Skip Navigation

Mixed Fungemia: Incidence, Risk Factors, and Mortality in a General Hospital

  1. Julia Jensen,
  2. Patricia Muñoz,
  3. Jesús Guinea,
  4. Marta Rodríguez-Créixems,
  5. Teresa Peláez, and
  6. Emilio Bouza
  1. Clinical Microbiology and Infectious Diseases Department, Hospital General Universitario “Gregorio Marañón,” Universidad Complutense, Madrid, Spain
  1. Reprints or correspondence: Dr. Patricia Muñoz, Servicio de Microbiologia y Enfermedades Infecciosas, Hospital General Universitario Gregorio Marañón, Doctor Esquerdo 47, 28007 Madrid, Spain (pmunoz{at}micro.hggm.es).

  1. Presented in part: 46th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), San Francisco, 27–30 September 2006 (abstract M-1616).

 
Next Section

Abstract

Background. Fungemia has been historically considered to be a disease caused by a single Candida species; the detection of >1 species of yeast in circulating blood was distinctly uncommon using traditional microbiological procedures. We describe episodes of mixed fungemia (MF), detected between 1985 and 2006, in a large teaching hospital.

Methods. The study was divided into 2 periods that were separated by the introduction, in January 2005, of the CHROmagar Candida medium (CHROMagar) for the routine subculturing of blood cultures in which yeast has been identified. Overall, we documented 747 cases of fungemia. During the first period (1985–1994), we identified 217 episodes of fungemia and no single episode of MF; during the second period (1995–2006), 15 episodes of MF were detected among 530 episodes of fungemia (2.8%). Candida albicans was isolated in 13 patients, non-albicans species of Candida in 16 patients, and Saccharomyces cerevisiae in 1 patient. Each episode of MF was compared with 2 control episodes of monomicrobial fungemia.

Results. Patients with MF had more frequently experienced organ transplantation (13% vs. 0%) and surgery (60% vs. 27%), had less frequently received parenteral nutrition (40% vs. 70%) or had intravenous lines (80% vs. 100%), and had a lower incidence of shock (6% vs. 37%) and a lower mortality (20% vs. 53%).

Conclusions. Despite the introduction of chromogenic agar, MF is still an uncommon disease and has a less severe outcome than does monomicrobial candidemia.

Bacterial bloodstream infections (BSIs) are generally monomicrobial. Polymicrobial bacteremia occurs in 5%–22% of cases that occur in large series [1,2–3] and more frequently in elderly patients who have multiple underlying diseases. Polymicrobial BSIs have been associated with a higher mortality than have monomicrobial episodes [2, 3].

During the past 2 decades, the frequency of invasive fungal infection has increased dramatically among hospitalized patients throughout the world, with Candida species being one of the leading causes of BSI [4, 5]. Data from North American and European surveillance programs have revealed that Candida species occupy the fourth to sixth positions in the list of leading causes of nosocomial BSI [3, 6,7–8].

Fungemia has been classically considered to be a monomicrobial disease [9,10,11,12,13,14,15–16], and the detection of >1 species of yeast in circulating blood samples is uncommonly reported using traditional microbiological blood culture procedures [10,11,12,13,14,15,16,17,18,19–20]. We describe the incidence, clinical consequences, and risk factors of mixed fungemia (MF) in a large teaching hospital.

Previous SectionNext Section

Materials and Methods

Study design and hospital characteristics. We retrospectively studied all episodes of MF that occurred at our institution from January 1985 to May 2006. Ours is a large general-tertiary teaching hospital that serves a population of ∼715,000 inhabitants in Madrid, Spain. The hospital has ∼1750 beds and includes all medical and surgical specialties, including very active solid-organ (e.g., heart, liver, and kidney) and bone marrow transplantation programs.

