Invasive Fungal Infections In Neonates Research Articles

End-organ damage from neonatal invasive fungal infection: a 14-year retrospective study from a tertiary center in China

BackgroundInvasive fungal infection (IFI) has become an increasing problem in NICU neonates, and end-organ damage (EOD) from IFI is one of the leading causes of morbidity and mortality in neonates. This study was conducted to summarize clinical data on epidemiology, risk factors, causative pathogens, and clinical outcomes of IFI-associated EOD among neonates in a center in China for the sake of providing references for prevention and treatment of fungal infections in neonates in future.MethodsThe clinical data of IFI neonates who received treatment in a tertiary NICU of China from January 2009 to December 2022 were retrospectively analyzed, including causative pathogens and the incidence of EOD. The neonates were divided into EOD group and non-EOD (NEOD) group. The general characteristics, risk factors and clinical outcomes of the two groups were compared.ResultsIncluded in this study were 223 IFI neonates (137 male and 86 female) with a median gestational age (GA) of 30.71 (29,35) weeks and a median birth weight (BW) of 1470 (1120,2150) g. Of them, 79.4% were preterm infants and 50.2% were born at a GA of ≥ 28, <32 weeks, and 37.7% with BW of 1000–1499 g. Candida albicans (C. albicans) was the most common Candida spp. in these neonates, accounting for 41.3% of all cases, followed by C. parapsilosis (30.5%) and C. glabrata (7.2%). EOD occurred in 40 (17.9%) of the 223 cases. Fungal meningitis was the most common EOD, accounting for 13.5% of the 40 EOD cases. There was no significant difference in the premature birth rate, delivery mode, GA and BW between EOD and NEOD groups, but the proportion of male infants with EOD was higher than that without. There was no significant difference in antenatal corticosteroid use, endotracheal intubation, invasive procedures, use of antibiotics, total parenteral nutrition, blood transfusion, postnatal corticosteroid use, fungal prophylaxis and the incidence of necrotizing enterocolitis between the two groups, but the proportion of C. albicans infection cases in EOD group was higher than that in NEOD group (57.5% vs. 37.7%). Compared with NEOD group, the proportion of cured or improved infants in EOD group was significantly lower (P < 0.05), and the number of infants who died or withdrew from treatment was larger (P < 0.05).ConclusionsOur retrospective study showed that preterm infants were prone to fungal infection, especially very preterm infants. C. albicans was the most common Candida spp. for IFI, and was a high-risk factor for EOD. EOD can occur in both full-term and premature infants, so the possibility of EOD should be considered in all infants with IFI.

Updates in Laboratory Identification of Invasive Fungal Infection in Neonates.

Invasive fungal infection (IFI) in immunocompromised neonates is significantly associated with high morbidity and mortality and has become the third most common infection in Neonatal Intensive Care Units. The early diagnosis of IFI for neonatal patients is difficult because of the lack of specific symptoms. The traditional blood culture remains the gold standard in clinical diagnosis for neonatal patients but it requires a long duration, which delays treatment initiation. Detections of fungal cell-wall components are developed for early diagnosis but the diagnostic accuracy in neonates needs to be improved. PCR-based laboratory methods, such as real-time PCR, droplet digital PCR, and the cationic conjugated polymer fluorescence resonance energy transfer (CCP-FRET) system, distinguish the infected fungal species by their specific nucleic acids and show a high sensitivity and specificity. Particularly, the CCP-FRET system, which contains a cationic conjugated polymer (CCP) fluorescent probe and pathogen-specific DNA labeled with fluorescent dyes, could identify multiple infections simultaneously. In the CCP-FRET system, the CCP and fungal DNA fragments can self-assemble into a complex with an electrostatic interaction and the CCP triggers the FRET effect under ultraviolet light to make the infection visible. Here, we summarize the recent laboratory methods for neonatal IFI identification and provide a new perspective for early clinical fungal diagnosis.

