 beds. ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI
4
Clinical Microbiology and Infection
rates to voriconazole, caspofungin and amphotericin B were not significantly different between low-volume and high-volume hospitals. Demographic and clinical features of patients with candidaemia varied with hospital size. Hospitalization at a hospital with >600 beds was associated with younger age, care in a haematology/oncology service, leukaemia, solid neoplasm, cytotoxic chemotherapy, neutropenia and haematopoietic stem-cell transplantation (Table 1). Three covariates were retained in the multivariable model with FRC bloodstream infection as the dependent variable: neutropenia (odds ratio 3.6, 95% confidence interval 1.6–8.2; p 0.002), receipt of fluconazole within 30 days before candidaemia (OR 4.1, 95% CI 1.8–9.2; p <0.0001) and hospitalization at a medical centre with >600 beds (OR 4.4, 95% CI 1.006– 19.5; p 0.049). Goodness-of-fit for this model was adequate as determined using the Hosmer–Lemeshow test (p 0.5). In this nationwide study in Israel, candidaemia was a frequent nosocomial event, but the proportion of FRC was highly variable across different medical centres, even among those in close geographical proximity. Strikingly, divergent epidemiological patterns were observed in low-volume and high-volume institutions. Hence, although the candidaemia rate was only slightly higher in high-volume institutions, bloodstream infection with FRC occurred almost exclusively at high-volume institutions. FRC infection was strongly associated with the oncological and stem-cell transplantation settings, which were limited to large, tertiary-care medical centres. Our study underscores the need for nationwide studies that incorporate data from diverse nosocomial settings and a range of hospital types [2,8,9,15]. Candidaemia surveillance programmes often only include data from tertiary-care medical centres [3,7,10,14], a design that may select for patient populations at high risk for drug-resistant Candida spp., resulting in potential overestimation of the overall incidence rate of drug resistance. In conclusion, this study, the first of its kind performed in the Middle East, illustrates the epidemiology of candidaemia in Israel and highlights the importance of tertiary-care medical centres, and specifically that of haematology/oncology units, as hotspots for the emergence of FRC. These findings may be used to direct empirical treatment strategies at the institutional level.
Acknowledgements The authors wish to thank Anna Novikov and Nathaniel Albert for excellent technical assistance. The principle investigator (MG) had full access to all of the data in the study
CMI
and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Funding This work was funded in part by a grant from Merck Sharp & Dohme Ltd, Israel and by the Israeli Society for Infectious Diseases.
Transparency Declaration RB-A has received consulting fees and research support from Pfizer; DPK has received research support from and honoraria from Merck, Fujisawa, Enzon, Pfizer, and Schering Plough. All other authors: no conflicts.
Appendix I: Additional Israeli Candidaemia Study Group Invesigators Tel Aviv Sourasky Medical Centre, Tel Aviv: Ruth Orni-Wasserlauf; Hadassah-Hebrew University Medical Centre, Jerusalem: Allon Moses, Karen Olshtain-Pops; Chaim Sheba Medical Centre, Tel Hashomer: Michal Krieger, Yasmin Maor, Anna Goldshmidt-Reuven; Rambam Medical Centre, Haifa: Renato Finkelstein, Hannah Sprecher; Rabin Medical Centre and Schneider Children’s Hospital, Petah Tikva: Michal Paul, Itzhak Levi, Zmira Samra, Elad Goldberg; Wolfson Medical Centre, Holon: Michael Dan, Orna Schwartz-Harari; Shaare Zedek Medical Centre, Jerusalem: Amos Yinon, Yonit Weiner Well; Assaf Harofeh Medical Centre, Zerifin: Tsilia Lazarovitch; Kaplan Medical Centre, Rehovot: Oren Zimhony; Meir Medical Centre, Kfar Saba: BatSheva Gottesman; Bnai Zion Medical Centre, Haifa: Israel Potasman; Emek Medical Centre, Afula: Raul Raz, Bibiana Chazan; Soroka Medical Centre, Beer Sheva: Klaris Riesenberg, Lisa Saidel; Bikur Holim Medical Centre, Jerusalem: Efraim Halperin; Hillel Yafe Medical Centre, Hadera: David Hassin; Poriya Medical Centre, Tiberias: Soboh Soboh.
