Impact of study oximeter masking algorithm on titration of oxygen therapy in the Canadian oxygen trial.

OBJECTIVE: To compare oxygen saturations as displayed to caregivers on offset pulse oximeters in the 2 groups of the Canadian Oxygen Trial. STUDY DESIGN: In 5 double-blind randomized trials of oxygen saturation targeting, displayed saturations between 88% and 92% were offset by 3% above or below the true values but returned to true values below 84% and above 96%. During the transition, displayed values remained static at 96% in the lower and at 84% in the higher target group during a 3% change in true saturations. In contrast, displayed values changed rapidly from 88% to 84% in the lower and from 92% to 96% in the higher target group during a 1% change in true saturations. We plotted the distributions of median displayed saturations on days with >12 hours of supplemental oxygen in 1075 Canadian Oxygen Trial participants to reconstruct what caregivers observed at the bedside. RESULTS: The oximeter masking algorithm was associated with an increase in both stability and instability of displayed saturations that occurred during the transition between offset and true displayed values at opposite ends of the 2 target ranges. Caregivers maintained saturations at lower displayed values in the higher than in the lower target group. This differential management reduced the separation between the median true saturations in the 2 groups by approximately 3.5%. CONCLUSIONS: The design of the oximeter masking algorithm may have contributed to the smaller-than-expected separation between true saturations in the 2 study groups of recent saturation targeting trials in extremely preterm infants.

Reference values for respiratory events in overnight polygraphy from infants aged 1 and 3months.

OBJECTIVE: We aimed to determine reference values for respiratory indices in polygraphies (PGs) performed in infants aged 1 and 3months. METHODS: Healthy full-term neonates were recruited on the maternity ward. They were followed up by overnight PG at the age of 1month and again at the age of 3months. Indices of respiratory events, such as apneas, hypopneas, and percentage of periodic breathing were determined in each PG. Interpretation of PGs was performed blinded to the subject's data and the time of measurement. PG indices at 1 and 3months of age were compared. RESULTS: PG recordings were obtained for 37 healthy infants (22 boys). At the age of 1month, the median (minimum-maximum) central, obstructive, and mixed apnea index was 5.5 (0.9-44.3), 0.8 (0.1-6.7), and 0.3 (0-1.2), respectively. The same figures at the age of 3months were 4.1 (1.2-27.3), 0.8 (0-2.3), and 0.1 (0-0.8), respectively. Mixed obstructive apnea-hypopnea index (MOAHI) was 1.5 (0.2-7.0) and 0.9 (0.2-4.4) at the first and second measurements, respectively (P=.017). Only 1.2% of central apneas lasted longer than 20s. Periodic breathing was present in more than 90% of subjects studied. CONCLUSIONS: The infants in our study aged ⩽3months had respiratory event indices that were different from older children or adults. MOAHI showed a significant decrease during the first 3months after birth. We recommend that scoring of PG in infants of 3months or younger should consider age-specific reference values.

Hepcidin concentrations in serum and urine correlate with iron homeostasis in preterm infants.

OBJECTIVES: To evaluate whether hepcidin concentrations in serum (Hep((S))) and urine (Hep((U))) correlate with iron metabolism, erythropoiesis, and inflammation in preterm infants. STUDY DESIGN: Thirty-one preterm infants (23-32 weeks gestational age) were included. The concentration of the mature, 25 amino-acid form of hepcidin was determined by enzyme-linked immunosorbent assay in serum, urine, blood counts, reticulocytes, and iron measurements. RESULTS: Median (IQR) Hep((S)) was 52.4 (27.9-91.9) ng/mL. The highest values were measured in patients with systemic inflammation. Hep((S)) and Hep((U)) correlated strongly (P = .0007). Hep((S)) and Hep((U)) also correlated positively with ferritin (P = .005 and P = .0002) and with reticulocyte hemoglobin content (P = .015 and P = .015). Hep((S)) and Hep((U)) correlated negatively with soluble transferrin receptor/ferritin-ratio (P = .005 and P = .003). Infants with lower hemoglobin concentrations and higher reticulocyte counts had lower Hep((S)) (P = .0016 and P = .0089). CONCLUSION: In sick preterm infants, iron status, erythropoiesis, anemia, and inflammation correlated with the mature 25 amino-acid form of hepcidin. Further evaluation of Hep((U)) for non-invasive monitoring of iron status in preterm infants appears justified.

A randomized clinical trial of a new orthodontic appliance to improve upper airway obstruction in infants with Pierre Robin sequence.

