Macronutrient content of pasteurised donor human milk: Variability between batches from single-donor pools at an Australian milk bank.

AIM: This study aimed to characterise the between-batch variability of pasteurised donor human milk (PDHM) produced from single-donor pools at Australian Red Cross Lifeblood's milk bank and identify key donor characteristics that predict macronutrient content. METHODS: Macronutrient content from 200 batches of PDHM was measured using a mid-infrared human milk analyser (Miris, Uppsala, Sweden). Linear mixed models were used to study the impact of stage of lactation and gestational age on macronutrient content. Coefficients of determination (R2 ) were calculated to estimate the impact of the individual donor on overall variability. RESULTS: Macronutrient content of PDHM varied considerably, with between-batch variations of 2.8 and 6.4-fold for protein and fat content, respectively. Mean crude protein content was 1.16 g/100 mL, ranging from 0.7 to 1.96 g/100 mL. Mean fat content was 3.85 g/100 mL, ranging from 1.46 to 9.39 g/100 mL. Stage of lactation was identified as a predictor for protein content and gestational age at birth for fat content. Individual donor effect explained 55 and 35% of the variance for fat and protein content, respectively. CONCLUSIONS: This study highlights the variation in macronutrient content in PDHM at an Australian milk bank. Variability could be reduced through the implementation of targeted multiple-donor pooling using the key donor characteristics identified in this study along with the measurement of macronutrient content of individual donors at the time of first donation. However, the clinical benefit of a reduction in between-batch variation, achieved through multiple-donor pooling, would need to be assessed to justify additional efforts associated with PDHM processing changes.

Donor and recipient safety in human milk banking.

AIM: Australian Red Cross Lifeblood supplies pasteurised donor human milk (PDHM) to more than 30 partner hospitals across Australia. Preterm infants who receive PDHM are a highly vulnerable population but formal biovigilance programs are rare in human milk banking. Lifeblood Milk performs ongoing surveillance for both donor and recipient adverse events. This study aimed to formally review adverse events reported to Lifeblood Milk since 2018. METHODS: Milk donor infectious diseases testing outcomes and donor adverse events (DAEs) are prospectively recorded at Lifeblood. Infant recipient adverse events are contractually reported back to Lifeblood Milk by hospitals and assessed according to severity and likelihood of relationship to PDHM administration. Donor and recipient adverse events over a 3.5-year period (July 2018 to December 2021) were reviewed. RESULTS: There were three DAEs (3/976 = 0.31%) related to phlebotomy; these included two vasovagal reactions and one phlebotomy site haematoma. Eight (8/976 = 0.81%) additional donors had biological false reactive (BFR) infectious diseases serology results. There were 10 reported suspected adverse events in recipients. Six were infection-related; other events included milk curd obstruction, high urinary iodine levels, sudden cardiac death and nasogastric tube obstruction. All reported suspected adverse events in recipients were classified as unlikely to be related, or definitely not related, to PDHM administration. CONCLUSIONS: Milk donor adverse events were rare but biological false reactive serology results were not uncommon. There were no recipient adverse events considered causally related to pasteurised donor human milk, which is generally a low-risk biological product. Ongoing biovigilance remains essential.

What are Optimal Bacteriological Screening Test Cut-Offs for Pasteurized Donor Human Milk Intended for Feeding Preterm Infants?

BACKGROUND: Definitive criteria for microbial screening of pasteurized donor human milk are not well established and international recommendations vary. AIMS: (1) To review pasteurized donor human milk batch discard due to failed microbial screening criteria at our milk bank (following United Kingdom National Institute of Clinical Excellence guidelines), and (2) to compare our known milk discard proportion with estimated milk discard proportions that would be required by other international milk bank guidelines. METHODS: We reviewed our microbial screening results (N = 783) over 18-months (July 2018-December 2019) and compared our known milk discard proportion with estimated milk discard proportions using other international milk bank guidelines. RESULTS: Of samples, n = 50 (6.4%) failed pre-pasteurization screening, most commonly due to the presence of >104 CFU/mL Enterobacterales in the pre-pasteurization sample (n = 30; 3.8%). Two (0.3%) samples failed post-pasteurization screening, with Bacillus cereus identified in both cases, resulting in total discard proportion of 6.7% (n = 52) of batches. Applying European Milk Bank Association recommended bacterial screening criteria, approximately 23.3% (n = 183) of milk batches would have been discarded. CONCLUSIONS: Further research is required to justify the stringent European Milk Bank Association recommendations for pre-pasteurization discard criteria, although we believe that a post-pasteurization acceptance criterion of <1 CFU/mL is appropriate and aligns with international guidance. Further work is needed to understand pasteurized donor human milk microbiological safety risks, to better integrate screening criteria within current food standards regulation, and to consider risk-based assessment including the impact on availability and affordability.

Development and evaluation of formal guidelines for donor selection for human milk banks.

AIM: Donor selection for milk banks is essential to ensure the safety and nutritional quality of the donor milk, and to ensure that the prospective donor and her breastfeeding infant do not come to harm through donating. Australian Red Cross Lifeblood Milk went through a robust process to develop a set of criteria for the selection and screening of potential breast milk donors, which included development of a Donor Questionnaire (DQ), supported by a formal set of Guidelines for the Selection of Milk Donors. Key screening questions from the DQ were made available to prospective donors to self-screen prior to the formal assessment process. The aim of this study was to review the outcomes of milk donor screening at Lifeblood Milk. METHODS: We reviewed the outcomes of our donor screening process over the first 12-months (July 2018-June 2019) of operations. RESULTS: A total of 50 out of 327 donors who responded to the self-screening questions were not able to proceed further; 201 donors were formally screened using the DQ and Guidelines for the Selection of Milk Donors, with 9 of 201 deferred based on their responses. An additional two donors were deferred (failed phlebotomy (n = 1) and reactive infectious disease serology (n = 1)), with 190 of 201 (95%) of prospective donors accepted after screening. CONCLUSIONS: Our experience highlighted international differences in practice between milk banks and lack of strong research to inform milk donor selection. Making a set of key screening questions available to donors for self-screening resulted in a high acceptance rate (95%) for donors who began the formal screening process. Further work is needed to better understand the impact of deferral on prospective milk donors.