Risk factors for candidaemia in critically ill patients: a prospective surveillance study

Candidaemia is frequently a life-threatening complication in patients admitted to the intensive care unit (ICU). To assess the risk factors for candidaemia in critically ill patients with prolonged ICU stay, a total of 1765 adult patients admitted for at least 7 days to 73 medical-surgical ICUs of 70 tertiary care hospitals in Spain participated in a prospective cohort study. Candidaemia was defined as recovery of Candida spp. from blood culture. Sixty-eight episodes of candidaemia occurred in 63 patients, representing 35.7 episodes per 1000 ICU patients admitted, with an incidence rate of 1.5 episodes per 1000 days of ICU stay. Causative fungi were C. albicans in 57.1% of cases and non-albicans Candida spp. in 42.9%. In the multivariate analysis, independent factors significantly associated with candidaemia were Candida colonisation (OR = 4.12, 95% CI: 1.82-9.33), total parenteral nutrition (OR = 3.89, 95% CI: 1.73-8.78), elective surgery (OR = 2.75, 95% CI: 1.17-6.45) and haemofiltration procedures (OR = 1.96, 95% CI: 1.06-3.62). In the ICU setting in Spain and in patients who have stayed in units for >7 days, more than half of cases of candidaemia were caused by C. albicans. Risk factors for candidaemia identified included Candida colonisation, elective surgery, total parenteral nutrition and haemodialysis.     

Neonatal candidaemia in Kuwait: a 12-year study of risk factors, species spectrum and antifungal susceptibility

A study of candidaemia in neonatal intensive care unit (NICU) over a 12-year period (1995–2006) taking into consideration demographic variables, risk factors, aetiological Candida species and therapeutic outcomes is presented. The yeast isolates were identified by VITEK2 yeast identification system and antifungal susceptibility was determined by E-test. Of 4815 neonates admitted in NICU, 182 cases of candidaemia were detected with an overall prevalence of 4.0% and crude mortality of 27.7%. The annual rate of candidaemia per 1000 admissions was the highest in 1997 (84 cases) and the lowest in 2004 (10 cases). Of the 112 assessable candidaemia cases, 78 (70%) occurred in very low birth weight neonates (≤1500 g), 65 (58%) were born with gestational age of ≤30 weeks. The main identifiable risk factors were use of ≥2 antibiotics (87%), total parenteral nutrition for >5 days (82%), placement of central venous catheter (78%) and prior colonisation with Candida species (54%). Candida albicans and non- albicans Candida species accounted for 43% and 57% of candidaemia cases, respectively, and C. parapsilosis emerged as a predominant species. No fluconazole resistance was observed in C. albicans and C. parapsilosis isolates. This is the first comprehensive study on the epidemiology of neonatal candidiasis in Kuwait.     

Duration and outcome of persistent candidaemia in newborn infants

The aims of the study were to define the duration of candidaemia in newborn infants and to determine the incidence of persistent hospital-acquired candidaemia and its associated morbidity and mortality compared with non-persistent candidaemia. This retrospective study, included 56 neonates admitted to the neonatal intensive care unit from 1996 to 2000 who had one or more positive blood cultures for Candida spp. The most prevalent pathogen was Candida albicans (64.3%). Mean disease duration was 6.9 +/- 5.5 days (median 6 days). Twenty-nine patients (52%) had a positive blood culture for >5 days (persistent candidaemia). There were no statistically significant differences between the infants with persistent and non-persistent candidaemia in background or predisposing clinical factors. Fungal endocarditis was present in four patients (13.7%) with persistent disease and one patient (3.7%) with non-persistent disease (odds ratio 4.19), and uveitis developed in one patient. Ten patients (17.8%) died--five with persistent and five with non-persistent disease (P = 1.0). These findings indicate that persistent neonatal candidaemia may be associated with an increased risk of Candida endocarditis, but not with other complications or increased mortality.     