Prophylaxis of Invasive Fungal Infection in Neonates: A Narrative Review for Practical Purposes

Candida albicans is the most frequent cause of invasive fungal disease in preterm and/or low birth weight neonates, followed by Candida parapsilosis, whilst infections from other species are rare. Considering the severity of the disease, associated with poor clinical signs and diagnostic difficulties, primary prophylaxis becomes relevant. This paper summarizes the pathogenesis and clinical presentation of invasive candidiasis in neonates, focusing on prophylaxis. For late onset invasive disease, e.g., those occurring after the 3rd (or 7th according to some definitions) day of life possible approaches are the use of fluconazole, recommended in case of weight <1000 g or <1500 g if the local incidence of invasive candidiasis is higher than 2%, or the use of nystatin (for patients < 1500 g). Micafungin must be used in case of colonization by Candida auris, or in centers with a high prevalence of this pathogen. Concurrently, correct management of the central venous catheter and isolation procedures, with special regard to patients colonized by resistant strains, are fundamental. Other approaches such as reduced use of H2 blockers and broad-spectrum antibiotics (e.g., 3rd generation cephalosporins or carbapenems) and promotion of breast feeding proved useful. Reduction of early-onset infections (those occurring in the first 3 days of life) can also be obtained by treating maternal vulvo-vaginal candidiasis, which can represent a fastidious problem during pregnancy. In this case, topic azoles (the only recommendable treatment) can represent a kind of "prophylaxis" of early neonatal candidiasis. However, it must always be remembered that prophylaxis reduces the risk of invasive candidiasis but can not completely eliminate its occurrence, with the parallel risk of selecting for antifungal-resistant strains. Clinicians must maintain a high level of suspicion to start an appropriate therapy and strict epidemiological surveillance to identify the occurrence of clusters and the appearance of strains resistant to prophylaxis.

Incidence and risk factors of bacterial sepsis and invasive fungal infection in neonates and infants requiring major surgery: an Italian multicentre prospective study

Limited data are currently available on the incidence rates and risk factors for bacterial sepsis and invasive fungal infections (IFIs) among neonates and infants undergoing major surgery. To assess the incidence of bacterial sepsis and IFI, fungal colonization, risk factors for sepsis, and mortality in neonates and infants aged <3 months undergoing major surgery. A multicentre prospective study was conducted involving 13 level-3 neonatal intensive care units in Italy, enrolling all infants aged ≤3 months undergoing major surgery. From 2018 to 2021, 541 patients were enrolled. During hospitalization, 248 patients had a bacterial infection, and 23 patients had a fungal infection. Eighty-four patients were colonized by fungal strains. Overall, in-hospital mortality was 2.8%, but this was higher in infected than in uninfected infants (P= 0.034). In multivariate analysis, antibiotic exposure before surgery, ultrasound-guided or surgical placement of vascular catheters, vascular catheterization duration, and gestational age ≤28 weeks were all associated with bacterial sepsis. The risk of IFI was markedly higher in colonized infants (odds ratio (OR): 8.20; P < 0.001) and was linearly associated with the duration of vascular catheterization. Fungal colonization in infants with abdominal surgery increased the probability of IFI 11-fold (OR: 11.1; P < 0.001). Preventive strategies such as early removal of vascular catheters and the fluconazole prophylaxis should be considered to prevent bacterial and fungal sepsis in infants undergoing abdominal surgery, and even more so in those with fungal colonization.

Invasive fungal infections in neonates: a review.

Invasive fungal infections remain the leading causes of morbidity and mortality in neonates, especially preterm and very low birth weight infants. Most invasive fungal infections are due to Candida or Aspergillus species, and other fungi are increasingly reported and described. Appropriate identification and treatment are required to augment activity and reduce the toxicity of antifungal drugs. Successful use of antifungals in the vulnerable neonatal population is important for both prevention and treatment of infection. Strategies for prevention, including prophylactic antifungal therapy as well as reducing exposure to modifiable risk factors, like limiting antibiotic exposure, discontinuation of central catheters, and hand hygiene are key techniques to prevent and decrease rates of invasive fungal infections. In conclusion, this is a review of the most common causes, prevention strategies, prophylaxis, and treatment of invasive fungal infections in neonates.

Use of Antifungals Other Than Amphotericin B for Invasive Fungal Infections in Neonates and Children

Use of Antifungals Other Than Amphotericin B for Invasive Fungal Infections in Neonates and Children

Simultaneous detection of fungal (1,3)-β-d-glucan and procalcitonin using a dual-label time-resolved fluorescence immunoassay.