A prospective database containing data for all episodes of BSI has been maintained by the microbiology service since 1985. Clinical reports of patients with MF were reviewed according to a preestablished protocol. We defined cases of MF as those with an isolation of ⩾2 different species of yeast with clinical significance from blood sample cultures. A patient was considered to have significant fungemia if the yeast was recovered from at least 1 blood culture specimen and was associated with fever or signs of infection. The origin of fungemia was established by means of positive culture results (regardless of whether both yeast species were recovered) and/or with clinical evidence of infection [21]. For example, when 1 or 2 yeasts were also isolated from a catheter tip or from urine samples, we assumed that the origin was the intravenous catheter or the urine, respectively. However, when fungi were isolated from skin, we assumed that this isolation reflected a metastatic skin lesion. Neutropenia was defined as an absolute neutrophil count of <500 cells/mm3. Related mortality was defined as death occurring within 5 days after the onset of fungemia, with concurrent signs of active infection and no other apparent cause.

We also included 2 cases of monomicrobial fungemia per case of MF as controls to identify the risk factors and clinical characteristics of mixed fungemia. The control cases were matched to cases of MF only by the year of positive blood culture result and were selected by means of a random-number table.

Microbiological identification and antifungal susceptibility testing. From 1985 to October 1995, blood samples were processed using the automated system Bactec-NR (Beckton Dickinson); thereafter, processing was done using the Bactec-9240 instrument (Beckton Dickinson). During the first period, all vials were incubated at 35°C for 7 days; in the second period, vials were shaken continuously for 5 days. CHROMagar Candida (CHROMagar), a differential and selective chromogenic agar medium for yeasts, was introduced in our institution in January 1995. Since then, yeast isolates have been systematically subcultured onto CHROMagar Candida plates; these plates are incubated for 5 days at 35°C to identify possible mixed infections. Identification of the isolates was confirmed using the ID 32 C system (bioMérieux), according to the manufacturer's recommendations. The isolates recovered from blood samples were stored at -70°C.

We studied the antifungal activity of amphotericin B, 5-fluorocytosine, ketoconazole, fluconazole, itraconazole, and voriconazole against the clinical isolates. The susceptibility study was performed using Sensititre YeastOne (Trek Diagnostic Systems), a modified microdilution method, according to CLSI M27-A 2 recommendations [22,23–24]. The trays were incubated for 24–48 h at 35°C. End point readings were defined visually as a full inhibition of growth (with amphotericin B) or prominent inhibition of growth (for the remaining antifungal agents).

Data analysis. Categorical data were analyzed using the χ2 test or Fisher's exact test (2-tailed), and the unpaired Student's t test was used for analysis of continuous variables. All comparisons were considered to be statistically significant at P values of ⩽.05. Statistical analysis was performed using the SPSS software package, version 11.5 (SPSS).

Previous SectionNext Section

Results

Incidence and clinical characteristics of MF. Between January 1985 and February 2006, 747 episodes of significant fungemia were detected in our institution (3% of all episodes of significant BSI). During the first period (1985–1994), we detected 217 episodes of fungemia and no episodes of MF. From 1995 to 2006, we detected 530 episodes of fungemia; 15 (2.8%) of these episodes, which occurred in 15 different patients, were MF. The annual distribution of these cases is summarized in table 1. No epidemiological links were found among episodes.

Table 1
View larger version:
Table 1

Clinical data for 15 patients with mixed fungemia.

Candida albicans was the most commonly isolated species from patients who had MF (n = 13); the other species that were isolated were Candida glabrata (n = 6), Candida parapsilosis (n = 4), Candida tropicalis (n = 2), Candida dubliniensis (n = 2), Candida krusei (n = 2), and Saccharomyces cerevisiae (n = 1). Resistance to fluconazole was only detected in 1 isolate of C. glabrata and in C. krusei. The fluconazole-resistant strain of C. glabrata was also resistant to voriconazole (MIC, 4 mg/L).

The clinical characteristics of the 15 patients with MF are shown in table 1. Eleven patients were male, and the median age was 41 years (range, 0–90 years). All patients had significant debilitating conditions. Commonly associated risk factors were the use of central vascular catheters (80%), having undergone previous therapy with broad-spectrum antibiotics (73%), the use of indwelling bladder catheters (73%), having undergone surgery (60%), and having received parenteral nutrition (40%). Five patients (33%) had received antifungal therapy before experiencing fungemia (4 received fluconazole and 1 received amphotericin B). Overall, 53% of patients with MF were admitted to medical wards, 34% to intensive care units, and 13% to surgical wards.

