Variations in transfusion practice in neonatal intensive care

Area of Science:

• Neonatal Medicine
• Pediatric Hematology
• Clinical Practice Variation

Background:

• Transfusion practices for very low birth weight (VLBW) infants differ across neonatal intensive care units (NICUs).
• Understanding these variations is crucial for optimizing patient care and resource utilization.

Purpose of the Study:

• To compare packed red blood cell transfusion practices between two NICUs.
• To assess the impact of local practices on transfusion timing, volume, and number in VLBW infants.
• To develop models for characterizing and improving transfusion and phlebotomy practices.

Main Methods:

• Retrospective review of 270 VLBW infant admissions (<1500 g) in two NICUs (A and B).
• Analysis of phlebotomy losses and transfusion rates, stratified by birth weight and illness severity (Score for Neonatal Acute Physiology - SNAP).
• Development of multivariate models to describe and compare NICU practices.

Main Results:

• NICU A had lower phlebotomy losses and a lower percentage of infants receiving transfusions compared to NICU B.
• NICU A transfused a greater total volume per kg, with different timing patterns than NICU B.
• Phlebotomy losses and transfusion volumes were significantly associated with gestational age (GA) and SNAP; NICU B had higher losses and transfusions for comparable infants.
• No significant differences in short-term outcomes (oxygen use, ventilation, chronic lung disease, growth) were observed between NICUs.

Conclusions:

• Significant differences in phlebotomy and transfusion practices exist between NICUs, with no apparent benefit to increased transfusion in one unit.
• Multivariate models can quantify practice variations and identify patients for targeted interventions like erythropoietin therapy.
• These models offer a tool for monitoring practice changes and potentially reducing donor exposure in VLBW infants.