Neonatal infectious diseases are a serious threat to the health of newborns. The aim was to establish a new detection method for the simultaneous measurement of (1,3)-β-d-glucan and procalcitonin in serum for the early screening and efficacy testing of neonatal infectious diseases. We established a sandwich dual-label time-resolved fluorescence immunoassay (TRFIA): anti-(1,3)-β-d-glucan/procalcitonin antibodies immobilized on 96-well plates captured (1,3)-β-d-glucan/procalcitonin antigens and then banded together with the detection antibodies labeled with europium(III) (Eu3+ )/samarium(III) (Sm3+ ) chelates. Finally, time-resolved fluorometry was used to measure the fluorescence intensity. The linear correlation coefficient (R2 ) of the (1,3)-β-d-glucan standard curve was 0.9913, and the R2 of the procalcitonin standard curve was 0.9911. The detection sensitivity for (1,3)-β-d-glucan was 0.4pg/mL (dynamic range: 0.6-90pg/mL), and the average recovery was 101.55%. The detection sensitivity for procalcitonin was 0.02ng/mL (dynamic range: 0.05-95ng/mL), and the average recovery was 104.61%. There was a high R2 between the present TRFIA method and a commercially available assay (R2 =0.9829 for (1,3)-β-d-glucan and R2 =0.9704 for procalcitonin). Additionally, the cutoff values for (1,3)-β-d-glucan and procalcitonin were 23.95pg/mL and 0.055ng/mL, respectively. The present TRFIA method has high sensitivity, accuracy, and specificity and is an effective method for early screening and efficient testing of neonatal invasive fungal infection.

1,3 β-D-Glucan versus blood culture for the diagnosis of invasive fungal infection in formula-fed preterm neonates

Abstract Background 1,3 β-D-Glucan is an important marker for diagnosis of invasive fungal infection (IFI).Its value in comparison to blood culture is not yet established in neonates. Aim primary aim is to study the value of 1,3 β-D-Glucan in comparison with culture on Sabouraud dextrose agar in diagnosis of IFI, Secondary aim is to study the prevalence of IFI in preterm neonates. Subjects and methods a prospective observational study carried on 75 non-exclusively breast fed preterm neonates ≤34 weeks gestational age. Tollner and Rodwell sepsis scores, SNAPII and blood culture were done to all neonates initially and on day 10&amp;17 of life. 1,3 β-D-Glucan was done for all neonates on day 10 of life. Results fungal growth was found in 9.3% of included preterm neonates.1,3β-D-Glucan was higher in culture proven fungal sepsis compared to culture proven bacterial sepsis and culture negative preterm neonates (83.28±26.22 pg\ µl, 51.42±16.71 pg\ µl, 43.07±12.87 pg\ µl respectively). 1,3 β-D-Glucan was related to Tollner (r = 0.286,p=0.013), Rodwell (r = 0.317,p=0.006) sepsis scores and to SNAPII score (r = 0.331,p=0.04). Its cut off point for diagnosis of IFI is 79pg\µl with sensitivity of 71% and specificity of 98%. Conclusion 1,3β-D-Glucan is an easy, rapid and specific marker for diagnosis of IFI in neonates. Further studies are needed to evaluate it as a therapeutic monitoring for IFI.

Review of Amphotericin B for Invasive Fungal Infections in Neonates and Children

Review of Amphotericin B for Invasive Fungal Infections in Neonates and Children

Clinical value of droplet digital PCR in rapid diagnosis of invasive fungal infection in neonates

To evaluate the clinical value of droplet digital PCR (ddPCR) in rapid and accurate diagnosis of invasive fungal infection (IFI) in neonates. The highly conserved sequence of fungi 18S RNA was selected as the target sequence, and primers were designed to establish a ddPCR fungal detection system. Blood samples were collected from 83 neonates with high-risk factors for IFI and/or related clinical symptoms in the neonatal intensive care unit (NICU) of a hospital in Shenzhen, China. Blood culture and ddPCR were used for fungal detection. The ddPCR fungal detection system had a specificity of 100% and a sensitivity of 3.2 copies/μL, and had a good reproducibility. Among the 22 blood samples from neonates with a confirmed or clinical diagnosis of IFI, 19 were detected positive by ddPCR. Among the 61 blood samples from neonates who were suspected of IFI or had no IFI, 2 were detected positive by ddPCR. The ddPCR technique can be used for the detection of neonatal IFI and is a promising tool for the screening and even diagnosis of neonatal IFI.

Invasive Fungal Infections in Neonates in Canada: Epidemiology and Outcomes.

