Neonatal Candidiasis Among Extremely Low Birth Weight Infants: Risk Factors, Mortality Rates, and Neurodevelopmental Outcomes at 18 to 22 Months
BACKGROUND. Neonatal candidiasis is associated with substantial morbidity and mortality rates. Neurodevelopmental follow-up data for a large multicenter cohort have not been reported.
METHODS. Data were collected prospectively for neonates born at <1000 g at National Institute of Child Health and Human Development-sponsored Neonatal Research Network sites between September 1, 1998, and December 31, 2001. Uniform follow-up evaluations, including assessments of mental and motor development with the Bayley Scales of Infant Development II, were completed for all survivors at corrected ages of 18 to 22 months. We evaluated risk factors for the development of neonatal candidiasis, responses to antifungal therapy, and the association between candidiasis and subsequent morbidity and death.
RESULTS. The cohort consisted of 4579 infants; 320 of 4579 (7%) developed candidiasis; 307 of 320 had Candida isolated from blood, 27 of 320 had Candida isolated from cerebrospinal fluid, and 13 (48%) of 27 of those with meningitis had negative blood cultures. In multivariate analysis of risk factors on day of life 3, birth weight, cephalosporins, gender, and lack of enteral feeding were associated with development of candidiasis. After diagnosis, most neonates had multiple positive cultures despite antifungal therapy, and 10% of neonates had candidemia for ≥14 days. Death or neurodevelopmental impairment (NDI) was observed for 73% of extremely low birth weight infants who developed candidiasis. Death and NDI rates were greater for infants who had delayed removal or replacement of central catheters (>1 day after initiation of antifungal therapy), compared with infants whose catheters were removed or replaced promptly.
CONCLUSIONS. Blood cultures were negative for approximately one half of the infants with Candida meningitis. Persistent candidiasis was common. Delayed catheter removal was associated with increased death and NDI rates.
Neonatal candidiasis is often fatal among premature infants.1–4 Most reports of neonatal candidiasis are single-center investigations.2,5–9 Because of the gravity of the condition, newer diagnostic and treatment strategies are needed.10 Well-defined, multicenter data regarding the epidemiologic and follow-up findings for patients with neonatal candidiasis are needed to facilitate design of such studies. In this report, we evaluate risks for the development of neonatal candidiasis, the clinical and microbiologic course for infants with documented candidiasis, and neurodevelopmental follow-up findings at corrected ages of 18 to 22 months among a large cohort of survivors with birth weights of <1000 g, from the National Institute of Child Health and Human Development Neonatal Research Network.
Data were collected prospectively from neonates born weighing <1000 g, between September 1, 1998, and December 31, 2001, at Neonatal Research Network sites.11 Trained research personnel collected maternal demographic, pregnancy, and delivery data soon after birth and infant data until 120 days of age, discharge, or death. Clinical data for these neonates were recorded prospectively. The registry includes data on late-onset sepsis, infecting organisms, and antibiotic therapy. In September 1998, infection surveillance was expanded to include detailed data on results of all blood and cerebrospinal fluid (CSF) cultures, postnatal age at the time of infection, antibiotic use, presence of indwelling catheters, and other risk factors for infection. Blood cultures were processed by the clinical microbiology laboratory at each center, with either Bactec (Becton Dickinson, Sparks, MD) or BacT/Alert (Organon Teknika, Durham, NC) systems. Decisions to obtain cultures through catheters or peripherally were at the discretion of the bedside clinician. Quantitative blood cultures were completed in accordance with local practices, but these data were not collected. One half to 1 mL of blood was inoculated per blood culture; an average of 6 blood cultures were obtained from the infants in this cohort, and an average of 1 CSF culture was obtained from each infant.