Nosocomial bloodstream infections associated with Candida species in a Turkish University Hospital

In recent years, a progressive increase in the frequency of nosocomial candidaemia has been observed, especially among the critically ill or immunocompromised patients. The aim of this study was to evaluate the trend in incidence of candidaemia together with potential risk factors in an 850-bed Turkish Tertiary Care Hospital in a 4-year period. A total of 104 candidaemia episodes were identified in 104 patients. The overall incidence was 0.56 per 1000 hospital admissions and the increase in incidence of candidaemia from 2000 to 2003 was found to be statistically significant (P = 0.010). Candida albicans was the most common species (57.7%) and non-albicans species accounted for 42.3% of all episodes. The most common non-albicans Candida sp. isolated was C. tropicalis (20.2%) followed by C. parapsilosis (12.5%). The most frequent risk factors possibly associated with the candidaemia were previous antibiotic treatment (76.9%), presence of central venous catheter (71.2%) and total parenteral nutrition (55.8%). Our results show the fact that the incidence of candidaemia caused by non-albicans species is frequent and increasing significantly, although the most common isolated Candida species were C. albicans and further investigations are necessary to evaluate the mechanisms of increasing incidence of candidaemia caused by non-albicans species.     

Nosocomial candidaemia in children: results of a 9-year study

The aim of this study was to determine changes in the incidence of nosocomial candidaemia and to evaluate the risk factors, demographic features, treatment and clinical outcome associated with candidaemia in a Turkish tertiary care paediatric unit within a 9-year period. The data of children who were diagnosed as nosocomial candidaemia, were examined in this study. Between January 1997 and December 2005, a total of 102 nosocomial candidaemia episodes were identified in 102 patients. The rate of nosocomial candidaemia in our clinic increased from 3.2 cases per 1000 admissions in 1997–1999, to 5.5 per 1000 admissions in 2000–2002 and to 6.9 per 1000 admissions in 2003–2005 ( P  = 0.003). The species most frequently causing candidaemia were Candida albicans (39.2%), Candida parapsilosis (21.6%) and Candida tropicalis (15.7%). The mortality of C. albicans (37.5%), was significantly higher than the mortality of non- albicans species (17.7%) ( P  = 0.04). Independent risk factors associated with candidaemia-related deaths by logistic regression analysis were disseminated candidiasis (odds ratio, 5.7; P  = 0.01), paediatric intensive care unit stay (odds ratio, 8.1; P  = 0.001), prolonged antibiotics therapy (odds ratio, 5.2; P  = 0.014), use of total parenteral nutrition (odds ratio, 4.4; P  = 0.038) and mechanical ventilation (odds ratio, 4.9; P  = 0.01). The rate of nosocomial candidaemia in our clinic increased >2-fold during the study period.     

Creation and assessment of a clinical predictive model for candidaemia in patients with candiduria

Candidaemia is the most common clinical presentation of invasive candidiasis and is a major cause of morbidity and mortality. Candiduria is a predictor for candidaemia; however, patient characteristics that are associated with concurrent candidaemia in the setting of candiduria are unclear. Identifying these characteristics could aid in the early detection of systemic disease. We performed a retrospective cohort analysis of hospitalised patients with candiduria at our institution over a 13‐year period. Our evaluation of patient characteristics included demographics, comorbidities, medications, procedures, devices, vital signs and laboratory values. We developed a multivariable logistic model to identify factors associated with candidaemia in patients with candiduria. We identified 4240 patients with candiduria, 263 (6.2%) of whom had candidaemia. Predictors for increased risk of candidaemia with candiduria included hospitalisations > 12 days, central venous catheter, parenteral nutrition, haematological and gynaecological malignancy, and receipt of β‐lactam/β‐lactamase inhibitors. Vital signs and laboratory values associated with candidaemia included elevated heart rate, temperature and creatinine, along with neutropenia and neutrophilia. Factors that demonstrated a decreased risk of candidaemia included diabetes mellitus, gastrostomy and urinary catheter with antibiotic use. The c‐statistic was 0.741 (95% CI, 0.710‐0.772). We identified a set of clinical characteristics that can predict the presence of candidaemia with candiduria.     