Neonatal fungemia is associated with adverse neonatal outcomes and higher overall healthcare expenditure. Our objective is to review the epidemiology of invasive fungal infections (IFIs) in neonates in Canada. A retrospective cohort study using data collected by the Canadian Neonatal Network (CNN) was conducted. Using a nested matched cohort study design, risk factors and outcomes of neonates born <33 weeks gestation (n = 39,305) during 2003-2013 were compared between neonates diagnosed with an IFI during their stay to infection-free controls. Overall incidence of IFI among all admitted neonates was 0.22% (n = 286), while the incidence of IFI in the group of neonates born <33 weeks gestation was 0.64%. Of the isolates, 170 (59%) had Candida albicans and 59 (21%) had Candida parapsilosis. Risk factors for IFI were lower gestation, male sex, Apgar score <7 at 5 minutes, higher severity of illness score, maternal diabetes and vaginal birth. Neonates with IFI had higher odds of mortality [adjusted odds ratio (aOR): 1.60; 95% confidence interval (CI): 1.06-2.43], necrotizing enterocolitis (aOR: 2.97; 95% CI: 1.76-5.01) and severe retinopathy of prematurity (aOR: 2.15; 95% CI: 1.26-3.67). The overall incidence of IFI in neonates was low in Canada in comparison to other large population cohort studies; however, the mortality and morbidity remained high.

Nationwide survey of neonatal invasive fungal infection in Japan.

Invasive fungal infection (IFI) is a life-threating infectious disease in high-risk neonates. Strategies for the treatment and prevention of IFI in neonates in Japan remain unclear. We conducted a nationwide retrospective survey to determine IFI incidence between January 2014 and October 2015. Primary survey questionnaires were submitted to 309 medical facilities that regularly treat high-risk neonates. The questionnaire assessed IFI incidence during the study period, methods for preventing fungal infection in early delivery neonates, and methods for preventing mother-to-child fungal transmission. The secondary questionnaire was for facilities that had IFI cases and replied to the primary questionnaire. In total, 128 medical facilities (41.4%) completed the primary questionnaire, 17/128 facilities recorded 23 proven or probable IFI cases. Estimated annual IFI incidence was 0.33/1000 live births of hospitalized neonates. Patient data at IFI onset were available for all 23 patients. Birth weight was <1000g in 18 patients. Causative microorganisms were identified in 22 patients. Candida species (n = 21) were the most common pathogens, and one patient had mucormycosis. The mortality rate was 17.4%. Regarding neonatal fungal prophylaxis, 55/128 facilities (43.0%) reported administering therapy. The most frequently used prophylactic drugs were fluconazole, then micafungin. Fungal prophylaxis for mothers who showed fungal colonization was performed in 30/128 facilities (23.4%). Oxiconazole vaginal tablets were most commonly used as prophylaxis for high-risk mothers. In Japan, the diagnosis, treatment, and prevention of neonatal IFI varied. Continuous surveillance and treatment regimen for neonatal IFI are required to improve outcomes in high-risk neonates.

Epidemiological changes in invasive fungal infection in a neonatal intensive care unit

