Why do four NICUs using identical RBC transfusion guidelines have different gestational age-adjusted RBC transfusion rates?

OBJECTIVE: To compare neonatal red blood cell (RBC) transfusion rates in four large Intermountain Healthcare NICUs, all of which adhere to the same RBC transfusion guidelines. STUDY DESIGN: This retrospective analysis was part of a transfusion-management quality-improvement project. De-identified data included RBC transfusions, clinical and laboratory findings, the anemia-prevention strategies in place in each NICU, and specific costs and outcomes. RESULT: Of 2389 NICU RBC transfusions given during the 4-year period studied, 98.9 ± 2.1% (mean ± S.D.) were compliant with our transfusion guidelines, with no difference in compliance between any of the four NICUs. However, RBC transfusion rates varied widely between the four, with averages ranging from 4.6 transfusions/1000 NICU days to 21.7/1000 NICU days (P < 0.00001). Gestational age-adjusted transfusion rates were correspondingly discordant (P < 0.00001). The lower-transfusing NICUs had written anemia-preventing guidelines, such as umbilical cord milking at very low birth weight delivery, use of cord blood for admission laboratory studies, and darbepoetin dosing for selected neonates. Rates of Bell stage ⩾ 2 necrotizing enterocolitis and grade ⩾ 3 intraventricular hemorrhage were lowest in the two lower-transfusing NICUs (P < 0.0002 and P < 0.0016). Average pharmacy costs for darbepoetin were $84/dose, with an average pharmacy cost of $269 per transfusion averted. With a cost of $900/RBC transfusion, the anemia-preventing strategies resulted in an estimated cost savings to Intermountain Healthcare of about $6970 per 1000 NICU days, or about $282,300 annually. CONCLUSION: Using transfusion guidelines has been shown previously to reduce practice variability, lower transfusion rates and diminish transfusion costs. Based on our present findings, we maintain that even when transfusion guidelines are in place and adhered to rigorously, RBC transfusion rates are reduced further if anemia-preventing strategies are also in place.

Causes of hemolysis in neonates with extreme hyperbilirubinemia.

OBJECTIVE: We instituted a quality improvement process to enhance our capacity to diagnose genetic hemolytic conditions in neonates with extreme hyperbilirubinemia. STUDY DESIGN: During a 1-year period, whenever the total serum bilirubin (TSB) was >25 mg dl(-1) a special evaluation was performed. If we deemed an erythrocyte membrane defect likely, based on red blood cell morphology, EMA-flow cytometry was performed. Otherwise 'next-generation' sequencing was performed using a panel of genes involved in neonatal hyperbilirubinemia. RESULT: Ten neonates had a TSB ⩾ 25 mg dl(-1). Two others were evaluated as part of this process at the request of their attending neonatologists, because each had a TSB >14 mg dl(-1) in the first hours after birth and required phototherapy for ⩾ 1 week. Explanations for the jaundice were found in all 12 neonates. Five had hereditary spherocytosis, three of which also had ABO hemolytic disease. Two had pyruvate kinase deficiency. One had severe G6PD deficiency. The other four had ABO hemolytic disease. CONCLUSION: On the basis of the present small case series, we suggest that among neonates with extreme hyperbilirubinemia, it can be productive to pursue a genetic basis for hemolytic disease.

Pyruvate kinase deficiency as a cause of extreme hyperbilirubinemia in neonates from a polygamist community.

Neonatal hemolytic jaundice is a risk factor for kernicterus. Pyruvate kinase (PK) deficiency is a rare cause of neonatal hemolytic jaundice, with a prevalence estimated at 1 case per 20,000 births in the United States, but with a higher prevalence among the Amish communities in Pennsylvania and Ohio. We discovered four neonates with PK deficiency born in a small community of polygamists. All four had early, severe, hemolytic jaundice. PK deficiency should be considered in neonates with early hemolytic, Coombs-negative, non-spherocytic jaundice, particularly in communities with considerable consanguinity. Such cases should be recognized early and managed aggressively to prevent kernicterus.

Adherence to NICU transfusion guidelines: data from a multihospital healthcare system.