Neonates eligible for inclusion in the study were born between September 1, 1998, and December 31, 2001, weighed ≤1000 g at birth, and survived beyond 3 days of life (DOL). A comprehensive follow-up evaluation was conducted at a corrected age of 18 to 22 months for surviving infants (Fig 1). The institutional review board at each center approved participation in the registry and the follow-up studies. Written informed consent was obtained from parents or legal guardians for follow-up evaluations.
An episode of candidemia was defined if an infant had a positive blood culture for Candida; an episode of Candida meningitis was defined if an infant had a positive CSF culture for Candida. Candidiasis was defined if a patient had a positive blood culture or CSF culture for Candida. The CSF cultures reported as positive were first positive results. Data for CSF (cell counts) were not collected. Therefore, cases of abnormal CSF parameters and positive blood cultures were not captured with these data. It is acknowledged that the meningitis burden is therefore underestimated with the exclusion of presumptive meningitis. The total burden of candidiasis is also underestimated, because we did not collect urine culture data and end-organ damage (eg, ophthalmoscopic examination or echocardiogram findings) was not recorded.
The comprehensive neurodevelopmental evaluation at 18 to 22 months of age included an interview with the primary caretaker for the infant, assessments of mental and motor development with the Bayley Scales of Infant Development II,12 a neurologic examination, and ascertainment of hearing and vision impairment. A Bayley Scale score of <70 (>2 SDs below the mean) in either the Mental Developmental Index (MDI) or the Psychomotor Developmental Index (PDI) was considered to indicate significant delay. Cerebral palsy was defined as a moderate or severe nonprogressive disorder characterized by abnormal tone in ≥1 extremity and abnormal control of movement and posture. Blindness was defined as no useful vision in either eye. Deafness was defined as disability, with hearing aids in both ears. One or more of these outcomes found in the neurodevelopmental evaluation was defined as neurodevelopmental impairment (NDI), a composite measure.
Successful antifungal therapy was defined as survival and clearance of Candida from the blood within 14 days after the start of antifungal therapy.13 Antifungal therapy failure was defined as either death within 14 days after initiation of antifungal therapy or positive blood or CSF cultures >14 days after initiation of therapy. A false-negative blood culture was defined as the following sequence of events: ≥1 blood culture positive for Candida, followed by ≥1 blood culture negative for Candida, followed by ≥1 blood culture positive for Candida. Central catheters included surgically or percutaneously placed central venous catheters and umbilical venous catheters. Central catheter removal was defined as removal of the catheter present at the time of infection. Central catheter replacement was defined as removal of the central catheter present at the time of infection and placement of a new central catheter, at the time of removal or any time after removal. Early catheter removal or replacement was defined as removal or replacement of all central catheters within 1 day after initiation of antifungal therapy. Late catheter removal or replacement was defined as retention of ≥1 central vascular catheter for ≥2 days after initiation of antifungal therapy.
Estimated gestational age (EGA) was determined with best obstetric estimates. Intraventricular hemorrhage (IVH) was graded according to the method described by Papile et al.14 Early risk factors were defined as the clinical and demographic features present or absent on DOL 3 and included race, birth weight, gender, race, EGA, IVH, parenteral alimentation, enteral nutrition, central venous catheter, mechanical ventilation, antibiotics/cephalosporins, center, and H2 receptor blockers.1–10 Risk factors evaluated for antifungal therapy failure and poor outcomes at 18 to 22 months included birth weight, gender, race, EGA, IVH, days of parenteral alimentation, age at first enteral feeding, days with central venous catheter, days of mechanical ventilation, days with peripherally inserted central catheter, center, Candida species, H2 receptor blockers, and patent ductus arteriosus (PDA).