References 1. Horn DL, Neofytos D, Anaissie EJ et al. Epidemiology and outcomes of candidemia in 2019 patients: data from the prospective antifungal therapy alliance registry. Clin Infect Dis 2009; 48: 1695–1703. 2. Zaoutis TE, Argon J, Chu J, Berlin JA, Walsh TJ, Feudtner C. The epidemiology and attributable outcomes of candidemia in adults and children hospitalized in the United States: a propensity analysis. Clin Infect Dis 2005; 41: 1232–1239.
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI
Ben-Ami et al. Fluconazole-resistant Candida
CMI
3. Colombo AL, Nucci M, Park BJ et al. Epidemiology of candidemia in Brazil: a nationwide sentinel surveillance of candidemia in eleven medical centers. J Clin Microbiol 2006; 44: 2816–2823. 4. Hajjeh RA, Sofair AN, Harrison LH et al. Incidence of bloodstream infections due to Candida species and in vitro susceptibilities of isolates collected from 1998 to 2000 in a population-based active surveillance program. J Clin Microbiol 2004; 42: 1519–1527. 5. St-Germain G, Laverdiere M, Pelletier R et al. Epidemiology and antifungal susceptibility of bloodstream Candida isolates in Quebec: report on 453 cases between 2003 and 2005. Can J Infect Dis Med Microbiol 2008; 19: 55–62. 6. Poikonen E, Lyytikainen O, Anttila VJ et al. Secular trend in candidemia and the use of fluconazole in Finland, 2004–2007. BMC Infect Dis 2010; 10: 312. 7. Poikonen E, Lyytikainen O, Anttila VJ, Ruutu P. Candidemia in Finland, 1995–1999. Emerg Infect Dis 2003; 9: 985–990. 8. Chen S, Slavin M, Nguyen Q et al. Active surveillance for candidemia, Australia. Emerg Infect Dis 2006; 12: 1508–1516. 9. Asmundsdottir LR, Erlendsdottir H, Gottfredsson M. Increasing incidence of candidemia: results from a 20-year nationwide study in Iceland. J Clin Microbiol 2002; 40: 3489–3492. 10. Pappas PG, Rex JH, Lee J et al. A prospective observational study of candidemia: epidemiology, therapy, and influences on mortality in hospitalized adult and pediatric patients. Clin Infect Dis 2003; 37: 634–643. 11. Diekema DJ, Messer SA, Brueggemann AB et al. Epidemiology of candidemia: 3-year results from the emerging infections and the epidemiology of Iowa organisms study. J Clin Microbiol 2002; 40: 1298–1302. 12. Marchetti O, Bille J, Fluckiger U et al. Epidemiology of candidemia in Swiss tertiary care hospitals: secular trends, 1991–2000. Clin Infect Dis 2004; 38: 311–320. 13. Pfaller MA, Moet GJ, Messer SA, Jones RN, Castanheira M. Geographic variations in species distribution and echinocandin and azole
14.
15.
16.
17.
18.
19.