OBJECTIVE: We critically reviewed every NICU blood component transfusion (packed erythrocytes, platelets, frozen plasma (FP) and cryoprecipitate) administered during a one-year period. This was done to determine the proportion of transfusions given out of compliance with the Intermountain Healthcare transfusion guidelines, and to look for patterns of non-compliance that could be addressed by quality improvement measures. STUDY DESIGN: A detailed review was made of every transfusion administered to patients with a date of birth of 1 January 2006 through 31 December 2006, in any of three level III, perinatal-center-associated NICUs within Intermountain Healthcare. RESULT: During 2006 the three NICUs cared for 1759 neonates. Seventeen percent of these received one or more (median 3) erythrocyte transfusions, 4% received one or more (median 3) platelet transfusions, 6% received one or more (median 1) FP infusions and 2% received cryoprecipitate (median 1 dose). Seventy percent of the erythrocyte transfusions were given in compliance with the guidelines, as were 69% of the platelet transfusions, 65% of the FP transfusions and 94% of the cryoprecipitate administrations. Patients who received large numbers of transfusions were more likely to receive transfusion that violated the guidelines. Forty-five percent of patients who received 1 to 3 transfusions received all transfusions within guidelines. However, only 18% of patients who received 4 to 10 transfusions received all within guidelines. No patient who received >10 transfusions received all within the guidelines. Erythrocyte transfusions given early in the hospital course were likely to be within guidelines; 72% (588/818) in the first 29 days were compliant with guidelines, but compliance fell to 61% (144/237) for transfusions administered after 29 days (P=0.002). About half of the platelet transfusions given early in the hospital course were in violation of guidelines, but after day 9, 83% of platelet transfusions were compliant with guidelines (P=0.000). CONCLUSION: Opportunities exist in our healthcare system to improve compliance with our transfusion guidelines. Such opportunities are greatest among neonates receiving multiple transfusions, among those receiving erythrocyte transfusions late in their NICU course and among those receiving platelet transfusions early in their NICU course.

Evaluation of transthyretin as a monitor of protein-energy intake in preterm and sick neonatal infants.

In the past, weight and weight gain have been the two parameters used frequently in neonatal units to monitor nutrition among high-risk infants. Our investigation sought to assess how several anthropometric measures (weight, length, head circumference, arm:head circumference ratio, and tricep skinfold) and serum albumin, transthyretin, and transferrin concentrations reflect protein and energy intake. After monitoring 42 preterm and 40 sick infants over 3 consecutive weeks, we found that transthyretin concentration proved the only serum protein which accurately measured energy and protein intakes in less than 1 week from dietary manipulation. Among preterm infants, as protein and energy intakes rose, transthyretin concentration increased significantly (p less than 0.001). Preterm infants ingested 79 +/- 39 kcal/kg/day and 2.04 +/- 1.02 g protein/kg/day at the first assessment point after birth and rose to 103 +/- 34 kcal/kg/day and 2.64 +/- 0.94 g protein/kg/day. Likewise, transthyretin measured 10.83 +/- 3.91 mg/dl at the initial measurement and rose to 14.80 +/- 4.44 at the second measurement time. Although protein intakes measured slightly lower in the sick group, their intakes correlated to transthyretin concentration (assessment time 1, r = 0.39; time 2, r = 0.33; time 3, r = 0.33). Thus, transthyretin concentration in neonatal infants offers a rapid, accurate, and moderately inexpensive way to monitor protein-energy adequacy.

A simplified method of obtaining umbilical cord blood for pH using a heparinized vacutainer versus heparinized syringe.

Two experimental methods for collecting cord blood for acid-base analysis using green top vacutainer tubes were compared with a standard method using heparinized syringes. Paired samples from 30 newborn infants were collected using one of the vacutainer methods and the standard heparinized syringe method. No significant differences in pH, pCO2, base deficit, or pO2 were noted when comparing the two vacutainer methods with the standard syringe method. Both vacutainer methods were substantially easier to use than the heparinized syringe method. A green top vacutainer may be used to obtain an accurate cord blood sample for assessment of an infant's acid-base status at birth, and to aid the neonatologist as an indicator of intrapartum asphyxia.

