Our objective was to survey neonatologists regarding international practice of red cell transfusion thresholds for premature infants with <1000-g birth weight and/or <28-week gestation. An invitation to fill out an 11-question web-based survey was distributed to neonatologists through their professional societies in 22 countries. Physicians were asked about which specific factors, in addition to hemoglobin levels, influenced their decisions about transfusing premature infants. These factors included gestational age, postnatal age, oxygen need, respiratory support, reticulocyte count, and inotropic support. Physicians were presented with 5 scenarios and asked to identify hemoglobin cutoff values for transfusing infants with <1000-g birth weight and/or <28-week gestation. One thousand eighteen neonatologists responded: the majority were from the United States (67.5%), followed by Germany (10.7%), Japan (8.0%), the United Kingdom (4.9%), Spain (3.9%), Italy (2.6%), Colombia (0.6%), Argentina (0.4%), Canada (0.4%), Belgium (0.1%), and the Netherlands (0.1%). Half of the respondents (51.1%) reported having a written policy with specific red cell transfusion guidelines in their unit. Factors considered “very important” regarding the need to administer blood transfusions included degree of oxygen requirement (44.7%) and need for respiratory support (44.1%). Erythropoietin was routinely used to treat anemia by 26.0% of respondents. Delayed cord clamping or cord milking was practiced by 29.1% of respondents. The main finding was of a wide variation in the hemoglobin values used to transfuse infants, regardless of postnatal age. Step-wise increments in the median hemoglobin cutoffs directly paralleled an increase in the need for levels of respiratory support. In the first week of life, there was a wider range in the distribution of hemoglobin transfusion thresholds for infants requiring no respiratory support and full mechanical ventilation compared with the thresholds used in the second, third, and fourth weeks of life. An international survey using hypothetical scenarios shows that red blood cell transfusion practices vary widely among practicing neonatologists in participating countries.

Context

Current guidelines, initially published in 1995, recommend antenatal corticosteroids for mothers with preterm labor from 24–34 weeks gestational age, but not before 24 weeks because of lack of data. However, many infants born before 24 weeks are provided intensive care now.

Objective

To determine if antenatal corticosteroids are associated with improvement in major outcomes in infants born at 22 and 23 weeks.

Design, Setting, Participants

Data for this cohort study were collected prospectively on 401–1000 gram inborn infants (N=10,541) of 22–25 weeks gestation born between 1993–2009 at 23 academic perinatal centers in the United States. Certified examiners unaware of exposure to antenatal corticosteroids performed follow-up examinations on 4,924 (86.5%) of the infants born in 1993–2008 who survived to 18–22 months. Logistic regression models generated adjusted odds ratios, controlling for maternal and neonatal variables.

Main Outcome Measures

Mortality and neurodevelopmental impairment at 18–22 months corrected age

RESULTS

Death or neurodevelopmental impairment at 18–22 months was lower for infants whose mothers received antenatal corticosteroids born at 23 weeks (antenatal corticosteroids, 83.4% vs no antenatal corticosteroids, 90.5%; adjusted odds ratio 0.58; 95% CI, 0.42–0.80), at 24 weeks (antenatal corticosteroids, 68.4% vs no antenatal corticosteroids, 80.3%; adjusted odds ratio 0.62; 95% CI, 0.49–0.78), and at 25 weeks (antenatal corticosteroids, 52.7% vs no antenatal corticosteroids, 67.9%; adjusted odds ratio 0.61; 95% CI, 0.50–0.74) but not at 22 weeks (antenatal corticosteroids, 90.2% vs no antenatal corticosteroids, 93.1%; adjusted odds ratio 0.80; 95% CI, 0.29–12.21). Death by 18–22 months, hospital death, death/intraventricular hemorrhage/periventricular leukomalacia, and death/necrotizing enterocolitis were significantly lower for infants born at 23, 24, and 25 weeks gestational age if the mothers had received antenatal corticosteroids but the only outcome significantly lower at 22 weeks was death/necrotizing enterocolitis (antenatal corticosteroids, 73.5% vs no antenatal corticosteroids, 84.5%; adjusted odds ratio 0.54; 95% CI, 0.30–0.97).

CONCLUSIONS

Among infants born at 23–25 weeks gestation, use of antenatal corticosteroids compared to non-use was associated with a lower rate of death or neurodevelopmental impairment at 18–22 months.

Aim

To explore the association of serum bilirubin level and breast milk feeding with retinopathy of prematurity (ROP) in preterm infants.