5
antifungal resistance rates among Candida bloodstream infection isolates: report from the SENTRY Antimicrobial Surveillance Program (2008 to 2009). J Clin Microbiol 2011; 49: 396–399. Pfaller MA, Messer SA, Boyken L, Tendolkar S, Hollis RJ, Diekema DJ. Geographic variation in the susceptibilities of invasive isolates of Candida glabrata to seven systemically active antifungal agents: a global assessment from the ARTEMIS Antifungal Surveillance Program conducted in 2001 and 2002. J Clin Microbiol 2004; 42: 3142–3146. Ben-Ami R, Olshtain-Pops K, Krieger M et al. Antibiotic exposure as a risk factor for fluconazole-resistant Candida bloodstream infection. Antimicrob Agents Chemother 2012; 56: 2518–2523. Clinical Laboratory Standards I. Reference Method for Broth Dilution Antifungal Susceptibility testing of Yeasts. Approved Standard M27-A3. Wayne, PA: National Committee for Clinical Laboratory Standards, 2008. Pfaller MA, Andes D, Diekema DJ, Espinel-Ingroff A, Sheehan D. Wild-type MIC distributions, epidemiological cutoff values and species-specific clinical breakpoints for fluconazole and Candida: time for harmonization of CLSI and EUCAST broth microdilution methods. Drug Resist Updat 2010; 13: 180–195. Pfaller MA, Diekema DJ, Andes D et al. Clinical breakpoints for the echinocandins and Candida revisited: integration of molecular, clinical, and microbiological data to arrive at species-specific interpretive criteria. Drug Resist Updat 2011; 14: 164–176. Pfaller MA, Andes D, Arendrup MC et al. Clinical breakpoints for voriconazole and Candida spp. revisited: review of microbiologic, molecular, pharmacodynamic, and clinical data as they pertain to the development of species-specific interpretive criteria. Diagn Microbiol Infect Dis 2011; 70: 330–343.
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI
10.1111/1469-0691.12004
Distribution of fluconazole-resistant Candida bloodstream isolates among hospitals and inpatient services in Israel R. Ben-Ami1, G. Rahav2, H. Elinav3, I. Kassis4, I. Shalit5, T. Gottesman6, O. Megged7, M. Weinberger8, P. Ciobotaro9, P. Shitrit10, G. Weber11, A. Paz12, D. Miron13, I. Oren4, J. Bishara5, C. Block3, N. Keller2, D. P. Kontoyiannis14 and M. Giladi1 for the Israeli Candidaemia Study Group* 1) Tel Aviv Sourasky Medical Centre, Tel Aviv, 2) Chaim Sheba Medical Centre, Tel Hashomer, 3) Hadassah Hebrew-University Hospital, Jerusalem, 4) Rambam Medical Centre, Haifa, 5) Rabin Medical Centre and Schneider Children’s Hospital, Petah Tikva, 6) Wolfson Medical Centre, Holon, 7) Shaare Zedek Medical Centre, Jerusalem, 8) Assaf Harofeh Medical Centre, Zerifin, 9) Kaplan Medical Centre, Rehovot, 10) Meir Medical Centre, Kfar Saba, 11) Carmel Medical Centre, Haifa, 12) Bnai Zion Medical Centre, Haifa, 13) Emek Medical Centre, Afula and Ziv Medical Centre, Zefat, Israel and 14) MD Anderson Cancer Centre, Houston, TX, USA
Abstract The emergence of fluconazole-resistant Candida (FRC) is worrisome, but little is known about susceptibility patterns in different nosocomial settings. We prospectively analysed Candida bloodstream isolates in 18 medical centres in Israel (six tertiary-care and 12 community hospitals). The study included 444 episodes of candidaemia (450 patient-specific isolates, 8.5% fluconazole-resistant). Institutional FRC bloodstream infection rates correlated with annual inpatient days, and were strongly associated with the presence and activity of haematology/oncology services. Infection with Candida krusei and fluconazole-resistant Candida glabrata occurred exclusively in hospitals with >600 beds. These findings suggest that empirical antifungal strategies should be tailored to the nosocomial setting. Keywords: Candidaemia, drug resistance, epidemiology, fluconazole Original Submission: 2 May 2012; Revised Submission: 8 August 2012; Accepted: 14 August 2012 Editor: E. Roilides Clin Microbiol Infect
Corresponding author: R. Ben-Ami, Infectious Disease Unit, Tel Aviv Sourasky Medical Centre, 6 Weizman St., Tel Aviv 64239, Israel E-mail: [email protected] *Listed in Appendix I.