OBJECTIVE: To test the hypothesis that a new orthodontic appliance with a velar extension that shifts the tongue anteriorly would reduce upper airway obstruction in infants with Pierre Robin sequence (PRS). STUDY DESIGN: Eleven infants with PRS (median age, 3 days) and an apnea index (AI) >3 were studied. The effect of the new appliance on the AI was compared with that of a conventional appliance without a velar extension by using a crossover study design with random allocation. RESULTS: Compared with baseline (mean AI, 13.8), there was a significant decrease in the AI with the new appliance (3.9; P value <.001), but no change with the conventional appliance (14.8; P = .842). Thus, the relative change in AI was -71% (95% CI, -84--49) for the new appliance and +8% (95% CI, -52-142) for the conventional appliance, which was significantly different (P = .004). No severe adverse effects were observed. CONCLUSION: This new orthodontic appliance appears to be safe and effective in reducing upper airway obstruction in infants with PRS.

Case-control study of symptoms and neonatal outcome of human milk-transmitted cytomegalovirus infection in premature infants.

OBJECTIVE: Preterm infants are at risk of acquiring human cytomegalovirus (CMV) infection through breast milk transmission, possibly leading to serious symptoms, as suggested by previous studies. Over a period of 8.5 years, we compared infants infected postnatally with CMV with noninfected controls to determine whether CMV infection transmitted through breast milk poses serious acute risks. STUDY DESIGN: CMV monitoring included maternal serologic testing and biweekly viral culture and polymerase chain reaction in breast milk and infant urine. Clinical and laboratory test findings were assessed retrospectively in infected infants and controls matched for gestational age during the initial hospital stay. RESULTS: Forty CMV-infected infants met the study criteria. They had lower minimal platelet and neutrophil counts and a higher frequency of C-reactive protein (CRP) elevations to 10 to 20 mg/L than their matched controls (P < or = .001). But no association of CMV infection with bronchopulmonary dysplasia, necrotizing enterocolitis, growth, or CRP elevations to > 20 mg/L was found. Cholestasis appeared in 3 infants in the CMV-infected group, but disappeared within 10 weeks. CONCLUSIONS: Neonatal symptoms related to postnatal CMV infection were transient and had no affect on neonatal outcome in these infants, in contrast with uncontrolled reports. Whether withholding or pasteurizing breast milk is warranted, however, depends on long-term outcome.

Bradycardia and desaturation during skin-to-skin care: no relationship to hyperthermia.

OBJECTIVE: We recently found increased temperature and increased bradycardia and desaturation during skin-to-skin care (SSC). We wanted to determine if these effects were related. STUDY DESIGN: Twenty-two infants (median gestational age at birth 28.5 weeks [range 24-31], median age at study 25.5 days [range 10-60 days], median birth weight 1025 g [range 550-1525 g], median weight at study 1320 g [range 900-2460 g]) underwent three 2-hour recordings of breathing movements, nasal airflow, heart rate, and pulse oximeter saturation (SpO 2 ): at thermoneutrality (TN) during incubator care, at TN during SSC, and at elevated temperature (ET) during incubator care. Core temperature was measured via a rectal probe. Recordings were analyzed for the summed rate of bradycardia and desaturation (heart rate <2/3 of baseline; SpO 2

Early feeding after necrotizing enterocolitis in preterm infants.

OBJECTIVE: To report our experience with an early initiation of enteral feedings after necrotizing enterocolitis (NEC). STUDY DESIGN: Over a 4-year period, all inborn infants with NEC Bell stage II or greater received enteral feedings, increased by 20 mL/kg/d, once no portal vein gas had been detected on ultrasound for 3 consecutive days (group 1). Infants were compared with a historic comparison group (group 2). RESULTS: Necrotizing enterocolitis rates were 5% (26/523) in the early feeding group and 4% (18/436) in the comparison group. One early feeding infant and two comparison group infants died of NEC, whereas two and one, respectively, had recurrent NEC. Enteral feedings were restarted at a median of 4 days (range, 3-14) versus 10 days (range, 8-22) after onset of NEC. Early feeding was associated with shorter time to reach full enteral feedings (10 days [range, 7-31] vs 19 days [range, 9-76], P<.001), a reduced duration of central venous access (13.5 days [range, 8-24] vs 26.0 days [range, 8-39], P<.01), less catheter-related septicemia (18% vs 29%, P<.01), and a shorter duration of hospital stay (63 days [range, 28-133] vs 69 days [range, 36-150], P<.05). CONCLUSION: Early enteral feeding after NEC was associated with significant benefits and no apparent adverse effects. This study was underpowered, however, to exclude a higher NEC recurrence risk potentially associated with this change in practice.

Influence of nasogastric tubes on gastroesophageal reflux in preterm infants: a multiple intraluminal impedance study.

To investigate whether a nasogastric tube predisposes to gastroesophageal reflux, 16 preterm infants underwent 48-hour recordings of multiple intraluminal impedance with the catheter tip in the lower esophagus or stomach for 24 hours each. There were 72 (range, 40-145) reflux episodes with the esophageal placement and 122 (range, 60-147) during the gastric position (P <.01).