Methods

We conducted a case-control study to examine the independent and combined effects of serum bilirubin and breast milk feeding on ROP risk in infants <32 weeks gestation or with birth weight <1500 grams. Cases (66 infants with ROP) were matched with controls (66 infants without ROP) based on factors know to affect ROP risk.

Results

When analyzed using the paired t test, the peak bilirubin levels were lower in ROP cases than in controls (mean 7.2 vs 7.9 mg/dl; P=0.045). Using conditional logistic regression, we found a negative association between highest serum bilirubin level and risk of ROP (OR=0.82 per 1-mg/dl change in bilirubin (P=0.06). There was no significant association between breast milk feeding and risk of ROP.

Conclusion

Bilirubin may help to protect preterm infants against ROP.

Objective

To examine pathogens and other characteristics associated with late-onset bloodstream infections (BSI) in infants with intestinal failure (IF) as a consequence of necrotizing enterocolitis (NEC).

Study design

Infants 401–1500 grams at birth who survived >72 hours and received care at NICHD Neonatal Research Network centers were studied. Frequency of culture positive BSI and pathogens were compared for infants with medical NEC, NEC managed surgically without IF, and surgical IF. Among infants with IF, duration of parenteral nutrition (PN) and other outcomes were evaluated.

Results

932 infants were studied (IF, n=78; surgical NEC without IF, n=452; medical NEC, n=402). The proportion with BSI after NEC diagnosis was higher in infants with IF than with surgical NEC (p=0.007) or medical NEC (p<0.001). Gram positive pathogens were most frequent. Among infants with IF, increased number of infections was associated with longer hospitalization and duration on PN (0, 1, ≥2 infections; median stay (days): 172, 188, 260, p=0.06; median days on PN: 90, 112, 115, p=0.003), and the proportion who achieved full feeds during hospitalization decreased (87%, 67%, 50%, p=0.03).

Conclusion

Recurrent BSIs are common in VLBW infants with IF. Gram positive bacteria were most commonly identified in these infants.

Objective

The safe lower limit of hematocrit or hemoglobin that should trigger a red blood cell (RBC) transfusion has not been defined. The objective of this study was to examine the physiological effects of anemia and compare the acute responses to transfusion in preterm infants who were transfused at higher or lower hematocrit thresholds.

Methods

We studied 41 preterm infants with birth weights 500-1300 g, who were enrolled in a clinical trial comparing high (“liberal”) and low (“restrictive”) hematocrit thresholds for transfusion. Measurements were performed before and after a packed RBC transfusion of 15 ml/kg, which was administered because the infant's hematocrit had fallen below the threshold defined by study protocol. Hemoglobin, hematocrit, red blood cell count, reticulocyte count, lactic acid, and erythropoietin were measured before and after transfusion using standard methods. Cardiac output was measured by echocardiography. Oxygen consumption was determined using indirect calorimetry. Systemic oxygen transport and fractional oxygen extraction were calculated.

Results

Systemic oxygen transport rose in both groups following transfusion. Lactic acid was lower after transfusion in both groups. Oxygen consumption did not change significantly in either group. Cardiac output and fractional oxygen extraction fell after transfusion in the low hematocrit group only.

Conclusions

Our results demonstrate no acute physiological benefit of transfusion in the high hematocrit group. The fall in cardiac output with transfusion in the low hematocrit group shows that these infants had increased their cardiac output to maintain adequate tissue oxygen delivery in response to anemia and, therefore, may have benefitted from transfusion.

Objective

Preterm infants are frequently transfused with red blood cells based on standardized guidelines or clinical concern that anemia taxes infants’ physiological compensatory mechanisms and thereby threatens their health and well-being. The impact of various transfusion guidelines on long-term neurocognitive outcome is not known. The purpose of this study is to evaluate long-term neurocognitive outcome on children born prematurely and treated at birth with different transfusion guidelines.

Methods

Neurocognitive outcomes were examined at school age for 56 preterm infants randomly assigned to a liberal (n = 33) or restrictive (n = 23) transfusion strategy. Tests of intelligence, achievement, language, visual-spatial/motor, and memory skills were administered. Between-group differences were assessed.

Results

Those in the liberal transfusion group performed more poorly than those in the restrictive group on measures of associative verbal fluency, visual memory, and reading.

Conclusions

Findings highlight possible long-term neurodevelopmental consequences of maintaining higher hematocrit levels.