Analyses were based on the presence or absence of a first episode of candidiasis. Neurodevelopmental differences were assessed with a t test, χ2 test, or Fisher's exact test, on the basis of sample size and data distribution. In comparisons of neurodevelopmental outcomes of neonates with candidemia, neonates with Candida meningitis, and infants without candidiasis, infants with both candidemia and Candida meningitis (n = 8) were included in the group with meningitis. For multivariate analyses, logistic regression was used, with either forward addition or backward elimination (P value for retention: <.05). To assess the distribution of candidemia in relation to the DOL of the infant, the DOL of the first positive blood culture was used and incremental risk periods (eg, DOL 3–6, 7–10, 11–20, 21–30, 31–40, or 41–50) were devised. These divisions were chosen in an effort not to provide a uniform risk rate but rather to assist in future targeted prophylaxis and early-diagnosis study design. Because center predicted candidiasis, we decided that a variable reflecting this association should be included in the analysis. However, the inclusion of 16 centers (in addition to the predictor variables) in a model with only 300 observations yields a model that is not robust. To balance the needs to include center variability and to obtain a robust model, we elected to use quartile ranks of the outcome of interest in the multivariate analyses presented. Analyses were completed with SAS software (SAS Institute, Cary, NC); P values and 95% confidence intervals (CIs) are 2-tailed.
Rates of Infection and Age at Diagnosis
As shown in Fig 1, 4579 infants (401–1000 g) born and/or cared for at Neonatal Research Network centers survived ≥72 hours. Invasive candidiasis was documented with blood or CSF cultures for 320 infants (7%), and 219 (68%) of those infants survived and were discharged from the nursery; 84% (178 of 211) of the survivors were examined in the follow-up clinics. Among those who did not have candidiasis, 83% (3556 of 4259 patients) survived and 83% of the survivors (2874 of 3471 patients) were examined in the follow-up clinics.
As shown in Table 1, Candida albicans and Candida parapsilosis were the most common species isolated from the blood and CSF. C albicans was isolated from nearly 50% of positive blood cultures and 81% of Candida-positive CSF cultures. C parapsilosis was isolated from 43% of blood cultures and 26% of CSF cultures. C albicans was associated with a higher mortality rate than C parapsilosis (42% vs 20%; P < .001). The organism distributions and subsequent mortality rates were similar for infants with positive blood cultures alone and infants with documented meningitis. Of the infants whose cultures grew both C parapsilosis and C albicans, the 2 organisms were isolated on the same day or within 2 days.
Blood cultures were an insensitive diagnostic tool to determine the presence of meningitis; 3972 of 4579 infants were evaluated for sepsis with a blood culture, and 2016 (51%) of 3972 underwent both lumbar puncture and blood culture. Of the 2016 lumbar punctures in which a blood culture was also obtained, 209 were blood culture positive for Candida and CSF culture negative for Candida, and 1780 were blood culture negative for Candida and CSF culture negative for Candida. Therefore, 27 CSF cultures (1.3%) were positive for Candida. Thirteen (48%) of the 27 infants with meningitis had negative blood cultures for Candida. Therefore, 320 infants had candidiasis (either meningitis or candidemia); 14 of 320 infants had candidemia and Candida meningitis, 293 of 320 had candidemia, and 13 of 320 had Candida meningitis with a negative blood culture. The timing of the first positive culture is outlined in Fig 2.
Risk Factors for Candidiasis Present on DOL 3
Table 2 shows the early risk factors (present at DOL 3) for subsequent development of candidiasis among infants with birth weights of 401 to 1000 g. After adjustment, infants with birth weights of 401 to 750 g had candidiasis more frequently than did infants of 751 to 1000 g (odds ratio [OR]: 3.22; 95% CI: 2.47–4.19; P < .001). Infants who were receiving enteral feeding by DOL 3 developed candidiasis less frequently than did those who began enteral feeding later, and third-generation cephalosporin use at DOL 3 was associated with subsequent candidiasis. Candidiasis was not independently associated with the following potential risk factors present or absent on DOL 3: race, gender, IVH, parenteral alimentation, H2 receptor blockers, mechanical ventilation, and central vascular access.
