High vs low CPAP strategy with aerosolized calfactant in preterm infants with respiratory distress syndrome.

BACKGROUND: Optimal CPAP strategy to prevent CPAP failure defined as need for endotracheal intubation is unknown. OBJECTIVE: To evaluate the risk of CPAP failure in infants treated with high vs low CPAP strategy while receiving aerosolized calfactant in the AERO-02 clinical trial and AERO-03 expanded access program. METHODS: Infants born between 29 0/7 to 36 6/7 weeks were included. Comparisons were made between low and high CPAP groups (Low, 4-7 cm H2O; High, 8-10 cm H2O). RESULTS: CPAP failure and pneumothorax were not different between the groups. Odds of CPAP failure were not different after adjustment for baseline characteristics (OR = 0.61; 95% CI: 0.29, 1.24). CONCLUSION: We found no difference in CPAP failure among infants who received aerosolized calfactant that were treated with high vs low CPAP strategy. Efficacy of high CPAP strategy with aerosolized surfactant treatment needs to be evaluated in future studies.

Recombinant tissue plasminogen activator to restore catheter patency: efficacy and safety analysis from a multihospital NICU system.

OBJECTIVE: In 2001, the US Food and Drug Administration approved recombinant tissue plasminogen activator (alteplase, Cathflo Activase) to reestablish patency of central catheters occluded, presumably, by a fibrin clot. We conducted a multicenter quality improvement study to determine the value of this procedure in our Neonatal Intensive Care Unit (NICUs), including analyses of efficacy, safety and costs. STUDY DESIGN: We conducted a retrospective quality analysis of neonates in level III NICUs, who received alteplase for the purpose of reestablishing patency of occluded central catheters. RESULTS: Alteplase was administered to 169 neonates, each given one to four doses, totaling 205 episodes of administration. The most common type of catheter where alteplase was used was percutaneously inserted central catheter (PICC) lines (78% of uses), 8% were umbilical venous catheters (UVCs), 6% arterial lines, 5% chest tubes and 3% other catheters. Postnatal age at first dose ranged from 0 to 132 days (median, 12); dosed patients were 22 to 41 weeks gestation at birth (median, 31). Fifty-eight percentage of administrations restored catheter function. Success was more likely at younger postnatal age (10±2 days old in successful vs 14±1 days in unsuccessful treatments; P=0.023). Seventy-two percentage of the re-canalized catheters remained functional until they were no longer needed (2 to 30 days later). Nine percentage of episodes were treated with a second dose 1 to 17 days later for re-occlusion and 50% of those were successful. Bleeding consequences were identified in only one case, where three separate lines were treated (chest tube, PICC and UVC) within a 6-h period. Costs to the health system of doses, minus savings to the system by not needing to replace lines, averaged a net of $34 per dose. CONCLUSIONS: The apparent safety and favorable value analysis prompted us to develop a consistent approach to alteplase usage in the Intermountain Healthcare NICUs, using the data in this report to standardize the guidelines across our health system.

Elevated fecal calprotectin levels during necrotizing enterocolitis are associated with activated neutrophils extruding neutrophil extracellular traps.

OBJECTIVE: Neonates with necrotizing enterocolitis (NEC) have higher calprotectin levels in stool than do healthy neonates. However, it is not known whether high stool calprotectin at the onset of bowel symptoms identifies neonates who truly have NEC vs other bowel disorders. STUDY DESIGN: Neonates were eligible for this study when an x-ray was ordered to 'rule-out NEC'. Stool calprotectin was quantified at that time and in a follow-up stool. Each episode was later categorized as NEC or not NEC. The location of calprotectin in the bowel was determined by immunohistochemistry. RESULTS: Neonates with NEC had higher initial and follow-up stool calprotectin levels than did neonates without NEC. Calprotectin in bowel from neonates with NEC was within neutrophil extracellular traps (NETs). CONCLUSION: At the onset of signs concerning for NEC, fecal calprotectin is likely to be higher in neonates with NEC. Calprotectin in their stools is exported from neutrophils via NETs.

Whole-blood viscosity in the neonate: effects of gestational age, hematocrit, mean corpuscular volume and umbilical cord milking.