BACKGROUND:

Extremely low birth weight twins have a higher rate of death or neurodevelopmental impairment than singletons. Higher-order extremely low birth weight multiple births may have an even higher rate of death or neurodevelopmental impairment.

METHODS:

Extremely low birth weight (birth weight 401–1000 g) multiple births born in participating centers of the Neonatal Research Network between 1996 and 2005 were assessed for death or neurodevelopmental impairment at 18 to 22 months' corrected age. Neurodevelopmental impairment was defined by the presence of 1 or more of the following: moderate to severe cerebral palsy; mental developmental index score or psychomotor developmental index score less than 70; severe bilateral deafness; or blindness. Infants who died within 12 hours of birth were excluded. Maternal and infant demographic and clinical variables were compared among singleton, twin, and triplet or higher-order infants. Logistic regression analysis was performed to establish the association between singletons, twins, and triplet or higher-order multiples and death or neurodevelopmental impairment, controlling for confounding variables that may affect death or neurodevelopmental impairment.

RESULTS:

Our cohort consisted of 8296 singleton, 2164 twin, and 521 triplet or higher-order infants. The risk of death or neurodevelopmental impairment was increased in triplets or higher-order multiples when compared with singletons (adjusted odds ratio: 1.7 [95% confidence interval: 1.29–2.24]), and there was a trend toward an increased risk when compared with twins (adjusted odds ratio: 1.27 [95% confidence: 0.95–1.71]).

CONCLUSIONS:

Triplet or higher-order births are associated with an increased risk of death or neurodevelopmental impairment at 18 to 22 months' corrected age when compared with extremely low birth weight singleton infants, and there was a trend toward an increased risk when compared with twins.

OBJECTIVE:

We compared neurodevelopmental outcomes at 18 to 22 months' corrected age of infants born with extremely low birth weight at an estimated gestational age of <25 weeks during 2 periods: 1999–2001 (epoch 1) and 2002–2004 (epoch 2).

PATIENTS AND METHODS:

We conducted a multicenter, retrospective analysis of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Perinatal and neonatal variables and outcomes were compared between epochs. Neurodevelopmental outcomes at 18 to 22 months' corrected age were evaluated with neurologic exams and Bayley Scales of Infant Development II. Logistic regression analyses determined the independent risk of epoch for adverse outcomes.

RESULTS:

Infant survival was similar between epochs (epoch 1, 35.4%, vs epoch 2, 32.3%; P = .09). A total of 411 of 452 surviving infants in epoch 1 and 405 of 438 surviving infants in epoch 2 were evaluated at 18 to 22 months' corrected age. Cesarean delivery (P = .03), surgery for patent ductus arteriosus (P = .004), and late sepsis (P = .01) were more common in epoch 2, but postnatal steroid use was dramatically reduced (63.5% vs 32.8%; P < .0001). Adverse outcomes at 18 to 22 months' corrected age were common in both epochs. Moderate-to-severe cerebral palsy was diagnosed in 11.1% of surviving infants in epoch 1 and 14.9% in epoch 2 (adjusted odds ratio [OR]: 1.52 [95% confidence interval (CI): 0.86–2.71]; P = .15), the Mental Developmental Index was <70 in 44.9% in epoch 1 and 51% in epoch 2 (OR: 1.30 [95% CI: 0.91–1.87]; P = .15), and neurodevelopmental impairment was diagnosed in 50.1% of surviving infants in epoch 1 and 58.7% in epoch 2 (OR: 1.4 [95% CI: 0.98–2.04]; P = .07).

CONCLUSIONS:

Early-childhood outcomes for infants born at <25 weeks' estimated gestational age were unchanged between the 2 periods.

OBJECTIVE

Individuals with Down syndrome (DS) are at increased risk of several morbidities with lifelong health consequences. Little is known about mortality or morbidity risks in early infancy among very-low-birth-weight (VLBW) infants with DS. Our objective was to compare survival and neonatal morbidities between VLBW infants with DS and VLBW infants with other non-DS chromosomal anomalies, other non-chromosomal birth defects, and VLBW infants without major birth defects.

METHODS

Data were collected prospectively for infants weighing 401-1500 grams born and/or cared for at one of the study centers participating in the NICHD Neonatal Research Network from 1994 through 2008. Risk of death and morbidities including patent ductus arteriosus (PDA), necrotizing enterocolitis (NEC), late onset sepsis (LOS), retinopathy of prematurity (ROP), and bronchopulmonary dysplasia (BPD), were compared between VLBW infants with DS and infants in the other groups.