Differences in patient risk factors and source of candidaemia caused by Candida albicans and Candida glabrata

There is an increasing frequency of candidaemia caused by Candida glabrata which has decreased in vitro susceptibility to fluconazole. Differences in risk factors for candidaemia caused by C. glabrata and C. albicans have not been formally evaluated in a diverse patient group. We performed a retrospective study of adult inpatients from January 1, 2003 to April 30, 2008 with C. glabrata and C. albicans candidaemia at a single tertiary care centre in Detroit, Michigan to evaluate for differences in risk factors and presumed source of infection in these groups. Patients’ underlying conditions, risk factors and source of infection (probable or definite) were compared. Among 119 patients, 80 (67.2%) were C. albicans and 39 (32.8%) C. glabrata. Using logistic regression analysis, patients with C. glabrata infection were more likely to have diabetes mellitus (OR 2.43; 95% CI, 1.06–5.54) and abdominal source of infection (OR 4.53, 95% CI, 1.72–11.92). Mortality rates in the two groups were similar. Patients with C. glabrata candidaemia are more likely to be diabetic and have an abdominal source of infection compared with patients with C. albicans.     

Epidemiology and prognosis of candidaemia in elderly patients

The aim of the study was to analyse the epidemiology and prognosis of candidaemia in elderly patients. We performed a comparison of clinical presentation of candidaemia according to age and a study of hazard factors within a prospective programme performed in 29 hospitals. One hundred and seventy‐six episodes occurred in elderly patients (>75 years), 227 episodes in middle‐aged patients (61‐75 years) and 232 episodes in younger patients (16‐60 years). Central venous catheter, parenteral nutrition, neutropenia, immunosuppressive therapy and candidaemia caused by Candida parapsilosis were less frequent in elderly patients. These patients received inadequate antifungal therapy (57.3%) more frequently than middle‐aged and younger patients (40.5% P < .001). Mortality during the first week (20%) and 30 days (42%) was higher in elderly patients. The variables independently associated with mortality in elderly patients during the first 7 days were acute renal failure (OR: 2.64), Pitt score (OR: 1.57) and appropriate antifungal therapy (OR: 0.132). Primary candidaemia (OR: 2.93), acute renal failure (OR: 3.68), Pitt score (OR: 1.38), appropriate antifungal therapy (OR: 0.3) and early removal of the central catheter (OR: 0.47) were independently associated with 30‐day mortality.In conclussion, inadequate antifungal treatment is frequently prescribed to elderly patients with candidaemia and is related with early and late mortality.     

Candidaemia in a London teaching hospital: analysis of 128 cases over a 7-year period

In a retrospective analysis of 128 cases of Candida bloodstream infections in a London teaching hospital between 1995 and 2001, the incidence of candidaemia increased from 0.2/1000 admissions in 1995 to 0.5 and 0.4/1000 admissions in 2000 and 2001, respectively. Risk factors for candidaemia included the presence of intravascular (IV) lines (88%), admission to intensive care (51%), parenteral nutrition (35%), multiple antibiotics (74%), corticosteroid therapy (12%), cancer chemotherapy (11%), renal transplantation (5%) and neutropenia (3%). The sources of infection were IV lines (77%), the urinary tract (7%) and the gastrointestinal tract (7%). Serious infective complications (endocarditis, endophthalmitis or brain abscess) were noted in 6% of cases. The most frequently isolated species were Candida albicans (64%), C. glabrata (20%), C. tropicalis (9%) and C. parapsilosis (5%). The overall fluconazole-resistance rate of Candida spp. was 7% (MIC > or = 64 mg l-1). All the C. albicans isolates were sensitive to fluconazole (MIC < or = 8 mg l-1) whereas 20% of non-C. albicans isolates (27% of C. glabrata and 14%C. tropicalis) were resistant. The mortality rate (35%) was lower than in other reports and may be due to the early recognition of candidaemia and the prompt removal of IV lines together with the initiation of appropriate antifungal therapy. Regular surveillance of local Candida species, resistance profiles and risk factors is important in order to identify patients at risk and to develop empirical treatment protocols to reduce the incidence and mortality of candidaemia.     

