Incidence of and Risk Factors for Perioperative Blood Transfusion During Index Neonatal Operations

Background: Blood transfusion in the neonatal period may be necessary due to blood loss during index neonatal operations. There is a paucity of data on the frequency of transfusion during these cases, and guidelines for pretransfusion testing prior to operation are not standardized. Furthermore, there is substantial financial cost associated with testing. At our institution, a type and cross costs $436.00 and a crossmatch costs $632.00. The purpose of this study is to determine the incidence of perioperative blood transfusions during index neonatal operations and identify risk factors associated with perioperative blood transfusion to determine which neonates would benefit from pretransfusion testing.

Methods: A retrospective review of infants who underwent index neonatal cases between 2013-2019 was performed. Data were collected for patients who underwent operations for Hirschsprung’s disease, esophageal atresia/tracheoesophageal fistula (EA/TEF), biliary atresia, anorectal malformation, omphalocele, gastroschisis, duodenal atresia, congenital diaphragmatic hernia (non-ECMO) or pulmonary lobectomy. Infants under 6 months were included except in the case of lobectomy where infants up to 12 months were included. Data collected included demographics, preoperative, operative, and postoperative transfusion details. Postoperative blood transfusion was defined as transfusion within 24 hours of operation. Prematurity was defined as prior to 37 weeks gestational age. Characteristics were stratified by transfusion status and compared using Fisher’s exact tests, Chi-squared tests, or T-tests depending on distribution. A logistic regression was performed using peri- or postoperative transfusion as the outcome adjusting for prematurity, structural heart disease, and preoperative transfusion. Probabilities for all combinations of risk factors were calculated with 95% confidence intervals from the logistic regression.

Results: Analysis was performed on 420 patients. A total of 313 (74.5%) had at least one preoperative type and screen and 144 (34.2%) underwent preoperative crossmatch. The average number of type and screens per patient was 1.4 ± 0.9 (mean ± standard deviation) and the average number of crossmatches was 0.5 ± 0.8. Twenty-five (6.0%) patients received intraoperative (n=4), postoperative (n=20) transfusion, or both (n=1). The majority of patients who received perioperative transfusion underwent EA/TEF repair (32.0%). No patients who underwent surgery for Hirschsprung’s disease or anorectal malformation received perioperative transfusion. Congenital heart disease (CHD), preoperative transfusion, and prematurity were associated with perioperative transfusion (Table 1). Presence of all three risk factors resulted in a 48% probability of requiring perioperative transfusion (Table 2).

Conclusion: Blood transfusion during the intraoperative and postoperative period of neonatal index operation is rare, but rates of pretransfusion testing are high. Type of operation, presence of CHD, prematurity, and preoperative transfusion are associated with perioperative transfusion. Institutions can consider these factors when establishing guidelines for preoperative testing as opportunities for preventing unnecessary testing and subsequent cost savings exist.