Nosocomial candidemia at a general hospital: prognostic factors and impact of early empiric treatment on outcome (2002–2005)
Candidemia nosocomial en un hospital general: factores pronósticos e impacto del tratamiento empírico precoz (2002-2005)
a Internal Medicine-Infectious Diseases Department, University Hospital “Virgen de la Arrixaca”, Murcia, España
b Microbiology Department, University Hospital “Virgen de la Arrixaca”, Murcia, España
c Department of Biostatistics, Faculty of Medicine, University of Murcia, Murcia, España
d Medicine Department, Faculty of Medicine, University of Murcia, Murcia, España
KeywordsNosocomial candidemia. Candida sp.. Mortality. Prognosis factors. Treatment.
Palabras ClaveMortalidad. Tratamiento. Factores pronósticos. Candidemia nosocomial. Candida sp.
To evaluate epidemiological and clinical prognosis factors related to mortality and impact of early empiric treatment on patients with nosocomial candidemia (NC).Patients and methods
Observational study of a cohort of 107 adult patients with NC admitted at a tertiary hospital (2002–5).Results
In bivariate analysis, risk factors significantly associated with mortality rate (49.5%) were: age >65 years, previous steroid treatment, solid organ transplant, acute severity of illness, shock, renal failure and respiratory distress at onset, delayed or inadequate antifungal treatment, non-removal of central venous catheter and associated post-surgical bacterial sepsis or respiratory infection. In multivariate analysis, risk factor associated with mortality was acute severity of illness at onset (OR 76.9; CI 12.5–500) being early and adequate treatment (OR 11.8; CI 1.7–81.2) and early (<48h) removing of central venous catheter (OR 12.2; CI 1.9–74.9) factors associated with cure; there was no statistically significant difference between fungistatic (azoles) or fungicidal (amphotericin or caspofungin) treatment.Conclusions
Acute severity of illness at onset is associated with mortality in patients with NC whereas early and adequate treatment and early removing of central venous catheter are associated with cure.
Candidemia represents an increasing problem in clinical practice. It is linked to mortality, with figures that oscillate between 40–60%, longer average hospital stays and a higher economic cost per process.1,2,3
Among the prognostic factors that have been described as associated with higher mortality, the following can be mentioned: a) patients factors: malignant haematological disease, need of stay at an Intensive Care Unit (ICU), mechanical ventilation and presence of central venous catheters4,5; b) infection factors: presence of septic syndrome, thrombocytopenia, renal failure and candidemia due to Candida krusei5,6,7 c) treatment factors: delayed or inadequate antifungal treatment and non-removal of the central venous catheter.3,4,8,9
In view of these clinical problems, the aim of our study is to evaluate the epidemiological and clinical prognosis factors that are associated with mortality in patients with nosocomial candidemia and especially the impact of early empiric treatment on their outcome.Patients and Methods Hospital characteristics
The Hospital Universitario Virgen de la Arrixaca in Murcia is a 944-beds facility, 611 of which belong to the general hospital. Its area of care encompasses a population of approximately 450,000 individuals. It is, in addition, a reference hospital for certain specialties such as neurosurgery, burn trauma, cardiovascular surgery, and organ transplantation.Study period
All episodes of bloodstream infection due to Candida sp. in adult patients during the period from January 2002 to may 2005 were included.Microbiological Study
The isolation and detection of the type of Candida sp. was done following the microbiological protocols and in accordance with the international regulations of the Centres for Diseases Control.10
The study of susceptibility to amphotericin and fluconazole was done by using the Yeast One method (TREK Diagnosis Systems, Cleveland. Ohio) following the NCCLS criteria.11 Thus, Candida sp. was considered to be resistant to fluconazole in the case of a MIC >64μg/mL, of intermediate susceptibility or of dose-dependent resistance if MIC was comprised between 16–32μg/mL and susceptible if MIC <16μg/mL. Resistance to amphotericin was defined if MIC >1μg/mL. The quality control was carried out using Candida parapsilosis (ATCC 22019).Patients’ assessment
Candidemia was considered to be of community acquisition whenever Candida sp. was isolated within the first 72h of hospital admission, provided the patient had not been hospitalised in the previous month. Candidemia was considered to be of nosocomial acquisition when a positive blood culture was obtained after the first 72h of hospitalisation in patients who showed no symptoms or signs of infection at the time of hospital admission. Likewise, when a positive culture was obtained within the first 72h of hospitalisation in a patient who had been hospitalised during the previous month, infection was understood to be of nosocomial acquisition. The retrospective review of the clinical records was done following a study protocol in which epidemiological, clinical and microbiological variables were collected.