As for clinical presentation, all patients had fever; 4 experienced renal failure; 2 had hepatic abscesses; 2 patients, one who had esophageal carcinoma and the other who was an injection drug user, had related skin lesions; and 1 patient had lung infiltrates and chorioretinitis (table 1). Vascular catheters were the most common source of fungemia (9 cases [60%]), followed by an unknown source (3 cases [20%]), the urine (2 cases [13%]), and the abdomen (1 case [7%]).

Empirical therapy was initiated with fluconazole in 13 patients (11 cases) or with amphotericin B (2 cases) (table 1). The overall mortality rate among patients with MF was 20% (3 patients); all 3 deaths were related to the episode of fungemia (table 2).

Table 2
View larger version:
Table 2

Comparison of the epidemiological characteristics of patients with mixed fungemia with those of patients with monomicrobial fungemia.

Risk factors for MF. For the identification of the differential characteristics of MF, we compared each patient with MF with 2 randomly selected control subjects who had experienced an episode of monomicrobial fungemia. The differences found in the univariate analysis are shown in table 2. Patients with MF more commonly underwent organ transplantation (13% vs. 0%; P = .04) and prior surgery (60% vs. 27%; P = .03), had a lower rate of use of parenteral nutrition (40% vs. 70%; P = .05) and intravenous lines (80% vs. 100%; P = .01), and had a lower incidence of shock (6% vs. 37%; P = .03). With regard to the portal of entry, we did not find any significant difference, although there was a trend toward vascular catheters more commonly being the origin of MF than of monomicrobial fungemia (60% vs. 37%; P = .13), whereas an unknown origin (20% vs. 43%; P = .12) and an abdominal origin (7% vs 13%; P = .50) were more frequently observed in cases of monomicrobial fungemia.

There were no significant differences in the percentage of patients who were treated with antifungal agents (13 [86%] vs. 25 [84%]; P = .77). We found a lower overall mortality rate among patients with MF (20% vs. 53%; P = .03), although we could not find a statistically significant difference in related mortality (20% vs. 40%; P = .18).

Previous SectionNext Section

Discussion

MF has occurred in 2.8% of all episodes of fungemia in our hospital since 1995, when CHROMagar Candida medium was introduced for use in the routine subculturing of positive blood cultures in our institution. Episodes of MF are neither clinically different from monomicrobial episodes nor an indication of a more severe disease.

The incidence of MF is only collaterally mentioned in articles on candidemia in the general population, and it ranges from 2% to 9.3% [9,10,11,12,13,14,15,16,17,18,19–20, 25, 26]. However, only 3 series have focused specifically on mixed candidemia. Guerra-Romero et al. [18], in a retrospective study in Spain that ended in 1988, documented MF in 3.4% of patients. Pulimood et al. [17], in a retrospective study from 1992 to 2000 in the United States, found that MF occurred in 5.2% of patients with candidemia. The frequency of MF does not appear to be higher in cohorts of classically immunocompromised patients with candidemia, according to the data presented by Boktour et al. [19]. They identified 6% of patients with cancer who had candidemia (from 1993 to 2000) due to >1 Candida species. None of these studies used the CHROMagar Candida medium for identification.

In our institution, we pursue a long follow-up period with patients in the general population and use CHROMagar Candida for the identification of fungal isolates. It is now accepted that mixed yeast cultures can remain undetected by routine isolation methods [27, 28]; therefore, many health care centers recommend the use of a differential subculture medium (e.g., CHROMagar Candida) when yeasts are detected in the Gram stain of a blood sample culture.

C. albicans was the most common species isolated, both in the present study and in our review of the literature on MF, followed by C. glabrata, C. parapsilosis, C. tropicalis, C. dubliniensis, C. krusei, and S. cerevisiae. The most common combinations of Candida species were C. albicans and C. glabrata (5 [33%] of 15 cases) and C. albicans and C. parapsilosis (4 [26%] of 15 cases). Among all of our patients, only 2 different fungal species were detected simultaneously, but other studies [17,18,19–20] have demonstrated the anecdotal presence of up to 3 different Candida species recovered from the same blood sample culture.