Objective To identify the epidemiological changes in invasive fungal infection (IFI) in a neonatal intensive care unit (NICU) to provide information for prevention and treatment of IFI. Methods A total of 102 cases who were diagnosed with IFI among 42 187 neonates hospitalized in the NICU of Affiliated BaYi Children's Hospital, Chinese People's Liberation Army General Hospital from January 1, 2009 to December 31, 2014 were enrolled in this study. Since January 1, 2012, the divisions of our NICU were more specific and intravenous fluconazole was administered as a routine preventive measure for high-risk infants. Clinical information of the IFI cases including general features, incidence, distribution of pathogens and drug (Amphotericin B, Fluconazole, Flucytosin, Itraconazole and Voriconazole) sensitivity were analyzed between former period (January 1, 2009 to December 31, 2011) and latter period (January 1, 2012 to December 31, 2014) by Chi-square test. Results The total incidence of IFI was 2.42‰ (102/42 187), and among the 102 IFI cases, 73.5% (75/102) were preterm infants and 75.5% (77/102) were low birth weight infants. The incidence of IFI in the latter period was lower than that in the former period [1.8‰ (48/26 046) vs 3.3‰ (54/16 141), χ2=9.329, P<0.01]. The incidences of IFI in neonates with gestation age <28, ≥28-<32 and ≥32-<37 weeks in latter period were decreased as compared with those in former period [10.6‰ (3/284) vs 76.9‰ (9/117), χ2=12.569; 6.1‰ (13/2 134) vs 21.9‰ (28/1 277), χ2=16.868; 1.4‰ (12/8 706) vs 1.9‰ (10/5 256), χ2=7.165] (all P<0.01). Altogether 103 pathogen strains were identified from 102 IFI cases as one Candida parapsilosis strain and one Laurent cryptococcus strain were both isolated from one patient. The most prevalent three pathogens were Candida albicans [51.5% (53/103)], Candida parapsilosis [24.3% (25/103)] and Candida glabrata [8.7% (9/103)]. The isolated rates of Candida albicans and Candida glabrata strains in the latter period were higher than those in the former period [63.3% (31/49) vs 40.7% (22/54), χ2=5.218; 18.4% (9/49) vs 0.0% (0/54), χ2=10.868], while the isolated rate of Candida parapsilosis strain was lower in the latter period than that in the former period [12.2% (6/49) vs 35.2%(19/54), χ2=7.355] (all P<0.05). All pathogen strains were sensitive strains except one Candida krusei strain which was isolated in the former period and was resistant to Fluconazole. Conclusions Premature infants born at lower gestational ages or with low birth weights are still at high-risk of IFI, but the incidence of IFI has declined in recent years. Routine administration of fluconazole in high-risk infants in NICU could prevente IFI without increasing drug resistance. Candida albicans is the predominant pathogen of IFI. Key words: Mycoses; Intensive care units, neonatal; Incidence; Infant, low birth weight; Candida

Recognition and diagnosis of invasive fungal infections in neonates

Fungal infections largely caused by Candida species are responsible for a significant disease burden in neonates and invasive candidiasis in hospitalised neonates has high associated morbidity and mortality. Early initiation of antifungal treatment improves outcome but the recognition and diagnosis of systemic fungal infection in this population is difficult due to the non-specific clinical presentation and the high false negative rate of cultures. There is a need for a practical, sensitive and rapid diagnostic test to enhance identification and early treatment. Serum detection of (1,3)-β-d-glucan and Candida PCR are promising candidates but at present limited data exists for their use in the neonatal intensive care setting. Until such investigations are validated, early initiation of antifungal treatment on the basis of risk factor profile and clinical features remains the safest practical approach.

Generalized erythroderma in a neonate: Rare presentation of nosocomial candidemia with Candida tropicalis

Candida species are currently one of the most common causes of nosocomial infection in neonatal intensive care units with Candida tropicalis emerging as a frequent offending agent. Colonization precedes systemic invasive fungal infections in neonates. It generally manifests with lethargy, increased apneic episodes, feeding intolerance, late-onset thrombocytopenia, hypoglycemia, unexplained hyperglycemia, poor perfusion, and need for increased ventilator requirement. We report a 10-day-old neonate, with late-onset sepsis by C. tropicalis, presenting as generalized erythroderma. We present this case to highlight that nosocomial candidemia with C. tropicalis can present as generalized erythroderma in neonates. Maintaining adequate oral or parenteral fluidintake with monitoring of serum electrolytes is mandatory.

Retrospective analysis of 76 neonates with invasive fungal infection in 2004-2014