RESULTS

Infants with DS were at increased risk of death (adjusted relative risk [RR] 2.47, 95% confidence interval [CI] 2.00-3.07), PDA, NEC, LOS, and BPD relative to infants with no birth defects. Decreased risk of death (RR 0.40, 95% CI 0.31-0.52) and increased risks of NEC and LOS were observed when comparing infants with DS to infants with other non-DS chromosomal anomalies. Relative to infants with non-chromosomal birth defects, infants with DS were at increased risk of PDA and NEC.

CONCLUSION

The increased risk of morbidities among VLBW infants with DS provides useful information for counseling parents and for caretakers in anticipating the need for enhanced surveillance for prevention of these morbidities.

Objective

To examine risk factors for neonatal clinical seizures and to determine the independent association with death or neurodevelopmental impairment (NDI) in extremely low birth weight (ELBW) infants.

Study design

A total of 6499 ELBW infants (401–1000 g) surviving to 36 weeks postmenstrual age (PMA) were included in this retrospective study. Unadjusted comparisons were performed between infants with (n=414) and without (n=6085) clinical seizures during the initial hospitalization. Multivariate logistic regression modeling examined the independent association of seizures with late death (after 36 weeks PMA) or NDI after controlling for multiple demographic, perinatal, and neonatal variables.

Results

Infants with clinical seizures had a greater proportion of neonatal morbidities associated with poor outcome, including severe intraventricular hemorrhage, sepsis, meningitis, and cystic periventricular leukomalacia (all P < .01). Survivors were more likely to have NDI or moderate-severe cerebral palsy at 18 to 22 months corrected age (both P < .01). After adjusting for multiple confounders, clinical seizures remained significantly associated with late death or NDI (odds ratio 3.15 [95% confidence interval 2.37–4.19]).

Conclusions

ELBW infants with clinical seizures are at increased risk for adverse neurodevelopmental outcome, independent of multiple confounding factors.

OBJECTIVE

To examine the impact of birth at night, on the weekend, and during July or August – the first months of the academic year – and the impact of resident duty-hour restrictions on mortality and morbidity of VLBW infants.

METHODS

Outcomes were analyzed for 11,137 infants with birth weight 501–1250 grams enrolled in the NICHD Neonatal Research Network registry 2001–2005. Approximately half were born before the introduction of resident duty-hour restrictions in 2003. Follow-up assessment at 18–22 months was completed for 4,508 infants. Mortality (7-day and 28-day), short-term morbidities, and neurodevelopmental outcome were examined with respect to the timing of birth: night vs day, weekend vs weekday, and July or August vs other months, and after vs before implementation of resident duty-hour restrictions.

RESULTS

There was no effect of hour, day, or month of birth on mortality and no impact on the risks of short-term morbidities except the risk of ROP requiring operative treatment was lower for infants born during the late night hours than during the day. There was no impact of timing of birth on neurodevelopmental outcome except the risk of hearing impairment or death was slightly lower among infants born in July or August compared with other months. The introduction of resident and fellow duty-hour restrictions had no impact on mortality or neurodevelopmental outcome. The only change in short-term morbidity after duty-hour restrictions were introduced was an increase in the risk of ROP (stage 2 or higher).

CONCLUSION

In this network of academic centers, the timing of birth and the introduction of duty-hour restrictions had little effect on the risks of mortality and morbidity of VLBW infants, suggesting that staffing patterns were adequate to provide consistent care.

OBJECTIVE

This report presents data from the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network on care of and morbidity and mortality rates for very low birth weight infants, according to gestational age (GA).

METHODS

Perinatal/neonatal data were collected for 9575 infants of extremely low GA (22–28 weeks) and very low birth weight (401–1500 g) who were born at network centers between January 1, 2003, and December 31, 2007.

RESULTS

Rates of survival to discharge increased with increasing GA (6% at 22 weeks and 92% at 28 weeks); 1060 infants died at ≤ 12 hours, with most early deaths occurring at 22 and 23 weeks (85% and 43%, respectively). Rates of prenatal steroid use (13% and 53%, respectively), cesarean section (7% and 24%, respectively), and delivery room intubation (19% and 68%, respectively) increased markedly between 22 and 23 weeks. Infants at the lowest GAs were at greatest risk for morbidities. Overall, 93% had respiratory distress syndrome, 46% patent ductus arteriosus, 16% severe intraventricular hemorrhage, 11% necrotizing enterocolitis, and 36% late-onset sepsis. The new severity-based definition of bronchopulmonary dysplasia classified more infants as having bronchopulmonary dysplasia than did the traditional definition of supplemental oxygen use at 36 weeks (68%, compared with 42%). More than one-half of infants with extremely low GAs had undetermined retinopathy status at the time of discharge. Center differences in management and outcomes were identified.