We could not find any factors that differentiate these patients from those with monomicrobial fungemia, and neither could other authors [17,18,19–20]. Interestingly, only 5 patients in this study were experiencing active malignancies, none of which were hematological; none of these patients had chemotherapy-associated neutropenia, and only 5 (33%) had received prior antifungal therapy, a frequently identified risk factor in patients with monomicrobial candidemia [10, 11, 29,30–31].

The portal of entry of MF was different from that of monomicrobial fungemia, although these differences were not statistically significant. Vascular catheter-related candidemia predominated in cases of MF (60% vs. 37%; P = .13), and an unknown origin was more frequently encountered in cases of monomicrobial fungemia (20% vs. 43%; P = .12). In only 1 patient did a culture of a sample from a catheter tip have positive results for both Candida species. Cultures of samples other than blood were not processed using a chromogenic medium, which may account for this discrepancy. Pulimood et al. [17] reported that catheters were the origin of infection in 8 of 12 patients who had MF, but this incidence was similar to that observed in patients who had monomicrobial fungemia—that is, 67% versus 53%. In addition, that study did not provide the number of patients in whom both Candida species were isolated from both sites.

We could not find significant differences in therapy or related mortality between MF and monomicrobial fungemia. However, the importance of the second strain could be better analyzed in the 3 patients who had a resistant strain. One patient was infected with C. glabrata and died, even though therapy with caspofungin was later initiated. Two patients with C. krusei infection survived; one patient was initially treated with amphotericin B, and the other received fluconazole. The patient who died of MF that consisted of an infection with C. albicans and C. parapsilosis had esophageal cancer and did not receive an antifungal agent in time. The overall mortality rate was higher for cases of monomicrobial fungemia (53%) than for MF (20%; P = .03). These data differ from those of Pulimood et al. [17] and Boktour et al. [19], who found similar mortality (∼50%) in both groups.

This study has several limitations. First, it is a retrospective study in which patients were recruited over a long period of time and in which different fungal culture media were used. We cannot exclude the possibility that the patient population changed. Second, the number of cases of MF is small, which may account for our inability to find differences in some comparisons. In addition, with such low numbers, only a few cases can markedly influence outcome.

In summary, despite the introduction of chromogenic agar, MF is still an uncommon disease and has a less severe outcome than does monomicrobial candidemia.

Previous SectionNext Section

acknowledgments

Financial support. Programa de Centros de Investigación Biomédica en Red (CIBER) de Enfermedades Respiratorias (CB06/06/0058) and Fondo de Investigación Sanitaria (FIS; CM05/00171, to J.G.).

Potential conflicts of interest. All authors: no conflicts.