Respiratory distress was defined as bilateral alveolar infiltrates without data of heart failure; renal failure was considered as an increase in previous creatinine >0.5mg/L (at time of NC). A case was defined as likely to be catheter related when a semiquantitative culture of the catheter tip yielded more than 15 CFU of the same Candida species isolated in the blood sample obtained through the catheter or a peripheral vein.
Patients were assessed with regard to the prognosis of their underlying disease according to the criteria of McCabe and Jackson12. Their condition was then classified as “rapidly fatal” when death was expected to take place in days or weeks, “ultimately fatal” when death was likely to take place in some months or years, and “nonfatal” when death was not predictable. The severity of the patient's condition at presentation was assessed according to Winston et al.,13 as follows: “critical” when the patient's clinical condition was rapidly deteriorating and the probability of death during the first 24h was high; “poor” when the clinical condition was deteriorating and death was probable but not imminent; “fair” when the clinical condition was deteriorating but death was not probable; and “good” when the clinical situation did not change during the first 24h and death was not probable.Prognostic Factors
In order to identify prognostic factors, all clinical and epidemiological characteristics, all complications, and the type of antifungal treatment were examined in relation to the final evolution of patients, with “recovery” understood to be the disappearance of all active signs and symptoms of infection.14 When the physician considered the death of the patient to be related to the infection, it was designated as “death related to candidemia.” If the physician considered that death had occurred after recovery from the infection and was related to the underlying disease or other medical or surgical complications, it was classified as “death not related to candidemia.” The death of the patient was taken into consideration when it took place during admission to hospital, in accordance with previous studies.9
Among the epidemiological factors, the presence of central venous and urinary catheters or nasogastric tube, the treatment with parenteral nutrition and the previous use of antibiotics were included. As for the latter, the type of agent and number of cycles received were taken into account, a cycle being considered when the patient had received antibiotic treatment with a minimum duration of 7 days in the previous 6 weeks before NC. The previous use of antifungal agents was defined as the administration of this type of drugs in the previous 6 weeks before candidemia.Treatment
The antifungal used was considered active when, according to microbiological data, it presented in vitro activity (MIC that allowed to define the strain as susceptible) against the Candida sp. isolated. We considered the initial empiric antifungal treatment to be appropriate when it exhibited in vitro activity against the corresponding Candida sp. strain isolated, when it was used in a correct dosage, and thus its use did not have to be modified after the antibiogram results were obtained. The efficacy of the treatment was considered to be assessable after at least 5 days of administration. The duration of the treatment was adjusted to the international norms, requiring a minimum of 14 days to be considered as complete and up to 30–40 days in the case of development of complications.15
“Early” treatment was defined as the empiric treatment initiated facing the presence of fever and before having any microbiological information or in the first 48h after receiving the notification of positive blood culture for yeasts. On the other hand, treatment initiated after 48h of receiving the microbiological information was estimated as “late” treatment.
The removal of the venous catheter was considered as appropriate when it was done in the first 48h after having received the microbiological information.Statistical study
Data were analysed using the statistics program SPSS 15.0 (SPSS Software, Chicago, IL, USA). A descriptive study was performed for the clinical and epidemiological characteristics as well as for the prognostic factors of patients with NC. The relation or association between pairs of qualitative variables was determined through analyses of contingency tables by means of Pearson's chisquared test, complemented by an analysis of residues with the aim of determining the directional dependence. In the case of quantitative variables, means were compared using Student's t test. The difference was considered significant at p<0.05. Bivariate and multivariate analyses of the cohort of patients with NC were performed in order to detect those factors related to mortality. The multivariate analysis comprised a non-conditioned logistic regression, where the death of the patient was taken as the dependent variable and all those variables statistically associated with a higher mortality rate in the bivariate analysis were included as independent variables.Results
The complete medical records of 107 patients with NC were available, 79 out of whom were men and 28 women, with an average age of 61 years (range 17–87); 53 patients died (49.5%). All patients had received previous antibiotic treatment. Culture of the catheter tip was positive in 40 patients.