CONCLUSION

Although the majority of infants with GAs of ≥24 weeks survive, high rates of morbidity among survivors continue to be observed.

This study is a retrospective, case-control study involving 535 preterm infants examining the roles of sequence polymorphisms in genes that mediate host immune responses to bacterial infection in newborn infants. A total of 49 single nucleotide polymorphisms (SNPs) in 19 candidate genes including inflammatory cytokines (IL6, IL10, IL1B, TNF), cytokine receptors (IL1RN), toll-like receptors (TLR2, TLR4, TLR5) and cell surface receptors (CD14) were genotyped. Subjects were stratified into 3 groups (sepsis, suspected sepsis, control). The data were analyzed using a family-based transmission disequilibrium test. We found that birth weight, gestational age, duration of rupture of membranes, and presence of clinical chorioamnionitis were strongly associated with sepsis. Polymorphisms in TLR2 (rs3804099), TLR5 (rs5744105), IL10 (rs1800896), and PLA2G2A (rs1891320) genes were associated with sepsis. Allelic variants in PLA2G2A and TLR2 were associated with Gram-positive infections, while IL10 was associated with Gram-negative infections (p< 0.05). We conclude allelic variations in PLA2G2A, TLR2, TLR5, and IL10 may moderate the predisposition to sepsis in preterm infants.

Background

Most neonates less than 1.0 kg birth weight need red blood cell (RBC) transfusions. Delayed clamping of the umbilical cord 1 minute after delivery transfuses the neonate with autologous placental blood to expand blood volume and provide 60 percent more RBCs than after immediate clamping. This study compared hematologic and clinical effects of delayed versus immediate cord clamping.

Study Design and Methods

After parental consent, neonates not more than 36 weeks' gestation were randomly assigned to cord clamping immediately or at 1 minute after delivery. The primary endpoint was an increase in RBC volume/mass, per biotin labeling, after delayed clamping. Secondary endpoints were multiple clinical and laboratory comparisons over the first 28 days including Score for Neonatal Acute Physiology (SNAP).

Results

Problems with delayed clamping techniques prevented study of neonates of less than 30 weeks' gestation, and 105 neonates 30 to 36 weeks are reported. Circulating RBC volume/mass increased (p = 0.04) and weekly hematocrit (Hct) values were higher (p < 0.005) after delayed clamping. Higher Hct values did not lead to fewer RBC transfusions (p ≥ 0.70). Apgar scores after birth and daily SNAP scores were not significantly different (p ≥ 0.22). Requirements for mechanical ventilation with oxygen were similar. More (p = 0.03) neonates needed phototherapy after delayed clamping, but initial bilirubin levels and extent of phototherapy did not differ.

Conclusions

Although a 1-minute delay in cord clamping significantly increased RBC volume/mass and Hct, clinical benefits were modest. Clinically significant adverse effects were not detected. Consider a 1-minute delay in cord clamping to increase RBC volume/mass and RBC iron, for neonates 30 to 36 weeks' gestation, who do not need immediate resuscitation.

Objective

Although many centers have introduced more restrictive transfusion policies for preterm infants in recent years, the benefits and adverse consequences of allowing lower hematocrit levels have not been systematically evaluated. The objective of this study was to determine if restrictive guidelines for red blood cell (RBC) transfusions for preterm infants can reduce the number of transfusions without adverse consequences.

Design, Setting, and Patients

We enrolled 100 hospitalized preterm infants with birth weights of 500 to 1300 g into a randomized clinical trial comparing 2 levels of hematocrit threshold for RBC transfusion.

Intervention

The infants were assigned randomly to either the liberal- or the restrictive-transfusion group. For each group, transfusions were given only when the hematocrit level fell below the assigned value. In each group, the transfusion threshold levels decreased with improving clinical status.

Main Outcome Measures

We recorded the number of transfusions, the number of donor exposures, and various clinical and physiologic outcomes.

Results

Infants in the liberal-transfusion group received more RBC transfusions (5.2 ± 4.5 [mean ± SD] vs 3.3 ± 2.9 in the restrictive-transfusion group). However, the number of donors to whom the infants were exposed was not significantly different (2.8 ± 2.5 vs 2.2 ± 2.0).