  • Received December 5, 2006.
  • Accepted February 28, 2007.
Previous Section
 

references

  1. ↵
    1. Cooper GS,
    2. Havlir DS,
    3. Shlaes DM,
    4. Salata RA
    . Polymicrobial bacteremia in the late 1980s: predictors of outcome and review of the literature. Medicine (Baltimore) 1990;69:114-23.
    Medline
  2. ↵
    1. Weinstein P,
    2. Reller L,
    3. Murphy J
    . Clinical importance of polymicrobial bacteremia. Diagn Microbiol Infect Dis 1986;5:185-96.
    CrossRefMedlineWeb of Science
  3. ↵
    1. Bouza E,
    2. Pérez-Molina J,
    3. Muñoz P
    . Report of ESGNI-001 and ESGNI-002 studies: bloodstream infections in Europe. Clin Microbiol Infect 1999;5:2S1-12.
    CrossRef
  4. ↵
    1. Beck-Sagué CM,
    2. Jarvis WR
    . Secular trends in the epidemiology of nosocomial fungal infections in the United States, 1980–1990. National Nosocomial Infections Surveillance System. J Infect Dis 1993;167:1247-51.
    MedlineWeb of Science
  5. ↵
    1. Harbarth S,
    2. Reuf C,
    3. Francioli P
    . Nosocomial infections in Swiss university hospitals: a multi-centre survey and review of the published experience. Swiss-Noso Network. Schweiz Med Wochenschr 1999;129:1521-8.
    MedlineWeb of Science
  6. ↵
    1. Edmond MB,
    2. Wallace SE,
    3. McClish D
    . Nosocomial bloostream infections in the United States hospitals: a three-year analysis. Clin Infect Dis 1999;29:239-44.
    MedlineWeb of Science
  7. ↵
    1. Coello R,
    2. Charlett A,
    3. Ward V
    . Deviced-related sources of bacteraemia in English hospitals—opportunities for the prevention of the hospital acquired bacteraemia. J Hosp Infect 2003;53:46-57.
    CrossRefMedlineWeb of Science
  8. ↵
    1. Marchetti O,
    2. Bille J,
    3. Fluckiger U,
    4. et al
    . Epidemiology of candidemia in Swiss tertiary care hospitals: secular trends, 1991–2000. Clin Infect Dis 2004;38:311-20.
    Abstract/FREE Full Text
  9. ↵
    1. Fraser VJ,
    2. Jones M,
    3. Dunkel J,
    4. Storfer S,
    5. Medoff G,
    6. Dunagan WC
    . Candidemia in a tertiary care hospital: epidemiology, risk factors, and predictors of mortality. Clin Infect Dis 1992;15:414-21.
    Abstract/FREE Full Text
  10. ↵
    1. Nguyen MH,
    2. Peacock JE Jr,
    3. Morris AJ,
    4. et al
    . The changing face of candidemia: emergence of non-Candida albicans species and antifungal resistance. Am J Med 1996;100:617-23.
    CrossRefMedlineWeb of Science
  11. ↵
    1. Abi-Said D,
    2. Anaissie E,
    3. Uzun O,
    4. Raad I,
    5. Pinzcowski H,
    6. Vartivarian S
    . The epidemiology of hematogenous candidiasis caused by different Candida species. Clin Infect Dis 1997;24:1122-8.
    Abstract/FREE Full Text
  12. ↵
    1. Viudes A,
    2. Peman J,
    3. Canton E,
    4. Ubeda P,
    5. Lopez-Ribot J,
    6. Gobernado M
    . Candidemia at a tertiary-care hospital: epidemiology, treatment, clinical outcome and risk factors for death. Eur J Clin Microbiol Infect Dis 2002;21:767-74.
    CrossRefMedlineWeb of Science
  13. ↵
    1. Viscoli C,
    2. Girmenia C,
    3. Marinus A
    . Candidemia in cancer patients: a prospective, multicenter surveillance study by the Invasive Fungal Infections Group B (IFIG) of the European Organization for Research and Treatment of Cancer (EORTC). Clin Infect Dis 1999;28:1071-9.
    Abstract/FREE Full Text
  14. ↵
    1. Garbino J,
    2. Kolarova L,
    3. Rohner P,
    4. Lew D,
    5. Pichna P,
    6. Pittet D
    . Secular trends of candidemia over 12 years in adult patients at a tertiary care hospital. Medicine (Baltimore) 2002;81:425-33.
    CrossRefMedline
  15. ↵
    1. Luzzati R,
    2. Allegranzi B,
    3. Antozzi L,
    4. et al
    . Secular trends in nosocomial candidaemia in non-neutropenic patients in an Italian tertiary hospital. Clin Microbiol Infect 2005;11:908-13.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Pappas PG,
    2. Rex JH,
    3. Lee J