Microbiologic Course (Clearance, Recurrence, and Catheter Management)
More than one half of the infants had >1 positive blood culture for Candida (median: 2; range: 1–30). The duration of candidemia after systemic antifungal agent treatment was often prolonged. The median number of days of positive blood cultures after systemic antifungal agent treatment among infants who survived was 3 days, and 10% of survivors had documented candidemia for ≥14 days. Intermittently negative cultures, ie, a negative blood culture obtained between days of positive blood cultures, were common occurrences. Intermittently negative (or false-negative) blood cultures occurred for 21% of the infants with documented candidemia (63 of 307 infants).
Prompt removal/replacement of the central catheter after candidemia was associated with better short-term and long-term outcomes. Each of the following was associated with delayed removal/replacement of the central catheter: mortality rate (delayed: 37%; prompt: 21%; P < .024), NDI rate among survivors (delayed: 63%; prompt: 45%; P = .08), and time to clear Candida from the blood (delayed: 7.3 days; prompt: 5.0 days; P = .11).
As shown in Table 3, NDI and death were associated with delayed removal/replacement of the catheter in multivariate logistic regression (OR: 2.69; 95% CI: 1.25–5.79). This association was observed after adjustment for gestational age, gender, and center. To develop the model, forward selection was used, and variables evaluated but not entered into the model during forward selection (P < .05) included birth weight, race, days with peripherally inserted central catheter, age at first enteral feeding, days of mechanical ventilation, PDA, IVH, days with central venous line, days of parenteral alimentation, Candida species, and H2 receptor blockers.
The increase in mortality rate observed with delayed catheter removal/replacement was observed across species, but mortality rates were relatively higher for neonates infected with C parapsilosis with delayed versus early management, compared with those infected with C albicans. Among infants with candidemia who had a central catheter at the time of culture and survived long enough to receive antifungal therapy, 100 had C albicans and 98 had C parapsilosis. Of the infants with C albicans, 31 had early removal/replacement and 11 (35%) died; 69 had delayed management and 33 (48%) died. Of the infants with C parapsilosis, 30 had prompt removal/replacement and 3 (10%) died; 68 had delayed management and 21 (31%) died. Species did not independently predict NDI, death, or microbiologic course (Tables 3 and 4).
Clinical and Microbiologic Responses to Systemic Antifungal Therapy
Three hundred eight of 320 infants with candidiasis received antifungal therapy. Ten infants died before receipt of antifungal therapy, and 2 infants were transferred after cultures were obtained but before antifungal therapy. A total of 241 infants received their first dose of antifungal therapy on the same day as or after documented positive cultures and 63 (26%) of 241 experienced antifungal therapy failure. Antifungal therapy failure (death within 14 days after initiation of antifungal therapy or ≥1 positive blood culture for Candida ≥14 days after initiation of antifungal therapy) was observed more frequently among neonates with birth weights of 401 to 750 g, compared with infants with birth weights of 751 to 1000 g (OR: 2.78; 95% CI: 1.17–6.59). As shown in Table 4, antifungal therapy failure was not associated independently with race, gender, IVH, PDA, H2 receptor blockers, mechanical ventilation, days with peripherally inserted central catheter, center, gestational age, Candida species, or central vascular access.
Of the infants who did not have documented meningitis and who received antifungal therapy (295 of 320 infants), 104 of 295 were given their first dose either on or before the day that the first blood culture that turned positive was obtained and 34 of 104 (33%) died. Of the 191 of 295 infants who received their first dose of antifungal therapy after the initial blood culture was obtained, 53 (28%) died.
Antifungal therapy was variable, and 4 infants received only fluconazole. Of the infants who received an amphotericin B product in combination with a second antifungal agent, 58% died or had prolonged candidemia; 32% of infants who received amphotericin B deoxycholate died or had prolonged candidemia; and 15% of infants who received only amphotericin B lipid complex product died (none had prolonged candidemia). Of the infants with meningitis, the median time to clear CSF was longer for those who received flucytosine plus amphotericin B deoxycholate (17.5 days; 6 infants), compared with those who received only amphotericin B (6 days; 18 infants).