Candidaemia is a serious nosocomial infection that carries significant morbidity and mortality [1,2]. Epidemiological surveys have revealed regional variations in candidaemia incidence rates, species distribution, drug-resistance patterns and mortality rates [3–14]. However, even within a defined geographical region, candidaemia may occur in widely disparate settings, from small community hospitals with basic inpatient care facilities to large tertiary centres that provide highly specialized services. Nevertheless, little is known about the characteristics of Candida spp. in different nosocomial settings. We performed a prospective nationwide hospital laboratory-based survey of candidaemia in Israel from November
2005 through to June 2007 [15]. Eighteen medical centres, which together account for 75% of the acute-care hospital beds in Israel, were included: six tertiary-care hospitals (6002 beds), and 12 smaller community hospitals (5678 beds). Hospitals were located in northern (n = 7), central (n = 9) and southern (n = 2) districts of Israel. Ten hospitals were clustered in three urban zones: zone A (greater Tel Aviv area, four hospitals), zone B (Jerusalem, three hospitals) and zone C (Haifa, three hospitals). Demographic and clinical data were prospectively entered into standardized data forms by on-site investigators and reviewed by the principle investigator. Candida isolates were sent to the central study site for species identification using standard microbiology methods. Susceptibility to fluconazole, voriconazole, amphotericin B and caspofungin was determined using the E-test (bioMerieux, Marcy l’Etoile, France) method. Because of trailing growth with the E-test, C. glabrata fluconazole susceptibility was determined using broth microdilution according to CLSI method M27-A3 [16]. We used
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases
2
Clinical Microbiology and Infection
CMI
the recently updated CLSI species-specific antifungal drug susceptibility thresholds [17–19]. The study was approved by the ethics committees of each participating centre. Candidaemia incidence rates were calculated using data obtained from the Israeli Ministry of Health (http:// www.health.gov.il). We analysed the correlation between fluconazole-resistant Candida (FRC) bloodstream infection and hospital volume of activity and the availability and activity of specialized services. Hospital and patient variables that were found to interact significantly with fluconazole-resistance in bivariable analyses were entered into a multivariable logistic regression model in a stepwise fashion, and those variables with a significant effect on the b coefficient
(p <0.05) were retained in the model. The time-at-risk (number of calendar days from the date of hospital admission to the date of candidaemia) was forced into the multivariable model. Chi-squared test or Fisher’s exact test and Student’s t test were used to assess the distribution of categorical and continuous variables, respectively. A type I error of <0.05 was considered statistically significant. Calculations were performed in STATA 11.1 (StataCorp, College Station, TX, USA). The study included 444 patients and 450 patient-specific Candida isolates (Table 1). The adjusted incidence rates of candidaemia were 11/100 000 inpatient-days, 5.57/100 000 population/year. The most frequent species were C. albicans, (n = 200, 44.4%), C. parapsilosis (n = 75, 16.6%), C. tropicalis