Objective To investigate the clinical features of neonatal invasive fungal infection(IFI)so as to guide diagnosis, prevention and treatment of IFI. Methods Seventy-six neonates with IFI admitted to the Neonatal Intensive Care Unit (NICU) at Children's Hospital of Fudan University from 2004 to 2014 were included in the study. Pathogens, clinical manifestation, risk factor exposure, laboratory findings, complications, and clinical outcome of neonatal IFI were analyzed. Results Seventy-six cases were diagnosed as IFI between 2004 and 2014, with an yearly increasing trend. Sixty-eight patients were premature infants (89.5%). Of the 76 cases, except one with unknown birth weight, 11(14.7%), 34(45.3%), 20(26.7%)and 10 (13.3%) cases had birth weight < 1 000 g, (≥1 000-<1 500) g, (≥1 500-<2 500) g and ≥2 500 g, respectively. The pathogens were mainly Candida (74/76, 97.4%), including 26 cases of Candida albicans (34.2%). However, the incidence of non-Candida albicans infection was increasing. Candida guilliermondii was the most common in non-Candida albicans, accounting for 29.2% (14/48). All Candida albicans were sensitive to fluconazole. One strain of Candida glabrata was resistant to fluconazole. The most common risk factors included use of broad-spectrum antibiotics(93.3%, 56/60), parenteral nutrition(70.0%, 42/60), central vein catheterization(53.3%, 32/60), invasive ventilation(40.0%, 24/60) and history of abdominal surgery(21.7%,13/60). Clinical manifestations of IFI included temperature instability, frequent apnea, increased requirement of respiratory support and feeding intolerance. Among all cases, six were diagnosed as central nervous system infection. Of the patients who received cranial MRI, 46.8%(22/47) showed multiple abnormal signals in cerebral parenchyma. Fifty-two patients were cured and seven patients died before discharge, including one death due to fungal infection. Conclusions There is an increasing trend of IFI cases in NICU, especially in premature infants. Non-Candida albicans has become the main pathogenic fungus. There are no specific clinical manifestations in neonatal IFI. Use of broad-spectrum antibiotics, parenteral nutrition and central venous catheterization are common risk factors, and preventive measures should be taken in high-risk infants. In addition, IFI in neonates may affect important organs such as central nervous system, thus early treatment is necessary in suspected patients. Key words: Candidiasis, invasive; Intensive care units, neonatal; Retrospective studies

Clinical characteristics of invasive fungal infection in neonates

Objective To analyze the clinical features of invasive fungal infection in neonates, in order to raise the level of diagnosis and treatment of this disease. Methods We collected the clinical data of 66 cases who were admitted to our neonatal ward between January 2012 and March 2012, and confirmed invasive fungal infection.The data which included general situation, onset time, the first symptoms, laboratory examination results, blood culture results, treatment and outcomes, prognosis and follow-up status were analyzed retrospectively. Results The average onset age of invasive fungal infection was(18.6±9.1) day-old.The clinical manifestations of the children patients were atypical, with the first clinical presentation of apnea, feeding intolerance, fever, poor response and less activity.The laboratory examination results were thrombocytopenia, elevated C-reactive protein and(1, 3) -β-D-glucan.The blood culture results mainly were Candida.And fluconazole treated was effective. Conclusion Fungal infection in neonate is lack of specific clinical manifestations.We should be on high alert for the neonate who has high risk factors of fungal infections.Early diagnosis and prompt treatment is the key to successful treatment. Key words: Neonate; Invasive fungal infection; Clinical characteristics

Research progress on the invasive fungal infections in neonates

Invasive fungal infections(IFI) is significantly increasing in the neonatal intensive care unit recent years.The characteristics of IFI are low diagnosis rate, high mortality and being prone to involve in the nervous system.In this paper, the epidemiology, pathogens, risk factors, diagnosis, treatment and prevention of neonatal IFI are reviewed to improve the understanding of this disease. Key words: Infant, newborn; Fungal infections; Diagnosis; Treatment; Prevention

A case of disseminated Candida dubliniensis in a preterm infant: The importance of early detection and management of invasive fungal infections in neonates

This case report highlights the importance of timely diagnosis of disseminated fungal infections in neonates, as well as the increased incidence of infection with non- albicans Candida , and the association with surgical conditions such as necrotising enterocolitis.

Neonatal invasive fungal infection in England 2004–2010

Rates of invasive fungal infection are highest among neonates, especially those of low birthweight. This study aimed to describe the current epidemiology of invasive neonatal fungal infections in a UK neonatal infection surveillance network. From 2004 to 2010 prospective multicentre surveillance was conducted by 14 neonatal units using a web-based database. Clinicians then completed a standardized pro forma for each positive fungal blood and/or cerebrospinal fluid culture. The overall incidence was 2.4/1000 neonatal unit admissions and was highest among babies <1000 g (extreme low birthweight, 18.8/1000). Only five infants (6%) were >1500 g. The majority of infections were caused by Candida albicans (59; 69%) and Candida parapsilosis (17; 20%); 33% of infants had received antifungal prophylaxis. Known risk factors (use of central venous catheter, parenteral nutrition, previous antibiotic use) were common among cases. The attributable case fatality rate was 21% (18/84). Extreme low birthweight infants remain at highest risk of invasive fungal infection and prophylaxis should be particularly considered for this group. The number needing to receive prophylaxis to prevent one case varies significantly among units, hence unit-specific decisions are required. Further research is still needed into the optimal empiric and therapeutic strategies.