In the bivariate study, the statistically significant factors related to death were the following: age >65 years, previous steroid treatment, antecedent of solid organ transplantation, acute severity of illness at onset, shock, renal failure, respiratory distress, delayed or inadequate antifungal treatment, non-removal of central venous catheter and concomitant respiratory infection or postsurgical infection. Patients with infection by resistant Candida sp. (MIC >64μg/mol) or with diminished susceptibility (MIC 16–32μg/mol) to fluconazole were linked to failure of empiric treatment with fluconazole in the bivariate analysis but it was not significant in the multivariate one. Nine patients with resistant or intermediate to fluconazole NC died, out of whom 2 had received adequate treatment (amphotericin) and 7 did not (5 treated with fluconazole and 2 who did not receive any treatment); 8 patients in this category survived, all having received adequate treatment (5 with amphotericin, 1 with caposfungin, 1 with voriconazol and 1 with high dose fluconazole). The assessment of NC by Candida albicans versus non C. albicans was not significantly associated with worse prognosis, neither was the use of fungistatics (azoles: n=67) or fungicidals (amphotericin; n=27) or caspofungin; n=1) (tabla 1).
Table 1. Nosocomial Candidemia: prognostic factors associated with mortality.
|Variables||Recovery (N=54) n (%)||Death (N=53) n (%)||p||OR (95% CI)|
|<65||30 (60)||20 (40)||<0.05||ns|
|>65||24 (42.1)||33 (57.9)|
|Male||37 (46.8)||42 (53,2)||ns||---|
|Female||17 (60.7)||11 (39.3)|
|Yes||14 (51.9)||13 (48.1)||ns||---|
|No||40 (50)||40 (50)|
|Previous steroid treatment|
|Yes||15 (35.6)||26 (63.4)||<0.05||ns|
|No||39 (59.1)||27 (40.9)|
|Solid organ neoplasia|
|Yes||11 (39.3)||17 (60.7)||ns||---|
|No||43 (54.4)||36 (45.6)|
|Solid organ transplant|
|Yes||0 (0)||6 (100)||<0.05||ns|
|No||54 (53.5)||47 (46.5)|
|Neutropenia <1000 cel/μl|
|Yes||3 (37.5)||5 (62.5)||ns||---|
|No||51 (51.5)||48 (48.5)|
|Yes||48 (49.5)||49 (50.5)||ns||---|
|No||6 (60)||4 (40)|
|Initial clinical condition (Winston et al.)|
|Critical||6 (11.5)||46 (88.5)||<0.05||76.9 (12.5–500)|
|Poor||34 (83)||7 (17)|
|Fair||14 (100)||0 (0)|
|Associated bacterial Infection|
|No||30 (60)||20 (40)||ns||---|
|Yes||24 (42.1)||33 (57.9)|
|Respiratory||0 (0)||5 (100)||<0.05||ns|
|Sepsis||22 (40.7)||22 (59.3)||ns||---|
|Postsurgical||1 (20)||4 (80)||<0.05||ns|
|Peritonitis||1 (50)||1 (50)||ns||---|
|Flebitis||0 (0)||1 (100)||ns|
|Yes||7 (19.5)||29 (80.5)||<0.05||ns|
|No||47 (66.2)||24 (33.8)|
|Yes||16 (34.8)||30 (65.2)||<0.05||ns|
|No||38 (62.3)||23 (37.7)|
|Yes||12 (28.6)||30 (71.4)||<0.05||ns|
|No||42 (64.6)||23 (35.4)|
|Medical||17 (63)||10 (37)||ns||---|
|Surgical||36 (50)||36 (50)|
|Oncology-Haematology||1 (12.5)||7 (87.5)|
|C. albicans||25 (51)||24 (49)||ns||---|
|C. parapsilosis||14 (53.8)||12 (46.2)|
|C. tropicalis||8 (66.7)||4 (13.3)|
|C. glabrata||6 (46.2)||7 (53.8)|
|C. krusei||0 (0)||2 (100)|
|C. lusitaniae||0 (0)||1 (100)|
|Various Candida sp.||1 (25)||3 (75)|
|Adequate||54 (62.8)||32 (37.8)||<0.05||ns|
|Inadequate||0 (0)||9 (100)|
|No treatment||0 (0)||12 (100)|
|Early and adequate||50 (92,6)||23 (43.4)||<0.05||11.8 (1.7–81.2) ¿|
|Late||4 (7,4)||18 (34)|
|No treatment||0 (0)||12 (22.6)|
|Fungistatic||36 (53.7)||31 (46.3)||ns||---|
|Fungicidal||18 (64.3)||10 (35.7)|
|Early removal of central venous catheter|
|Yes||45 (71.4)||18 (28.6)||<0.05||12.2 (1.9–74.9) ¿|
|No||4 (11.8)||30 (88.2)|
|Empiric Treatment in case of resistant or intermediate susceptibility to fluconazole Candida sp.|
|Adequate||8 (80)||2 (20)||<0.05|
|Non adequate||0 (0)||7 (100)|
¿ Factors associated with lower mortality (protectors).