Follow-Up Data for Survivors
Table 5 shows neurodevelopmental follow-up findings at 18 to 22 months of age. Of the 3049 tested infants, 178 had candidiasis and 15 of 178 had meningitis. The median Bayley MDI and PDI scores were lower for the neonates who had been infected with Candida (P < .001). Children infected with Candida were more likely to have Bayley MDI or PDI scores of <70, were more likely to have moderate or severe cerebral palsy, and were more likely to be blind or deaf than were neonates not infected with Candida. Death/NDI rates at a corrected age of 18 to 22 months across antifungal therapies were as follows: all 4 infants who received fluconazole died or had NDI, 81% of infants who received amphotericin in conjunction with other antifungal therapies died or had NDI, 75% of infants who received amphotericin B deoxycholate monotherapy died or had NDI, and 45% of infants who received amphotericin lipid complex died or had NDI.
The goals of this analysis were to present the epidemiologic features of neonatal candidiasis in a large multicenter cohort of extremely low birth weight (ELBW) infants, including incidence of infection, hospital course, and in-hospital mortality rates, and to evaluate postdischarge neurodevelopmental outcomes at 18 to 22 months in a cohort of neonates with candidiasis. It is hoped that these data will provide information helpful for the design and implementation of prospective randomized trials that will reduce the risk of invasive candidiasis and improve therapy for infected infants.
In a study of 46 ELBW infants with candidiasis from Ontario, Canada, the neurodevelopmental outcomes of neonates with a history of Candida infection were compared with those of a group of selected control subjects.3 The infected infants developed candidiasis in the years 1988 to 1996, and the control group was a cohort of infants monitored for a separate study of surfactant use in the years 1990 to 1994. Our neurodevelopmental outcome results are more detailed in presentation but similar to the findings reported from Ontario, ie, neonates with candidiasis had worse neurodevelopmental outcomes. However, the Ontario group found that infants infected with Candida were no more likely to die than control subjects.3 There are several important methodologic differences between the study conducted in Ontario and this report. The control subjects in the report from Ontario were all infants admitted to the hospital, whereas the infants infected with Candida in that study were infants who survived long enough (>3 days) to develop nosocomial candidiasis. A large proportion of the deaths in the ELBW population occur in the first 72 hours of life. To develop nosocomial candidiasis, an infected infant must survive the first 72 hours of life, but this survival was not guaranteed in the control group from Ontario. Other methodologic differences include that fact that our study was a complete cohort study and the data were collected from a geographically diverse group of NICUs. The observed survival differences, which were not observed in the Ontario study, can be explained by differences in study methods.
The diversity and number of centers represent a strength of this study. With few exceptions,1,3,10 most of the published reports on neonatal candidiasis involved studies conducted at <5 sites, with most being single-center reports. A substantial limitation of single-center studies is the wide disparity in neonatal outcomes.15 The data presented in this multicenter report have important implications for the design of neonatal candidiasis clinical trials and for treatment and long-term prognosis for infants with invasive candidiasis.
Implications for the Design of Prophylaxis and Treatment Studies
Single-center pilot studies of neonatal candidiasis prophylaxis have been conducted.5,6 Several key questions regarding prophylaxis trial design include selection of risk factors for inclusion criteria and the timing and length of prophylaxis to limit the length of exposure to antifungal drugs. One strategy might be to focus on the time exposed to a prophylactic agent. However, these data establish that the risk period for neonatal candidiasis starts at the end of the first week of life and extends well into the third month of life. An alternative strategy that has been used5 has been to evaluate risk factors at the beginning of the period of prophylaxis (DOL 3) and to limit prophylaxis to high-risk infants. For example, Kaufman et al5 required central vascular access or mechanical ventilation in the first 1 week of life for study inclusion. Although central vascular access and mechanical ventilation are likely to be risk factors for the development of candidiasis, the presence or absence of these foreign bodies at DOL 3 would have minimal effects on sample size in a randomized prophylaxis trial.