    . A prospective observational study of candidemia: epidemiology, therapy, and influences on mortality in hospitalized adult and pediatric patients. Clin Infect Dis 2003;37:634-43.
    Abstract/FREE Full Text
  17. ↵
    1. Pulimood S,
    2. Ganesan L,
    3. Alangaden G,
    4. Chandrasekar PH
    . Polymicrobial candidemia. Diagn Microbiol Infect Dis 2002;44:353-7.
    CrossRefMedlineWeb of Science
  18. ↵
    1. Guerra-Romero L,
    2. Telenti A,
    3. Thompson R
    . Polymicrobial fungemia: microbiology, clinical features and significance. Rev Infect Dis 1989;11:208-12.
    MedlineWeb of Science
  19. ↵
    1. Boktour M,
    2. Kontoyiannis DP,
    3. Hanna HA,
    4. et al
    . Multiple-species candidemia in patients with cancer. Cancer 2004;101:1860-5.
    CrossRefMedlineWeb of Science
  20. ↵
    1. Jonhson M,
    2. Gottfredsson M,
    3. Perfect J
    . Program and abstracts of the 42nd Interscience Conference on Antimicrobial Agents and Chemotherapy (San Diego, CA). Washington, DC: American Society for Microbiology; 2002. Polyfungemia at a tertiary care medical center 1992–2001: a decade of experience [abstract M881]; p. 391.
  21. ↵
    1. Raad I
    . Management of intravascular catheter-related infections. J Antimicrob Chemother 2000;45:267-70.
    FREE Full Text
  22. ↵
    1. NCCLS
    . Approved standard M27-A. Wayne, PA: NCCLS; 1997. Reference method for broth dilution antifungal susceptibility testing of yeasts.
  23. ↵
    1. Rodriguez-Tudela JL,
    2. Martinez-Suarez JV
    . Defining conditions for microbroth antifungal susceptibility tests: influence of RPMI and RPMI-2% glucose on the selection of endpoint criteria. J Antimicrob Chemother 1995;35:739-49.
    Abstract/FREE Full Text
  24. ↵
    1. Rodriguez-Tudela JL,
    2. Berenguer J,
    3. Martinez-Suarez JV,
    4. Sanchez R
    . Comparison of a spectrophotometric microdilution method with RPMI-2% glucose with the National Committee for Clinical Laboratory Standards reference macrodilution method M27-P for in vitro susceptibility testing of amphotericin B, flucytosine, and fluconazole against Candida albicans. Antimicrob Agents Chemother 1996;40:1998-2003.
    Abstract/FREE Full Text
  25. ↵
    1. Bedini A,
    2. Venturelli C,
    3. Mussini C,
    4. et al
    . Epidemiology of candidaemia and antifungal susceptibility patterns in an Italian tertiary-care hospital. Clin Microbiol Infect 2006;12:75-80.
    CrossRefMedlineWeb of Science
  26. ↵
    1. Rex JH,
    2. Bennett JE,
    3. Sugar AM,
    4. et al
    . A randomized trial comparing fluconazole with amphotericin B for the treatment of candidemia in patients without neutropenia. Candidemia Study Group and the National Institute. N Engl J Med 1994;331:1325-30.
    CrossRefMedlineWeb of Science
  27. ↵
    1. Yera H,
    2. Poulain D,
    3. Lefebvre A,
    4. Camus D,
    5. Sendid B
    . Polymicrobial candidaemia revealed by peripheral blood smear and chromogenic medium. J Clin Pathol 2004;57:196-8.
    Abstract/FREE Full Text
  28. ↵
    1. Odds FC,
    2. Bernaerts R
    . CHROMagar Candida, a new differential isolation medium for presumptive identification of clinically important Candida species. J Clin Microbiol 1994;32:1923-9.
    Abstract/FREE Full Text
  29. ↵
    1. Karabinis A,
    2. Hill C,
    3. Leclerq B
    . Risk factors for candidemia in cancer patients. J Clin Microbiol 1988;26:429-32.
    Abstract/FREE Full Text
  30. ↵
    1. Harvey R,
    2. Myers J
    . Nosocomial fungemia in a large community teaching hospital. Arch Intern Med 1987;147:2117-20.
    Abstract/FREE Full Text
  31. ↵
    1. Richet H,
    2. Andremont A,
    3. Tancrede C
    . Risk factors for candidemia in patients with acute lymphocytic leukemia. Rev Infect Dis 1991;13:211-5.
    MedlineWeb of Science
| Table of Contents

This Article

  1. Clin Infect Dis. (2007) 44 (12): e109-e114. doi: 10.1086/518175
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most