In multivariate analysis, acute severity of illness at onset was associated with mortality (OR 76.9; CI 12.5–500), being early and adequate treatment (OR 11.8; CI 1.7–81.2) and early removal (<48h) of central venous catheter (OR 12.2; CI 1.9-74.9) the factors related to survival (tabla 1).Discussion
Mortality in our study (49.5%) is similar to other series,1,2,3,4,5 in which this variable has not undergone significant changes in the late decades, oscillating between 40–60%. The high mortality in patients with NC is probably related to the greater seriousness of their underlying diseases2,4,5,16. Patients with cancer are prone to this infectious complication and neoplasia is one of the factors that has been associated with higher mortality, yet it was not the case in our series. This fact is probably related to the protocolized prophylactic use of fluconazole in these subjects. The influence that the increase in the consumption of fluconazole has on the development of resistance represents a controversial aspect.17 Like other researchers,18 we have not found differences of mortality between the group of patients with C. albicans and those with non C. albicans.
In our cohort of patients with nosocomial candidemia acute severity of illness at onset, non removal of central venous catheter and late antifungal treatment are the only statistically significant factors associated with death. These results are similar to those of other studies of the literature2,3,4,5,19,20. The importance of catheter removal may be related to the high rate of patients with parenteral nutrition in our cohort. Therefore the initial clinical assessment of these patients is of great relevance as it is the need of providing them with early empiric treatment that allows improving their prognosis and evolution. We consider that initial clinical seriousness of the patient must be taken into account when taking therapeutic decisions, especially if we bear in mind that in our cohort early and adequate antifungal treatment was determinant in the prognosis of candidemia (OR 11.8; 95% CI 1.7–81.2), as similar studies have concluded.2,4,5,9,17,21,22 Thus, in a retrospective study of 157 cases of candidemia Morrell et al.9 found that when antifungal treatment was initiated after 12 or 48h of knowing the existence of positive blood culture for yeasts, mortality rose to 31% and 35% respectively, while it significantly decreased to 11% when treatment was initiated within the first 12h. Moreover, Almirante et al.4 noticed that late removal (>5 days after the beginning of treatment) or non-removal of central venous catheter were associated with higher mortality, whereas early antifungal treatment and early removal of venous catheter were related to lower mortality. In our series early removal of central venous catheter was linked to survival (OR 12.2; 95% CI 1.9–75).
In our cohort, patients with infection by resistant or diminished susceptibility to fluconazol Candida sp. were also linked to failure of empiric treatment with fluconazole in the bivariate analysis as it has been described in other cohorts,22,23 but it was not significant in the multivariate one. Nevertheless the small simple size makes difficult to reach any conclusion. In this regard, the knowledge of the previous use of fluconazole is essential, as in these cases the MIC of other azoles (e.g. voriconazole) increases, thus possibly determining their clinical effectiveness in the treatment of Candidemias, especially in patients with acute seriousness of illness or seriously immune-depressed . Once susceptibility data are known, a voriconazole MIC <1μg/mL is considered to have a good correlation with its clinical effectiveness24 and to offer an effectiveness of 58% in the cases of Candida sp. infection refractory to previous treatment with other azoles.25 In our series, similar to other cohorts,26,27 we have not found significant differences among the patients who received fungistatic (azoles) or fungicidal treatment (amphotericin or caspofungin) but this may be related to the small sample size.