OBJECTIVE: The American College of Obstetrics and Gynecology Committee on Obstetric Practice recently endorsed delayed cord clamping at preterm delivery. However, the committee report expressed the concern by some practitioners that delayed clamping or cord milking might induce hyperviscosity in preterm neonates. To address this issue we: (1) established reference ranges for whole-blood viscosity among preterm neonates (viscosity reference ranges had previously been reported only in term neonates) and (2) determined the effect of umbilical cord milking at deliveries <32 weeks gestation on subsequent blood viscosity measurements. STUDY DESIGN: This was a prospective study in two Neonatal Intensive Care Units. Blood viscosity was measured using a cone and plate viscometer. Associations were sought with gestation, hematocrit/hemoglobin and mean corpuscular volume. Reference ranges were determined for preterm infants <32 weeks gestation. Then, after umbilical cord milking at deliveries <32 weeks, viscosity was measured at birth and again during the 12 h after birth. In neonates with viscosities >95th % range, we sought signs of hyperviscosity (plethora, hypotonia, hypoglycemia, hyperbilirubinemia, thrombocytopenia). RESULT: Viscosity at higher and lower sheer rates were linearly related (n=32, r=0.971). Within the range of hematocrits measured (29-63%) viscosity correlated with hematocrit (r=0.877) and hemoglobin (r=0.853) but not with erythrocyte size (r=0.179). Viscosity was related to gestational age (n=58), primarily due to the lower hematocrits at lower gestational ages. In the 12 h after cord milking viscosity ranged from 3.1 to 9.5 centipoise. Three of twenty preterm, neonates had viscosities >95th % reference range. However, all values were well below those where hyperviscosity is defined in term neonates and all lacked features of hyperviscosity. CONCLUSION: Cord blood viscosity is directly proportional to hematocrit/hemoglobin, lower at early gestation and not associated with erythrocyte size. Cord milking at preterm delivery is associated with a low risk of clinical hyperviscosity. Practioners should not refrain from cord milking at preterm delivery because of a concern that it will commonly cause neonatal hyperviscosity.

Using umbilical cord blood for the initial blood tests of VLBW neonates results in higher hemoglobin and fewer RBC transfusions.

OBJECTIVE: We previously described a method for reducing early phlebotomy losses from very low birth weight (VLBW) neonates by obtaining the initial blood tests from otherwise discarded fetal blood from the placenta. In the present study we sought to; (1) measure the feasibility of performing this method in actual practice, (2) test the hypothesis that this method would result in higher hemoglobin concentrations and lower erythrocyte transfusion rates in the first week after birth. METHODS: We conducted two studies in three Intermountain Healthcare NICUs. The first was a feasibility analysis involving 96 VLBW neonates, measuring the success of obtaining the NICU admission laboratory blood tests this way. The second study used case-control methodology to test the hypothesis that this method would result in a higher blood hemoglobin 12 to 24 h after birth, and a lower proportion receiving an erythrocyte transfusion in the first week. RESULT: In 91 of 96 VLBW neonates (95%) the initial blood tests were successfully obtained with this method. The success rate was not diminished by delayed cord clamping or cord milking, as it was successful in 35 of 36 (97%) such instances. Cases and controls were well matched on demographic and level of illness comparisons. Among cases the hemoglobin generally increased between birth and 12 to 24 h later, but among controls the hemoglobin generally decreased (P<0.05). In the week following birth fewer cases received vasopressors (P<0.01) and erythrocyte transfusions (P<0.001). CONCLUSION: We judge that it is feasible to collect the initial blood tests of VLBW neonates using otherwise discarded umbilical cord/placental blood, in that this can be accomplished in about 95% of VLBW deliveries. This method, which can be used in addition to either delayed clamping of the umbilical cord or cord milking, results in higher hemoglobin concentrations, less vasopressor use and fewer transfusions in the first week.

Fulminant necrotizing enterocolitis in a multihospital healthcare system.

OBJECTIVE: A subset of necrotizing enterocolitis (NEC) cases is fulminant, characterized by rapid progression to death with massive bowel necrosis found at laparotomy or autopsy. We sought to identify and report all such cases in a multihospital healthcare system during the past 9 years and to characterize this entity using case-control methodologies. STUDY DESIGN: This was a multicentered, cross-sectional, historic cohort study conducted using Intermountain Healthcare hospital patient data. All neonates who died of NEC within 48 h of onset, during 2001 to 2009, were compared with two matched control groups: (1) demographically matched controls who developed non-fulminant NEC, (2) demographically matched controls that did not develop NEC. RESULT: During this period, 2 71 327 live births occurred in the Intermountain Healthcare hospitals. Of these, 318 had a diagnosis of NEC (Bell stage ≥II). Also during this period, 205 other neonates were transferred into an Intermountain hospital for treatment of NEC. Of these 523 NEC cases, 35 (6.7%) had a fulminant course. Compared with the non-fulminant cases, the fulminant group were born at lower weight (1088±545 vs 1652±817 g, P=0.000) and earlier gestational age (27.5±3.3 vs 31.1±4.4 weeks, P=0.000), and were more likely to have: (1) radiographic evidence of portal venous air (P=0.000), (2) hematocrit <22% (P=0.000), (3) increase in feeding volume >20 ml/kg/day (P=0.003), (4) immature to total (I/T) neutrophil ratio >0.5 (P=0.005), (5) blood lymphocyte count <4000/µl (P=0.018), (6) an increase in concentration of human milk fortifier within 48 h before developing NEC (P=0.020). CONCLUSION: Portal venous air, anemia, rapid feeding escalation, a high I/T neutrophil ratio, a low lymphocyte count and recent increases in fortifier may all be associated with fulminant NEC.