Similarly, although lack of enteral feeding and use of cephalosporins on DOL 3 were associated with subsequent candidiasis, their presence or absence likely would not have a substantive impact on sample size in a randomized prophylaxis trial. The key risk factor known by DOL 3 was lower birth weight, even in this cohort of infants born at <1000 g. These data suggest that a trial of antifungal prophylaxis should target infants with birth weights of <750 g.
Candidemia is an important cause of infection-related death in the nursery,1,4 but there has not been a well-powered, multicenter, antifungal treatment trial among patients with neonatal candidiasis. Candidiasis treatment trials among older patients rely on a design that has both survival and microbiologic response to therapy as part of the primary outcome. Microbiologic response to therapy is clearance of Candida documented by negative cultures. The composite outcome of survival and microbiologic response to antifungal therapy based on a randomized trial in neonates is not known, but the overall failure rate in this cohort of neonates of <1000 g was 26%. These data also suggest that birth weight is the strongest predictor of clinical and microbiologic responses. Prolonged candidemia and/or death was observed for 31% of infants with birth weights of <750 g, compared with 14% of infants of 750 to 1000 g. These data also suggest that microbiologic responses should be documented by 2 negative cultures separated by ≥48 hours.
There was some variation among antifungal regimens and antifungal responses; P values are not provided for these data because we would like to emphasize the caution with which these results should be interpreted. Antifungal therapy choices were not standardized, and dosage changes within and between regimens were not recorded and were quite likely to be variable. One point worth emphasizing from this component of the study is that clearance of CSF was longer among neonates who received amphotericin B and flucytosine, compared with those who received only amphotericin B. In a small case series, flucytosine was reported to be associated with more rapid clearance in cases of meningitis. These data emphasize that optimal antifungal management is not known.
Implications for Clinical Care and Follow-Up Monitoring
These data support the need for improved diagnostic testing for neonatal candidiasis16,17; 21% of the neonates with candidemia had false-negative blood cultures while receiving antifungal therapy. Therefore, 2 negative blood cultures, separated by ≥48 hours, should be obtained to document sterilization of the bloodstream (similar to treatment for older patients14). The optimal length of therapy, however, is not answered by these data. Negative blood cultures were also seen for almost one half of the neonates with Candida meningitis. These data suggest that lumbar puncture is especially important for neonates at risk for candidiasis, if the clinical stability of the infant is such that the procedure can be performed. However, negative lumbar puncture findings should not be viewed as proof that the central nervous system is free of involvement; the hallmark of central nervous system candidiasis among neonates is meningoencephalitis rather than strictly meningitis.
Persistent candidemia was common; 10% of the neonates with candidemia had documented positive blood cultures for ≥2 weeks despite antifungal therapy. Persistent candidemia was more common among neonates whose catheters were not removed or replaced promptly. Prompt catheter removal was also associated with lower mortality rates, improved neurodevelopmental outcomes, and higher Bayley Scale scores. The duration of candidemia (5.1 days) in the prompt-removal group was longer than that observed in a study that reported time to clearance with prompt catheter removal.9 That report, however, included both premature and term infants; greater prematurity has been associated consistently with poor outcomes in candidiasis, and the younger infants of this cohort may help explain that difference. These data should be interpreted with caution, because this cohort study was not a randomized trial of catheter management and, once an infant had documented candidiasis, blood culture acquisition and catheter removal were not standardized, relative to severity of illness or other factors. Nonetheless, these data support the single-center work outlined previously by others,8,9 strongly suggesting prompt catheter removal in this population.