Current recommendations for the empiric treatment of patients with NC before having the susceptibility study of an already documented candidemia include stratification according to the risk group, just as it is reflected in different works and in the new IDSA therapeutic guidelines:28,29 a) candidemia in non-critically immunocompromised patient, high-dose fluconazole or voriconazole, with special attention to patients with previous use of azoles; b) candidemia in critically immunocompromised patient, amphotericin-B liposomal or equinocandins (caspofungin or anidalofungin), taking into account the resistance in case of risk of infection due to Candida parapsilosis.
We can conclude that patients with NC had a high mortality and, therefore, when facing a situation of fever and sepsis with acute seriousness at onset and in patients with risk factors already described in other series30,31 (central venous catheter, parenteral nutrition, previous use of broad-spectrum antibiotics, previous surgery, immunocompromised patients and prolonged hospital stay), a quick action with early diagnostic suspicion and timely adequate initial empiric treatment is determinant to improve patients’ care.Conflict of interest
The authors have no conflict of interest to declare.
Received 10 March 2009
Accepted 16 June 2009
Corresponding author. email@example.com
Bibliografía1.Moreira-Oliveira MS, Mikani Y, Miyagi M, Imai T, Schreiber AZ, Maretti M.L. Diagnosis of candidemia by polymerase chain reaction and blood culture: prospective study in a high-risk population and identification of variables associated with development of candidemia. Eur J Clin Microbiol Infect Dis. 2005; 24:721-6.
2.Gudlaugson D, Gillespie S, Lee K, Berg JV, Hu J, Messer S, et-al. Attributable mortality of Nosocomial candidemia. Clin Infect Dis. 2003; 37:1172-7.
3.Pfaller MA, Diekema D.J. Epidemiology of Invasive Candidiasis: a persistent Public Health Problem. Clin Microbiol Rev. 2007; 20:133-63.
4.Almirante B, Rodríguez D, Park BJ, Cuenca-Estrella M, Planes AM, Almela M, et-al. Epidemiology and Predictors of Mortality in Cases of Candida Bloodstream Infection: Results from population-Based Surveillance, Barcelona, Spain, from 2002-2003. J Clin Microbiol. 2005; 43:1829-35.
5.Alonso-Valle H, Acha O, García-Palomo JD, Fariñas-Alvarez C, Fernández-Mazarrasa C, Fariñas M.C. Candidaemia at a tertiary care hospital: epidemiology and factors influencing mortality. Eur J Clin Microbiol Infect Dis. 2003; 22:254-7.
6.Ben-Abraham R, Keller N, Teodorovitch N, Barzilai A, Harel R, Barzilai Z, et-al. Predictors of adverse outcome from Candida infection in a tertiary care hospital. Journal of Infection. 2004; 49:317-27.
7.Peman J, Canton E, Gobernado M. and the Spanish ECMM Working Group on Candidaemia. Eur J Clin Microbiol Infect Dis. 2005; 24:23-30.
8.Boo TW, O’Reilly B, O’Leary J, Cryan B. Candidaemia in an Irish tertiary referral hospital: epidemiology and prognostic factors. Mycoses. 2005; 48:251-9.
9.Morrell M, Fraser VJ, Kollef M.H. Delaying the empiric treatment of Candida Bloodstream infection until blood culture results are obtained: a Potential risk factor for hospital mortality. Antimicrob Ag Chemother. 2005; 49:3640-5.
10.Warren NG, Hazen K.C. Candida, Cryptococcus and others yeast of medical importance. En: Murray R.P., Baron E.J., Pfaller M.A., Tenover F.C., Yolken R.H., editors. Manual of Clinical Microbiology. 7th. American Society for Microbiology: Washington DC; 1999. 1184-99.
11.National Committee for Clinical Laboratory Standards. Reference method for broath dilution antifungal susceptibility testing of yeast: approved standard. Document M27-A2. 2nd ed. Wayne Pa: National Committee for Clinical Laboratory Standard; 2002.
12.McCabe WR, Jackson G.G. Gram negative bacteremia: I Ecology and etiology. Arch Intern Med. 1962; 110:847-55.
13.Winston DJ, Murphy W, Young L.S. Piperacillin therapy for serious bacterial infections. Am J Med. 1980; 69:225-31.
14.Medoff G, Dismukes WE, Pappagianis D, Diamond R, Gallis HA, Drutz D. General guidelines for the evaluation of new antifungal drugs for the treatment of systemic fungal infections. Clin Infect Dis. 1992; 15:S274-81.