Risk factors and outcome for breakthrough candidaemia in patients with cancer

While candidaemia that develops during systemic antifungal prophylaxis or therapy [breakthrough (BT) candidaemia] has been reported increasingly, the risk factors and outcome of BT candidaemia are not yet known definitely. We evaluated a consecutive series of cancer patients with candidaemia at Asan Medical Center during 6 years to identify risk factors and outcome in BT candidaemia comparing with non-BT candidaemia. Fifty-four episodes of candidaemia in cancer patients diagnosed during this period were reviewed retrospectively. There were 10 episodes (18.5%) of BT candidaemia in which the species distribution and frequency of catheter-related infection in the BT and non-BT groups were similar. BT candidaemia had a tendency to occur more frequently in patients with haematological than non-haematological diseases. Profound neutropenia and disseminated candidiasis were more common in the BT group; however, these differences were not statistically significant (P = 0.17 and 0.07 respectively). The duration of prior antibiotic therapy and duration of profound neutropenia (<100 mm(-3)) were identified as risk factors for BT candidaemia (P < 0.01 and 0.02 respectively) in univariate analysis and the latter was the only risk factor in multivariate analysis. The death rates due to candidaemia were 85.7% in the BT group and 42.9% in the non-BT group (P = 0.08); however, BT candidaemia alone did not increase mortality rate.     

Epidemiology,species distribution,clinical characteristics and mortality of candidaemia in a tertiary care university hospital in Turkey, 2007‐2014

Candidaemia still continues to be a serious medical concern and the epidemiology of candidaemia varies according to geographical areas. We aim to determine the incidence, local epidemiology, Candida species distribution and crude mortality rates of candidaemia. We retrospectively evaluated candidaemia episodes in between January 2007 and August 2014. We compared demographic, clinical, microbiological findings and mortality rates of episodes caused by Candida albicans and non‐albicans Candida species. Overall the candidaemia incidences were 1.23 episodes/1000 admissions. A significant negative slope among candidaemia episodes and years was determined. Overall C. albicans (54.6%) was the most common species followed by Candida glabrata, Candida tropicalis and Candida parapsilosis respectively. Preinfection hospital stay and length of hospital stay were statistically longer in patients with non‐albicans Candida candidaemia than in patients with C. albicans candidaemia. The source of candidaemia was unknown in 52.5% of all episodes. Central venous catheters among non‐albicans Candida candidaemia episodes and urinary system among C. albicans candidaemia episodes were common source of candidaemia compared to each other. Previous antifungal therapy preceding candidaemia and concomitant bacteraemia were significantly associated with non‐albicans Candida candidaemia. Continuous local surveillance will preserve its pivotal importance in formulating empirical antifungal therapy and improving management of candidaemia.     

Risk factors,clinical presentation and prognosis of mixed candidaemia: a population‐based surveillance in Spain

The low incidence of mixed candidaemia (MC) may have precluded a better knowledge of its clinical presentation. The aim of the study was to analyse the risk factors, clinical presentation and prognosis of MC episodes. A comparison between MC and monomicrobial candidaemia within a prospective programme on candidaemia was performed in 29 hospitals between April 2010 and May 2011. In fifteen episodes of candidaemia corresponding to 15 patients, out of 752, two species of Candida (1.9%) were isolated. MC was more frequent in patients with HIV infection (12%, P = 0.038) and those admitted due to extensive burns (23%, P = 0.012). The Candida species most frequently identified in MC were C. albicans 12 patients (40%), C. glabrata seven patients (23.3%) and C. parapsilosis six patients (20%). Early mortality was higher (nine patients, 60%) in patients with MC than in patients with MMC (223 patients, 30.3%, P = 0.046). In conclusion, MC was was independently associated with increased mortality even after considering other prognostic factors. MC is an infrequent event that is more common in HIV infection and in patients suffering from burns, and is associated with increased mortality.     