Short-term mortality rates were substantial; >30% of the infants with candidiasis died before discharge from the nursery. The 18- to 22-month follow-up prognosis for <1000-g infants with candidiasis was extremely poor; 73% of the children for whom follow-up data were available died or had NDI. The 18- to 22-month follow-up outcomes were worse for infants with candidiasis than for infants who did not develop candidiasis, although the group that did not develop candidiasis included both uninfected infants and infants infected with bacterial pathogens. These data suggest that close follow-up monitoring of infants diagnosed as having candidiasis is warranted.
D.K.B. received support from the National Institute of Child Health and Human Development (grant HD044799-01) and the Thrasher Research Fund (02819-5). The National Institute of Child Health and Human Development provided oversight of the project through a cooperative agreement (grant wherein the funding federal agency is substantially involved in carrying out the research program and federal scientists [R.D.H.] collaborate directly with the researchers on a joint research project).
Members of the National Institute of Child Health and Human Development Neonatal Research Network are as follows: Alan Jobe, MD, Chairman (principal investigator), University of Cincinnati; University of California, San Diego (grant U10 HD40461): Neil N. Finer, MD (principal investigator), and Wade Rich; Case Western Reserve University (grant U10 HD21364): Avroy A. Fanaroff, MB, BCh (principal investigator), Michele Walsh, MD, and Nancy Newman, RN; University of Cincinnati (grant M01 RR 08084): Edward F. Donovan, MD (principal investigator), Vivek Narendran, MD, MRCP, and Cathy Grisby, RN; Duke University (grant U10 HD40492): Ronald N. Goldberg, MD (principal investigator), Michael Cotten, MD, and Kathy Auten, RN; Emory University (grant M01 RR 00750): Barbara J. Stoll, MD (principal investigator), and Ellen Hale, RN; Indiana University (grant M01 RR 00039): James A. Lemons, MD (principal investigator), Brenda Poindexter, MD, and Lucy Miller, RN; University of Miami (grant U10 HD21397): Shahnaz Duara, MD (principal investigator), Charles R. Bauer, MD, and Ruth Everett, RN; National Institute of Child Health and Human Development: Rosemary D. Higgins, MD (principal investigator), and James Hansen, MD; University of New Mexico (grant M01 RR 00997): Lu-Ann Papile, MD (principal investigator), Jean Lowe, RN, and Conra Backstrom, RN; Research Triangle Institute (grant U01 HD36790): W. Kenneth Poole, PhD (principal investigator), Beth McClure, MEd, Betty Hastings, and Carolyn M. Petrie, MS; University of Rochester (grant M01 RR 00044): Dale L. Phelps, MD (principal investigator), Gary Myers, MD, Diane Hust, RN, PNP, and Linda Reubens, RN; Stanford University (grant M01 RR 00070): David K. Stevenson, MD (principal investigator), Krisa Van Meurs, MD, and Bethany Ball, BS; University of Tennessee at Memphis (grant U10 HD21415): Sheldon B. Korones, MD (principal investigator), Henrietta Bada, MD, and Tina Hudson, RN; University of Texas Health Science Center at Houston (grant U10 HD21373): Jon E. Tyson, MD, MPH (principal investigator), Kathleen Kennedy, MD, MPH, and Georgia McDavid, RN; University of Texas Southwestern Medical Center (grant U10 HD40689): Abbot R. Laptook, MD (principal investigator), Gay Hensley, RN, Sue Broyles, MD, and Jackie Hickman, RN; Wake Forest University (grant U10 HD40498): T. Michael O'Shea, MD (principal investigator), Robert Dillard, MD, Barbara Jackson, RN, and Nancy Peters, RN; Wayne State University (grant U10 HD21385): Seetha Shankaran, MD (principal investigator), Rebecca Bara, RN, Yvette Johnson, MD, MPH, and Deborah Kennedy, RN; Women and Infants Hospital (grant U10 HD27904): William Oh, MD (principal investigator), Barbara Stonestreet, MD, and Angelita Hensman, RN; Yale University (grant M01 RR 06022): Richard A. Ehrenkranz, MD (principal investigator), Patricia Gettner, RN, Linda Mayes, MD, and Elaine Romano, PNP; University of Alabama at Birmingham (grant U10 HD34216): Waldemar A. Carlo, MD (principal investigator), Namasivayam Ambalavanan, MD, and Monica V. Collins, RN; Harvard University (grant U10 HD34167, M01 RR 02635, M01 RR 02172, M01 RR 01032): Ann R. Stark, MD (principal investigator), and Kerri Fournier, RN.