TABLE 1. Demographic and clinical characteristics of patients with candidaemia by hospital size Hospital size Characteristic Medical centres Tertiary-care centre Urban zone A B C Other Patient days/year, mean (SD) Intensive care Oncology HSCT Total Patients with candidaemia Gender Male Female Age, median (IQR), years Functional disabilitya Days in hospital, median (IQR)b Inpatient service Medicine Surgery Intensive care Haematology/oncology Charlson score, median (IQR) Malignancy Leukaemia Lymphoma Solid tumour Diabetes mellitus Renal failure Antimicrobial exposurec Azole exposurec Central vascular catheter Parenteral nutrition Mechanical ventilationc Surgeryc Abdominal surgery Chest surgery Cytotoxic chemotherapyc Neutropeniac Renal replacement therapy Prematurity Stem cell transplant Solid organ transplant
<600 beds
>600 beds
Odds ratio (95% CI)
p
0.009
9 (50) 0 (0)
9 (50) 6 (67)
NA
0 2 2 5
4 1 1 3
(44) (11) (11) (33)
NA 0.4 0.4 0.4
(3658) (7573) (2992) (82 755)
NA NA NA NA
2772 0 0 130 733
(0) (22) (22) (55) (2629) (0) (0) (48 302)
6017 11 147 2394 319 446
Total 18 (100) 6 (33) 4 3 3 8
0.20 0.04 0.002 0.058 <0.0001
94 (21.1)
356 (80.1)
49 44 70 39 14
(53) (47) (49–79) (51) (6–27)
189 162 62 138 16
(54) (46) (32–76) (47) (6–30)
NA 0.8 (0.4–1.4) NA
35 21 52 4 3
(37.6) (22.5) (55.9) (4.3) (1–6)
110 76 145 49 3
(31.3) (21.6) (41.3) (13.9) (2–6)
0.7 0.9 0.5 3.6 NA
0 5 16 28 10 83 11 66 32 52 39 23 5 6 3 9 10 0 1
(0) (5.3) (17.2) (30.1) (10.7) (89.2) (11.8) (70.9) (34.4) (55.9) (41.9) (24.7) (5.3) (6.4) (3.2) (9.6) (10.7) (0) (1.0)
34 18 97 80 50 327 48 265 115 147 136 64 21 76 51 30 19 22 3
(9.6) (5.1) (27.6) (22.7) (14.2) (93.1) (13.6) (75.5) (32.7) (41.8) (38.7) (18.2) (5.9) (21.6) (14.5) (8.5) (5.4) (6.2) (0.8)
2.5c 0.9 1.8 0.6 1.3 1.6 1.1 1.2 0.9 0.5 0.8 0.6 1.1 4.0 5.1 0.8 0.4 1.6d 0.7
79 100 19 4 051
(22) (17) (17) (44)
096 (3512) 323 (7738) 150 613 (117 249) 444 (100)
1.0 (0.6–1.6)
(0.4–1.2) (0.5–1.7) (0.3–0.9) (1.2–14.1)
(0.3–3.3) (1.0–3.5) (0.4–1.1) (0.6–3.1) (0.6–3.7) (0.5–2.6) (0.7–2.1) (0.5–1.5) (0.3–0.9) (0.5–1.4) (0.3–1.2) (0.3–3.9) (1.6–11.6) (1.5–26.0) (0.3–2.1) (0.2–1.1) (0.06–42.0)
0.9 0.9 <0.0001 0.6 0.3
238 206 65 177 15
(53.6) (46.4) (33–77) (48.1) (6–30)
0.2 0.8 0.01 0.01 0.8
145 74 197 53 3
(32.6) (16.6) (44.4) (11.9) (2–6)
0.001 1.0 0.04 0.1 0.5 0.2 0.7 0.4 0.8 0.02 0.6 0.1 1.0 0.0008 0.003 0.6 0.09 0.01 1.0
34 23 113 108 60 410 59 331 147 199 175 87 26 82 54 39 29 22 4
(7.6) (5.1) (25.4) (24.3) (13.5) (92.3) (13.2) (74.5) (33.1) (44.8) (39.4) (19.6) (5.8) (18.5) (12.2) (8.7) (6.5) (5.0) (0.9)
HSCT, haematopoietic stem-cell transplant; IQR, interquartile range; NA, not applicable. All values denote number of patients (percentage), except where specified otherwise. p values are for Fisher’s exact test for categorical variables and Student’s t test for scalar variables. Values in bold represent variables that were significantly differently distributed between small (<600 beds) and large (>600 beds) hospitals. a Defined as partial or complete dependence on others for activities of daily living. b Time (days) from date of hospital admission to date of first blood culture positive for Candida spp. c Within 30 days before the onset of candidaemia. d Odds ratio calculated with the Cornfield method. Confidence interval cannot be detrmined because there were no cases in patients from hospitals with <600 beds.