Species distribution and antifungal susceptibility of blood Candida isolates at a tertiary hospital in southern Taiwan, 1999–2006

The aim of this study was to evaluate the incidence of candidaemia, consumption of fluconazole and susceptibility of blood Candida isolates at a tertiary hospital. From January 1999 to September 2006, all candidaemic episodes were identified and available strains were evaluated for the susceptibilities of antifungal agents. Annual defined daily doses of antifungal agents were collected. There had been 909 Candida isolates detected from the bloodstream of 843 patients during the study period. Among them, 740 isolates were available for the susceptibilities of antifungal agents. The incidence density of candidaemia was 28 episodes per 10 000 patient‐days. Species distribution of 909 isolates did not vary annually, but varied greatly in the units of the hospital. Candida parapsilosis was the more prominent (30.1%) isolate in the paediatric units, where C. tropicalis and C. glabrata were less common (12.3% and 1.4% respectively). Resistance rates for itraconazole, fluconazole and voriconazole were 6.9%, 3.8% and 3.8% respectively. There were 25 (3.4%) isolates resistant to amphotericin‐B. Although fluconazole usage increased over time (r² = 0.45; P = 0.07), fluconazole resistance did not increase accordingly (P = 0.33). In our institution in which the incidence of candidaemia was high, fluconazole resistance among blood Candida isolates remained rare.     

An open‐label study of anidulafungin for the treatment of candidaemia/invasive candidiasis in Latin America

Incidence and mortality of candidaemia/invasive candidiasis (C/IC) is relatively high in Latin America versus North America and Europe. To assess efficacy and safety of intravenous (IV) anidulafungin in Latin American adults with documented C/IC. All patients in this open‐label study received initial IV anidulafungin with optional step‐down to oral voriconazole after 5 days; total treatment duration was 14–42 days. The primary endpoint was global response (clinical + microbiological response) at end of treatment (EOT); missing/indeterminate responses were failures. The study enrolled 54 patients; 44 had confirmed C/IC within 96 h before study entry and comprised the modified intent‐to‐treat population. Global response at EOT was 59.1% (95% CI: 44.6, 73.6), with 13 missing/indeterminate assessments. Thirty‐day all‐cause mortality was 43.1%. Fourteen patients (31.8%) were able to step‐down to oral voriconazole; these patients had lower baseline acute physiological assessment and chronic health evaluation (APACHE) II scores and were less likely to have solid tumours or previous abdominal surgery. Anidulafungin was generally well tolerated with few treatment‐related adverse events. Anidulafungin was associated with relatively low response rates influenced by a high rate of missing/indeterminate assessments and mortality comparable to other recent candidaemia studies in Latin America. In a subset of patients with lower APACHE II scores, short‐course anidulafungin followed by oral voriconazole was successful.     

Candidaemia in a paediatric centre and importance of central venous catheter removal

The aim of this study is to identify differences in distribution of Candida species, resistance to antifungals and clinical outcome, as well as the identification of potential risk factors associated with candidaemia in children. We conducted a retrospective analysis in children ≤18 years with blood culture proven candidaemia identified between 2004 and 2012. Patients were divided into two groups (Group 1, <3 months, n = 51; Group 2, ≥3 months, n = 197) to identify any potential difference between the neonatal and early infantile periods in terms of risk factors and distribution of Candida species. A total of 248 distinct episodes of candidaemia were identified over the study period. The most frequently isolated Candida species were C. albicans (53.2%), followed by C. parapsilosis (26.2%), C. tropicalis (8.1%). Of the 248 episodes, 71 episodes (28.6%) resulted in death within 30 days from the onset of candidaemia. In Group 1, failure of central venous catheter (CVC) removal was found to be associated with a 20.5‐fold increase in mortality [95% CI (3.9, 106.5); P < 0.001], compared to a 5.9‐fold increased risk with hypoalbuminaemia [95% CI (1.03, 34.1); P = 0.046]. For Group 2, the increased risk was 23‐fold for failure of CVC removal [95% CI (7.48, 70.77); P < 0.001], 7.4‐fold for mechanical ventilation [95% CI (2.64, 21.08); P < 0.001], 4.4‐fold for hypoalbuminaemia [95% CI (1.56, 12.56); P = 0.005], 3.1‐fold for neutropaenia [95% CI (1.31, 7.69); P = 0.010] and 2.2‐fold for male gender [95% CI (1.02, 4.71); P = 0.043]. Therapeutic choices should be guided by sound knowledge of local epidemiological trends in candidaemia. Removal of CVC significantly reduces mortality and is an essential step in the management of candidaemia.     