- Accepted March 29, 2005.
- Address correspondence to Daniel K. Benjamin, MD, MPH, PhD, PO Box 17969, Duke Clinical Research Institute, Durham, NC 27705. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
The authors had full access to all of the data. The listed authors contributed to generation of the hypotheses (all authors), data collection (B.J.S., A.A.F., W.O., S.D., and R.G.), data analysis (D.K.B., S.A.M., and K.P.), data interpretation (all authors), and writing of the report (all authors). The members of the National Institute of Child Health and Human Development Neonatal Research Network contributed to the collection data.
- ↵Stoll BJ, Hansen N, Fanaroff AA, et al. Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics.2002;110 :285– 291
- ↵Benjamin DK Jr, Poole C, Steinbach WJ, Rowen JL, Walsh TJ. Neonatal candidemia and end-organ damage: a critical appraisal of the literature using meta-analytic techniques. Pediatrics.2003;112 :634– 640
- ↵Friedman S, Richardson SE, Jacobs SE, O'Brien K. Systemic Candida infection in extremely low birth weight infants: short term morbidity and long term neurodevelopmental outcome. Clin Infect Dis.2000;19 :499– 504
- ↵Stoll BJ, Hansen N, Fanaroff AA, et al. To tap or not to tap: high likelihood of meningitis without sepsis among very low birth weight infants. Pediatrics.2004;113 :1181– 1186
- ↵Kicklighter SD, Springer SC, Cox T, Hulsey TC, Turner RB. Fluconazole for prophylaxis against candidal rectal colonization in the very low birth weight infant. Pediatrics.2001;107 :293– 298
- Fernandez M, Moylett EH, Noyola DE, Baker CJ. Candidal meningitis in neonates: a 10-year review. Clin Infect Dis.2000;31 :458– 463
- ↵Benjamin DK Jr, Ross K, Benjamin DK, McKinney RE Jr, Auten R, Fisher RG. When to suspect fungal infections in neonates: a clinical comparison of Candida albicans and Candida parapsilosis fungemia with coagulase-negative staphylococcal bacteremia. Pediatrics.2000;106 :712– 718
- ↵Karlowicz MG, Hashimoto LN, Kelly RE Jr, Buescher ES. Should central venous catheters be removed as soon as candidemia is detected in neonates? Pediatrics.2000;106 (5). Available at: www.pediatrics.org/cgi/content/full/106/5/e63
- ↵Hack M, Horbar JD, Malloy MH, Tyson JE, Wright E, Wright L. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Network. Pediatrics.1991;87 :587– 597
- ↵Bayley N. Bayley Scales of Infant Development II. San Antonio, TX: Psychological Corp; 1993
- ↵Lemons JA, Bauer CR, Oh W, et al. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Research Network, January 1995 through December 1996: NICHD Neonatal Research Network. Pediatrics.2001;107 (1). Available at: www.pediatrics.org/cgi/content/full/107/1/e1
- ↵Odabasi Z, Mattiuzzi G, Estey E, et al. β-d-Glucan as a diagnostic adjunct for invasive fungal infections: validation, cutoff development, and performance in patients with acute myelogenous leukemia and myelodysplastic syndrome. Clin Infect Dis.2004;39 :199– 205
- Copyright © 2006 by the American Academy of Pediatrics