15.Oude-Lashof A.M.L, Donnelly JP, Meis J.F.G.M, Meer J.W.M, Kullberg B.J. Duration of antifungal treatment and development of delayed complications in patients with candidaemia. Eur J Clin Microbiol Infect Dis. 2003; 22:43-8.
16.Safdar A, Bannister TW, Safdar Z. The predictors of outcome in immunocompetent patients with hematogenous candidiasis. Intern J of Infect Dis. 2004; 8:180-6.
17.Marr KA, Seidel K, White TH, Bowden R.A. Candidemia in allogenic blood and marrow transplant recipients: evolution of risk factors after the administration of prophylactic fluconazole. J Infect Dis. 2000; 181:309-16.
18.Pappas PG, Rex JH, Lee J, Hamill RJ, Larsen RA, Powderly W, et-al. A prospective observational study of Candidemia: Epidemiology, Therapy, and Influence on mortality in Hospitalized adult and pediatric patients. Clin Infect Dis. 2003; 37:634-43.
19.Asmundsdottir LR, Erlendsdottir H, Gottfredsson M. Improving survival of patients with candidaemia: analysis of prognosis factors from a long-term, nationwide study in Iceland. Scand J Infect Dis. 2005; 37:111-20.
20.Alvarez-Lerma F, Palomar M, León C, Olaechea P, Cerdá E, Bermejo B. Colonización y/o infección por hongos en unidades de cuidados intensivos. Estudio multicéntrico de 1562 pacientes. Med Clin. 2003; 12:161-6.
21.MacCallum DM, Odds F.C. Need for early antifungal treatment confirmed in experimental disseminated Candida albicans infections. Antimicrob Ag Chemother. 2004; 48:4911-4.
22.Morgan J, Meltzer MI, Plikaytis BD, Sofair AN, Huie-White S, Wilcox S, et-al. Excess mortality, hospital stay, and cost due to candidemia: a case-control study using date from population-based candidemia surveillance. Infect Control and Hosp Epidemiol. 2005; 26:540-7.
23.Magill SS, Shields CH, Sears CL, Choti M, Merz W.G. Triazole cross-resistance among Candida spp: case report, occurrence among bloodstream isolates, and implications for antifungal therapy. J Clin Microbiol. 2006; 44:529-35.
24.Pfaller MA, Diekema DJ, Rez JH, Espinelle-Ingroff A, Johnson EM, Anes D, et-al. Correlation between CMI and response for Candida sp. Studies in front of voriconazole: Analysis and proposed of breakpoint interpretative. Antimicrob Ag Chemother. 2006; 44:819-26.
25.Ostrosky-Zeichner L, Oude-Lashof A.M.L, Kullberg BJ, Rex J.H. Voriconazole salvage treatment of invasive candidiasis. Eur J Clin Microbiol Infect Dis. 2003; 22:651-5.
26.Rex JH, Pappas PG, Karchmer AW, Sobel J, Edwards JE, Hadleyt S, et-al. A randomized and blinded multicenter trial of high-dose fluconazol plus placebo versus fluconazole plus amphotericin B as therapy for Candidemia and its consequences in non-neutropenic subjects. Clin Infect Dis. 2003; 36:1221-8.
27.Gómez J, Baños C, Simarro E, Ruiz J, Requena L, Pérez J, et-al. Fungemias nosocomiales en un hospital general: Epidemiología y factores pronóstico 1993–1998. Enf Inf Microbiol Clin. 2001; 19:304-7.
28.Gómez J, Ortega N, Herrero J.A. Bases uso racional de antifúngicos. Rev Esp Quimioterap. 2004; 17:241-6.
29.Pappas PG, Kauffman CA, Andes D, Benjamin DK, Calandra TF, Edwards JE, et-al. Infectious Diseases Society of America. Clinical Practice Guidelines for the Management of Candidiasis: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2009; 48:503-35.
30.Wenzel RP, Gennings C.H. Bloodstream infections due Candida species in the Intensive Care Unit: Identifying especially high-risk patients to determine prevention strategies. Clin Infect Dis. 2005; 41:S389-93.
31.Schelenz S, Garnsden W.R. Candidaemia in a London teaching hospital: analysis of 128 cases over a 7-year period. Mycoses. 2003; 46:390-6.