Acute myeloid leukemia in the elderly: etoposide when anthracyclins are contraindicated

We investigated whether etoposide might be a suitable alternative to anthracyclins for the treatment of elderly patients with acute myeloid leukemia (AML) in whom anthracyclins are contraindicated. In total, 88 patients over 60 years old were treated in our department between 1988 and 2000 for de novo or secondary AML. Of these 88 patients, 21 had severe cardiac disease and received a combination of etoposide and cytosine arabinoside (group A). They were compared with 23 patients who received daunorubicin plus cytosine arabinoside (group B) and 44 patients who received mitoxantrone, cytosine arabinoside and etoposide (group C). In group A, the complete remission (CR) rate was 33% (7/21), median disease-free survival (DFS) was 365 days and median overall survival (OS) was 214 days. These results were not statistically different from those for the anthracyclin-based regimens (CR 43%, DFS 210 days, OS 150 days for group B and CR 47%, DFS 216, OS 295 days for group C). Moreover, there was no significant difference in the number of treatment-related toxic events and deaths in the 3 treatment groups. Etoposide combined with a standard dose of cytosine arabinoside is relatively safe and effective in elderly patients with AML and offers an alternative approach to patients with cardiac contraindications to anthracyclin treatment.     

Incidence and predictors of invasive candidiasis associated with candidaemia in children

Risk factors for invasive candidiasis in children with candidaemia are poorly defined. We performed a retrospective cohort study of all children with candidaemia at our tertiary children’s hospital from 2000 to 2006. Invasive candidiasis was diagnosed by review of the medical record and standardised EORTC/MSG criteria. A variety of risk factors for invasive candidiasis were explored. Of 194 episodes of candidaemia in the microbiology laboratory database, 180 clinical records were available. Evaluation for invasive candidiasis consisted of 174 (97%) echocardiograms, 167 (93%) dilated ophthalmological examinations, 136 (76%) chest CT scans and 108 (60%) abdominal ultrasounds (complete, hepatosplenic or renal). Of the 180 patients, 15 (8%) were identified with invasive candidiasis (4 proven, 1 probable, 10 possible). Prematurity <32 weeks (P < 0.01), an underlying immunocompromising disorder (P < 0.01), and ≥2 days of candidaemia (P = 0.05) were significantly associated with invasive candidiasis. Invasive candidiasis, especially proven or probable, in the setting of candidaemia was not common in our hospital, but premature infants and immunocompromised children were at significantly higher risk. Based on our findings, extensive imaging and examination by an ophthalmologist were particularly low‐yield for invasive candidiasis in immunocompetent children beyond infancy.     

Anidulafungin for the treatment of candidaemia caused by Candida parapsilosis: Analysis of pooled data from six prospective clinical studies

Concerns with echinocandin use for infections caused by Candida parapsilosis complex species have driven the need for data to support echinocandin clinical efficacy in such patients. Data from six prospective studies were pooled to assess efficacy and safety of anidulafungin in patients with candidaemia caused by C. parapsilosis. Patient‐level data were pooled from patients with microbiologically confirmed candidaemia due to C. parapsilosis treated with anidulafungin. Patients received a 200 mg intravenous (IV) loading dose of anidulafungin (day 1) and 100 mg daily thereafter. IV treatment could be switched to oral azole therapy after ≥5 or ≥10 days. Primary endpoint was global response at end of IV therapy (EOIVT). Seventy patients had candidaemia caused by C. parapsilosis. Global response was 77.1% (95% CI: 67.3, 87.0) at EOIVT and 70.0% (95% CI: 59.3, 80.7) at end of treatment. Three of 55 isolates (with MICs available) were resistant to anidulafungin (MIC ≥8 mg/L). All‐cause mortality was 5.7% (n=4/70) by day 14 and 14.3% (n=10/70) by day 28. IV anidulafungin was effective for the treatment of C. parapsilosis candidaemia in this population, consistent with efficacy previously demonstrated for other Candida species. (ClinicalTrials.gov identifiers: NCT00496197, NCT00548262, NCT00537329, NCT00689338, NCT00806351, NCT00805740).     