Home phototherapy treatment of neonatal jaundice.

A home phototherapy program for healthy, term newborns with hyperbilirubinemia (n = 62) was implemented, and results were compared prospectively with a group of term neonates who met enrollment criteria but were treated in-hospital (n = 55). The purpose of the study was to prospectively assess the feasibility, safety, and effectiveness of home phototherapy in treating uncomplicated neonatal jaundice. Infants were required to be greater than 24 hours old and to weigh at least 2,270 g (5 lb). Enrolling physicians were instructed to select infants whose clinical diagnoses and bilirubin levels allowed an adequate margin of error for a trial of home phototherapy and whose parents were capable of managing the added responsibilities of home therapy. At the time the phototherapy equipment was delivered and set up in their home, parents received extensive instruction, including how to record pertinent ongoing data on a home flow sheet. Serum bilirubin levels were measured at least daily. The preponderant diagnosis listed by enrolling physicians was "physiologic jaundice." No parent reported significant complications, and no infant required rehospitalization. Bilirubin levels decreased as rapidly in the home group as in the hospitalized control group, and duration of treatment averaged 2.8 days. About $18,000 was saved by treating these 62 infants at home compared with in-hospital costs in our community. We found home phototherapy to be a feasible, safe, and effective alternative to in-hospital phototherapy for otherwise healthy, jaundiced infants with motivated and capable parents.

New approach to management of unilateral tension pulmonary interstitial emphysema in premature infants.

Conservative management of unilateral tension pulmonary interstitial emphysema is reported. Treatment consisted of (1) positioning the infant on his or her side with the emphysematous lung dependent, (2) minimal chest physiotherapy and endotracheal suctioning, and (3) appropriate ventilator management. Conservative therapy is effective and appears to be safer than selective bronchial intubation or surgical therapy.

Calcium and phosphorus compatibility in parental nutrition solutions for neonates.

Factors affecting calcium-phosphate solubility in parenteral nutrition solutions used in neonates were studied. Six neonatal parenteral nutrition solutions were prepared using either Aminosyn or FreAmine III and various amino acid and dextrose concentrations. Phosphorus (as mono- and dibasic potassium phosphate) and calcium (as 10% calcium gluconate) were added in concentrations of calcium 2.5-100 meq/liter and phosphorus 2.5-100 mmol/liter. Duplicate samples were prepared and analyzed either after they were heated in a water bath (37 degrees C) for 20 minutes or after 18 hours at 25 degrees C followed by 30 minutes in a water bath (37 degrees C). Precipitation was detected visually and spectrophotometrically, and pH was measured. Lipid emulsion was added to two Fre-Amine III solutions in a ratio of 7.5:1 (parenteral nutrition solution:lipid) and the resulting pH was measured. Time and temperature affected calcium-phosphate solubility in all solutions tested. Precipitation curves of amount of calcium versus amount of phosphate added were prepared for each solution. Amino acid and dextrose concentrations affected the pH of the solutions, and when a lipid emulsion was added, the pH rose more in the 1% than in the 2% FreAmine III solution. In selected solutions, as much as 120 mg/kg/day calcium and 55 mg/kg/day phosphate can be administered, approximating daily third-trimester accumulation of these minerals. Use of the precipitation curves in this paper, with attention to their limitations, should aid in the safe delivery of calcium and phosphorus intravenously to neonates.

Surgical treatment of patent ductus arteriosus in preterm infants. Four-year experience with ligation in the newborn intensive care unit.

Seventy-nine consecutive patent ductus arteriosus (PDA) ligations were performed in our newborn intensive care unit (NBICU) over a four-year period. Seven infants (9%) died within one month after surgery. The deaths were not attributed to the surgery but were associated with preexisting medical complications or new intracranial hemorrhages. Significant surgical morbidity was rare. We conclude that ligation of the PDA in the premature infant can be a safe and effective procedure, when pharmacologic closure is ineffective or contraindicated. Furthermore, PDA ligation may be safely performed in the NBICU.