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI
Ben-Ami et al. Fluconazole-resistant Candida
CMI
(n = 77, 17.1%), C. glabrata (n = 69, 15.3%), and C. krusei (n = 14, 3.1%). Six patients had dual-species candidaemia. Fifty-four candidaemia episodes (12.1%) were caused by isolates non-susceptible to fluconazole: 16 (3.6%) susceptible dose-dependent and 38 (8.5%) FRC. Seven centres (38.8%) had oncology services and five (27.7%) had stem-cell transplantation services (Table 1). FRC were found in ten hospitals (rate of resistance in those institutions, 7.1–13.6%). Hospitals where FRC were detected had significantly more total acute beds than hospitals with no FRC—median (interquartile ranges), 752 (657–970) beds versus 405 (310–465) beds, respectively; p 0.009). FRC were detected in 2/9 hospitals (22%) with <600 beds but in 8/9
3
hospitals (88%) with >600 beds (Fig. 1). The pooled incidence of candidaemia was marginally greater in larger hospitals than in smaller ones (11.9 versus 7.3/100 000 patientdays, p 0.050). However, the incidence rate of FRC bloodstream infection was nearly ten-fold higher in larger hospitals (1.2 versus 0.14/100 000 patient-days, p 0.0004). Moreover, there was a significant linear correlation between hospital volume (annual inpatient-days) and the incidence rate of FRC bloodstream infection (Pearson’s coefficient, 0.73, p 0.0005). Fluconazole-resistant C. glabrata and C. krusei occurred exclusively in hospitals with >600 beds: 0.23 versus 0/100 000 patient-days for C. glabrata (p 0.02), and 0.44 versus 0/ 100 000 patient-days for C. krusei (p 0.009; Fig. 1). Resistance
(a)
(b)
(c)
(d)
FIG. 1. Incidence rates of candidaemia by hospital size. Each marker represents a single medical centre. (a) Incidence rates of all candidaemia episodes; (b) incidence rates of fluconazole-resistant Candida bloodstream infection; (c) incidence rates of Candida krusei bloodstream infection; (d) incidence rates of fluconazole-resistant Candida glabrata bloodstream infection. Vertical dashed line marks the median hospital size of 600 beds. ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI
4
Clinical Microbiology and Infection
rates to voriconazole, caspofungin and amphotericin B were not significantly different between low-volume and high-volume hospitals. Demographic and clinical features of patients with candidaemia varied with hospital size. Hospitalization at a hospital with >600 beds was associated with younger age, care in a haematology/oncology service, leukaemia, solid neoplasm, cytotoxic chemotherapy, neutropenia and haematopoietic stem-cell transplantation (Table 1). Three covariates were retained in the multivariable model with FRC bloodstream infection as the dependent variable: neutropenia (odds ratio 3.6, 95% confidence interval 1.6–8.2; p 0.002), receipt of fluconazole within 30 days before candidaemia (OR 4.1, 95% CI 1.8–9.2; p <0.0001) and hospitalization at a medical centre with >600 beds (OR 4.4, 95% CI 1.006– 19.5; p 0.049). Goodness-of-fit for this model was adequate as determined using the Hosmer–Lemeshow test (p 0.5). In this nationwide study in Israel, candidaemia was a frequent nosocomial event, but the proportion of FRC was highly variable across different medical centres, even among those in close geographical proximity. Strikingly, divergent epidemiological patterns were observed in low-volume and high-volume institutions. Hence, although the candidaemia rate was only slightly higher in high-volume institutions, bloodstream infection with FRC occurred almost exclusively at high-volume institutions. FRC infection was strongly associated with the oncological and stem-cell transplantation settings, which were limited to large, tertiary-care medical centres. Our study underscores the need for nationwide studies that incorporate data from diverse nosocomial settings and a range of hospital types [2,8,9,15]. Candidaemia surveillance programmes often only include data from tertiary-care medical centres [3,7,10,14], a design that may select for patient populations at high risk for drug-resistant Candida spp., resulting in potential overestimation of the overall incidence rate of drug resistance. In conclusion, this study, the first of its kind performed in the Middle East, illustrates the epidemiology of candidaemia in Israel and highlights the importance of tertiary-care medical centres, and specifically that of haematology/oncology units, as hotspots for the emergence of FRC. These findings may be used to direct empirical treatment strategies at the institutional level.