There was no difference between the groups in the percentage of infants who avoided transfusions altogether (12% in the liberal-transfusion group versus 10% in the restrictive-transfusion group). Infants in the restrictive-transfusion group were more likely to have intraparenchymal brain hemorrhage or periventricular leukomalacia, and they had more frequent episodes of apnea, including both mild and severe episodes.

Conclusions

Although both transfusion programs were well tolerated, our finding of more frequent major adverse neurologic events in the restrictive RBC-transfusion group suggests that the practice of restrictive transfusions may be harmful to preterm infants.

Periventricular leukomalacia (PVL) is a complication of prematurity that carries a high risk of long-term neurologic morbidity. We report the first case, to our knowledge, of unexpected PVL associated with in utero methamphetamine exposure in a 30-week gestation premature infant with a benign hospital course, who subsequently developed cerebral palsy by 24 months of life.

BACKGROUND:

Guidelines for prevention of group B streptococcal (GBS) infection have successfully reduced early onset (EO) GBS disease. Study results suggest that Escherichia coli is an important EO pathogen.

OBJECTIVE:

To determine EO infection rates, pathogens, morbidity, and mortality in a national network of neonatal centers.

METHODS:

Infants with EO infection were identified by prospective surveillance at Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Network centers. Infection was defined by positive culture results for blood and cerebrospinal fluid obtained from infants aged ≤72 hours plus treatment with antibiotic therapy for ≥5 days. Mother and infant characteristics, treatments, and outcomes were studied. Numbers of cases and total live births (LBs) were used to calculate incidence.

RESULTS:

Among 396 586 LBs (2006–2009), 389 infants developed EO infection (0.98 cases per 1000 LBs). Infection rates increased with decreasing birth weight. GBS (43%, 0.41 per 1000 LBs) and E coli (29%, 0.28 per 1000 LBs) were most frequently isolated. Most infants with GBS were term (73%); 81% with E coli were preterm. Mothers of 67% of infected term and 58% of infected preterm infants were screened for GBS, and results were positive for 25% of those mothers. Only 76% of mothers with GBS colonization received intrapartum chemoprophylaxis. Although 77% of infected infants required intensive care, 20% of term infants were treated in the normal newborn nursery. Sixteen percent of infected infants died, most commonly with E coli infection (33%).

CONCLUSION:

In the era of intrapartum chemoprophylaxis to reduce GBS, rates of EO infection have declined but reflect a continued burden of disease. GBS remains the most frequent pathogen in term infants, and E coli the most significant pathogen in preterm infants. Missed opportunities for GBS prevention continue. Prevention of E coli sepsis, especially among preterm infants, remains a challenge.

OBJECTIVE

Invasive candidiasis is a leading cause of infection-related morbidity and mortality in extremely low-birth-weight (<1000 g) infants. We quantify risk factors predicting infection in high-risk premature infants and compare clinical judgment with a prediction model of invasive candidiasis.

METHODS

The study involved a prospective observational cohort of infants <1000 g birth weight at 19 centers of the NICHD Neonatal Research Network. At each sepsis evaluation, clinical information was recorded, cultures obtained, and clinicians prospectively recorded their estimate of the probability of invasive candidiasis. Two models were generated with invasive candidiasis as their outcome: 1) potentially modifiable risk factors and 2) a clinical model at time of blood culture to predict candidiasis.

RESULTS

Invasive candidiasis occurred in 137/1515 (9.0%) infants and was documented by positive culture from ≥ 1 of these sources: blood (n=96), cerebrospinal fluid (n=9), urine obtained by catheterization (n=52), or other sterile body fluid (n=10). Mortality was not different from infants who had positive blood culture compared to those with isolated positive urine culture. Incidence varied from 2–28% at the 13 centers enrolling ≥ 50 infants. Potentially modifiable risk factors (model 1) included central catheter, broad-spectrum antibiotics (e.g., third-generation cephalosporins), intravenous lipid emulsion, endotracheal tube, and antenatal antibiotics. The clinical prediction model (model 2) had an area under the receiver operating characteristic curve of 0.79, and was superior to clinician judgment (0.70) in predicting subsequent invasive candidiasis. Performance of clinical judgment did not vary significantly with level of training.

CONCLUSION

Prior antibiotics, presence of a central catheter, endotracheal tube, and center were strongly associated with invasive candidiasis. Modeling was more accurate in predicting invasive candidiasis than clinical judgment.