Comparative effectiveness of fungicidal vs. fungistatic therapies for the treatment of paediatric candidaemia

Adult data suggest that echinocandins for treatment of candidaemia are associated with decreased mortality, attributed to their fungicidal activity. There are limited data comparing antifungals in children. We compared 30‐day all‐cause mortality among paediatric candidaemia patients treated with fungicidal vs. fungistatic agents. All inpatients (>6 months and <19 years of age) with candidaemia between 2000 and 2012 at The Children's Hospital of Philadelphia were retrospectively identified. Definitive therapy with fungicidal (amphotericin B and caspofungin) agents was compared with fungistatic (fluconazole) agents. A propensity score model generated the inverse probability of receiving a fungicidal agent, which was included in a weighted logistic regression model. Among 203 children meeting inclusion criteria, 151 (74.4%) and 52 (25.6%) received a fungicidal and fungistatic agent, respectively. Overall, 18 (8.9%) patients died within 30 days. There was no statistically significant difference in mortality between patients started on a fungicidal or fungistatic agent (OR: 2.19, 95% CI: 0.42–11.48). In a propensity score–weighted model, definitive therapy with a fungicidal agent did not result in a significant decrease in mortality. These data suggest that both agents can be considered definitive therapy for paediatric candidaemia. The results should be interpreted with caution given the small sample size. Larger cohort studies are needed.     

Candidaemia in an Irish tertiary referral hospital: epidemiology and prognostic factors

There were two parts to this study. Part 1 evaluated the epidemiology of Candida bloodstream isolates within the Southern Health Board (SHB) of Ireland from 1992 to 2003 by retrospective surveillance of all such isolates of patients reported from SHB hospitals to our laboratory database during that period. Part 2 reviewed candidaemia cases occurring in Cork University Hospital (CUH) from 1999 to 2003 using surveillance of all positive blood culture isolates in CUH microbiology laboratory during the 5-year period. In part 1, 250 Candida bloodstream isolates were reported in the SHB over 12 years. There was a pattern of decreasing percentage of C. albicans with time. Whereas in part 2, 63 cases of candidaemia were identified in CUH from 1999 to 2003. Candida albicans constituted 50% of all isolates, while C. parapsilosis and C. glabrata accounted for 21.2% and 18.2% respectively. Average annual incidence rate was 0.48 episodes/1000 admissions and 0.70 episodes/10 000 patient-days. Vascular catheters were the commonest source of candidaemia (61.9%) followed by the urinary tract (12.7%). Risk factors included exposure to multiple antibiotics (75%); central vascular catheterization (73%); multiple colonization sites (71%); severe gastrointestinal (GI) dysfunction (54%) and acute renal failure (43%). Crude 7-day and 30-day mortality rates were 20.6% and 39.7% respectively. Logistic regression multivariate analysis identified the following to be independent predictors for mortality: age > or =65 years [odds ratio (OR) 7.2, P = 0.013]; severe GI dysfunction (OR 10.6, P = 0.01); acute renal failure (OR 7.6, P = 0.022); recent/concurrent bacteraemia (OR 5.2, P = 0.042); endotracheal intubation (OR 7.7, P = 0.014); while major surgery was associated with a better prognosis (OR 0.05, P = 0.002). Appropriate antifungal treatment was also found to be associated with survival (Fisher's exact test, P < 0.001). The epidemiology of Candida bloodstream isolates within our health board had changed over the years. Incidence and mortality of candidaemia were relatively high in our hospital. Dysfunction of major organ systems and recent bacteraemia were found to predict mortality.