Acknowledgements The authors wish to thank Anna Novikov and Nathaniel Albert for excellent technical assistance. The principle investigator (MG) had full access to all of the data in the study
CMI
and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Funding This work was funded in part by a grant from Merck Sharp & Dohme Ltd, Israel and by the Israeli Society for Infectious Diseases.
Transparency Declaration RB-A has received consulting fees and research support from Pfizer; DPK has received research support from and honoraria from Merck, Fujisawa, Enzon, Pfizer, and Schering Plough. All other authors: no conflicts.
Appendix I: Additional Israeli Candidaemia Study Group Invesigators Tel Aviv Sourasky Medical Centre, Tel Aviv: Ruth Orni-Wasserlauf; Hadassah-Hebrew University Medical Centre, Jerusalem: Allon Moses, Karen Olshtain-Pops; Chaim Sheba Medical Centre, Tel Hashomer: Michal Krieger, Yasmin Maor, Anna Goldshmidt-Reuven; Rambam Medical Centre, Haifa: Renato Finkelstein, Hannah Sprecher; Rabin Medical Centre and Schneider Children’s Hospital, Petah Tikva: Michal Paul, Itzhak Levi, Zmira Samra, Elad Goldberg; Wolfson Medical Centre, Holon: Michael Dan, Orna Schwartz-Harari; Shaare Zedek Medical Centre, Jerusalem: Amos Yinon, Yonit Weiner Well; Assaf Harofeh Medical Centre, Zerifin: Tsilia Lazarovitch; Kaplan Medical Centre, Rehovot: Oren Zimhony; Meir Medical Centre, Kfar Saba: BatSheva Gottesman; Bnai Zion Medical Centre, Haifa: Israel Potasman; Emek Medical Centre, Afula: Raul Raz, Bibiana Chazan; Soroka Medical Centre, Beer Sheva: Klaris Riesenberg, Lisa Saidel; Bikur Holim Medical Centre, Jerusalem: Efraim Halperin; Hillel Yafe Medical Centre, Hadera: David Hassin; Poriya Medical Centre, Tiberias: Soboh Soboh.
References 1. Horn DL, Neofytos D, Anaissie EJ et al. Epidemiology and outcomes of candidemia in 2019 patients: data from the prospective antifungal therapy alliance registry. Clin Infect Dis 2009; 48: 1695–1703. 2. Zaoutis TE, Argon J, Chu J, Berlin JA, Walsh TJ, Feudtner C. The epidemiology and attributable outcomes of candidemia in adults and children hospitalized in the United States: a propensity analysis. Clin Infect Dis 2005; 41: 1232–1239.