Neonates presenting with bloody stools and eosinophilia can progress to two different types of necrotizing enterocolitis.

OBJECTIVE: We hypothesized that neonates with bloody stools and concomitant eosinophilia are likely to have atopic enteropathy rather than necrotizing enterocolitis (NEC). STUDY DESIGN: This was a retrospective cross-sectional study using electronic medical records and paper charts. Records of neonates admitted to any Intermountain Healthcare NICU between 1 January 2005 and 30 June 2010 were eligible if 'bloody stools' were listed in any archive. Qualifying records were divided into two groups depending on whether or not within 72 h of passing bloody stool eosinophil counts were above the 95th percentile reference range limit for age. RESULT: Bloody stools were identified in 275 predominantly Caucasian neonates. Fifty-four of these had eosinophilia and 221 had normal eosinophil counts. Those with eosinophilia were born at a slightly younger gestational age (31.3 ± 4.6 vs 32.6 ± 4.0 weeks, mean ± s.d., P=0.032). Contrary to our hypothesis, those with eosinophilia did not have a lower rate of pneumatosis or bowel resection, or death ascribed to NEC. Eosinophilia was more common among those who had a red blood cell (RBC) transfusion within 48 h before passing bloody stools (P<0.001). Those with a recent RBC transfusion were the only neonates to have NEC surgery or to die from NEC. Preceding the bloody stools, those with no antecedent transfusion had been fed a larger volume (P=0.014), and had trends toward receiving calorically enriched feedings (P=0.055) and recent addition of human milk fortifier (P=0.060). Eosinophil counts following RBC transfusion tended to increase for 3-6 days, but when bloody stools were not preceded by transfusion the eosinophil counts were more static over that period. CONCLUSION: In this predominantly Caucasian group of neonates with bloody stools, the presence of eosinophilia did not identify a benign condition distinct from NEC. A total of 44% of these neonates had transfusion-associated NEC. Eosinophils could have a previously unrecognized role in the pathogenesis of this NEC subtype.

Antecedents of Bell stage III necrotizing enterocolitis.

OBJECTIVE: New biopharmaceuticals hold promise for preventing or treating necrotizing enterocolitis. However, it is unclear whether any such biopharmaceutical that requires enteral administration could be administered using an 'early-treatment' paradigm. This study was undertaken to assess this issue based on data from every case of Bell stage III NEC cared for during the past 7 years at Intermountain Healthcare. STUDY DESIGN: Patients with Bell stage III NEC were identified from electronic medical record repositories and the diagnosis was validated using operative reports. Electronic and paper records of each patient were then used to identify potential clinical and laboratory antecedents occurring within the 48 h period preceding the diagnosis of NEC. RESULT: One hundred eighteen patients had Stage III NEC. The earliest recognized antecedents were nonspecific for NEC (apnea/bradycardia, skin mottling and irritability). These were recorded at 2.8+/-2.1, 4.5+/-3.1 and 5.4+/-3.7 (mean+/-s.d.) hours, respectively, before NEC was diagnosed. The most commonly identified gastrointestinal antecedents were blood in the stools, increased abdominal girth and elevated pre-feeding gastric residuals or emesis. These were identified 2.0+/-1.9, 2.8+/-3.1 and 4.9+/-4.0 h before NEC was recognized. Thirty-eight percent had a blood transfusion (18+/-12 h) preceding the NEC. Tachycardia, tachypnea, hypotension and diarrhea were rarely identified as antecedents and no consistent laboratory antecedents were discovered. CONCLUSION: We judge that an 'early treatment of NEC' paradigm testing any pharmacological agent that must be administered enterally is not feasible. The first recognized antecedents of Bell stage III NEC are nonspecific for gastrointestinal pathology and insufficient time exists for dosing between the first gastrointestinal signs and placement of the gastric decompression tube.