Aim To characterise the current transfusion practice among clinicians in Australasian neonatal units and factors that influence their decision‐making. Methods Members of the Australia and New Zealand Neonatal Network (ANZNN) and practitioners at their local institutions were invited to participate in a 15‐question web‐based survey between 1 June and 31 July 2016. The survey was designed to assess (i) haemoglobin‐based transfusion thresholds; (ii) presence of local guidelines; (iii) preference for a restrictive or liberal transfusion policy; (iv) perceived benefits and risks of transfusion; and (v) use of clinical adjuncts to assist decision‐making. Results Overall, 130 participants responded to at least one question. Haemoglobin transfusion thresholds for anaemia of prematurity (AOP) varied significantly from <60 to <120 g/L. Of 103 participants, 36 (35%) reported that they do not have access to local transfusion guidelines. The majority utilise multiple clinical and haematological parameters to guide their decision‐making, and approximately half (45/84, 54%) believe that tissue hypoxia detected by near‐infrared spectroscopy (NIRS) may better inform transfusion thresholds. Of 102 participants, 51 (50%) support a restrictive rather than liberal transfusion policy. The most commonly reported perceived risks of transfusion for AOP were suppression of endogenous erythropoiesis and increased rates of necrotising enterocolitis. Conclusions There is a significant variation in transfusion practice in Australasian neonatal units. Quality and safety initiatives may assist with improved consistency of transfusion practice across the ANZNN. However, further research is required to better define optimal transfusion thresholds, quantify potential risks of transfusion and determine clinical utility of newer adjuncts such as NIRS.

Introduction Animal studies have shown that glucagon like peptide-2 (GLP-2) may reduce mucosal infl ammation; it decreases proinfl ammatory cytokines and ameliorates chronic colitis 1 . However, we do not yet know whether this anti-infl ammatory effect occurs in humans. If so, it potentially opens the door for use of GLP-2 as therapy in conditions like infl ammatory bowel disease. Therefore we studied the immunomodulatory functions of GLP-2 in humans. Methods Dendritic cells (DC) enriched from human blood of healthy volunteers were cultured in-vitro for 24 h with GLP-2 at concentrations of 1 pM, 1 nM and 1 mM. The effect of GLP-2 on DC survival was determined using apoptosis experiments. Phenotype and functions of DC were then assessed by fl ow cytometry and mixed leucocyte reaction (MLR), respectively. Each experiment was performed independently at least 3 times and analysed for statistically signifi cant effects. Results Apoptosis experiments showed that GLP-2 at all concentrations did not have a toxic effect on DC; their survival after in-vitro culture with GLP-2 was similar to that in basal control culture (p=NS). GLP-2 conditioning changed the phenotype of DC with reduction in HLA-DR intensity (p=0.0243) and increase in CD14 expression (p=0.0237), compared with basal control culture. However, the down-regulation of HLA-DR intensity and up-regulation of CD14 expression did not correlate with an increase in the phagocytic capacity (p=NS). Other markers of immature DC, ILT3 and DC SIGN, were not affected. TLR2/4 expression was also not affected by the treatment (p=NS). Finally, MLR experiments showed that GLP-2 treatment on DC did not have an effect on their stimulation of T cell proliferation (p=NS).