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI
Ben-Ami et al. Fluconazole-resistant Candida
CMI
3. Colombo AL, Nucci M, Park BJ et al. Epidemiology of candidemia in Brazil: a nationwide sentinel surveillance of candidemia in eleven medical centers. J Clin Microbiol 2006; 44: 2816–2823. 4. Hajjeh RA, Sofair AN, Harrison LH et al. Incidence of bloodstream infections due to Candida species and in vitro susceptibilities of isolates collected from 1998 to 2000 in a population-based active surveillance program. J Clin Microbiol 2004; 42: 1519–1527. 5. St-Germain G, Laverdiere M, Pelletier R et al. Epidemiology and antifungal susceptibility of bloodstream Candida isolates in Quebec: report on 453 cases between 2003 and 2005. Can J Infect Dis Med Microbiol 2008; 19: 55–62. 6. Poikonen E, Lyytikainen O, Anttila VJ et al. Secular trend in candidemia and the use of fluconazole in Finland, 2004–2007. BMC Infect Dis 2010; 10: 312. 7. Poikonen E, Lyytikainen O, Anttila VJ, Ruutu P. Candidemia in Finland, 1995–1999. Emerg Infect Dis 2003; 9: 985–990. 8. Chen S, Slavin M, Nguyen Q et al. Active surveillance for candidemia, Australia. Emerg Infect Dis 2006; 12: 1508–1516. 9. Asmundsdottir LR, Erlendsdottir H, Gottfredsson M. Increasing incidence of candidemia: results from a 20-year nationwide study in Iceland. J Clin Microbiol 2002; 40: 3489–3492. 10. Pappas PG, Rex JH, Lee J et al. A prospective observational study of candidemia: epidemiology, therapy, and influences on mortality in hospitalized adult and pediatric patients. Clin Infect Dis 2003; 37: 634–643. 11. Diekema DJ, Messer SA, Brueggemann AB et al. Epidemiology of candidemia: 3-year results from the emerging infections and the epidemiology of Iowa organisms study. J Clin Microbiol 2002; 40: 1298–1302. 12. Marchetti O, Bille J, Fluckiger U et al. Epidemiology of candidemia in Swiss tertiary care hospitals: secular trends, 1991–2000. Clin Infect Dis 2004; 38: 311–320. 13. Pfaller MA, Moet GJ, Messer SA, Jones RN, Castanheira M. Geographic variations in species distribution and echinocandin and azole
14.
15.
16.
17.
18.
19.
5
antifungal resistance rates among Candida bloodstream infection isolates: report from the SENTRY Antimicrobial Surveillance Program (2008 to 2009). J Clin Microbiol 2011; 49: 396–399. Pfaller MA, Messer SA, Boyken L, Tendolkar S, Hollis RJ, Diekema DJ. Geographic variation in the susceptibilities of invasive isolates of Candida glabrata to seven systemically active antifungal agents: a global assessment from the ARTEMIS Antifungal Surveillance Program conducted in 2001 and 2002. J Clin Microbiol 2004; 42: 3142–3146. Ben-Ami R, Olshtain-Pops K, Krieger M et al. Antibiotic exposure as a risk factor for fluconazole-resistant Candida bloodstream infection. Antimicrob Agents Chemother 2012; 56: 2518–2523. Clinical Laboratory Standards I. Reference Method for Broth Dilution Antifungal Susceptibility testing of Yeasts. Approved Standard M27-A3. Wayne, PA: National Committee for Clinical Laboratory Standards, 2008. Pfaller MA, Andes D, Diekema DJ, Espinel-Ingroff A, Sheehan D. Wild-type MIC distributions, epidemiological cutoff values and species-specific clinical breakpoints for fluconazole and Candida: time for harmonization of CLSI and EUCAST broth microdilution methods. Drug Resist Updat 2010; 13: 180–195. Pfaller MA, Diekema DJ, Andes D et al. Clinical breakpoints for the echinocandins and Candida revisited: integration of molecular, clinical, and microbiological data to arrive at species-specific interpretive criteria. Drug Resist Updat 2011; 14: 164–176. Pfaller MA, Andes D, Arendrup MC et al. Clinical breakpoints for voriconazole and Candida spp. revisited: review of microbiologic, molecular, pharmacodynamic, and clinical data as they pertain to the development of species-specific interpretive criteria. Diagn Microbiol Infect Dis 2011; 70: 330–343.
ª2012 The Authors Clinical Microbiology and Infection ª2012 European Society of Clinical Microbiology and Infectious Diseases, CMI