Does washing blood for transfusion make a difference to preterm babies?

INTERVENTION: Group O Rhesus negative, washed leukodepleted red blood cells will be provided by Australian Red Cross Lifeblood. Washing of red blood cells will occur as per current established protocols. The washed leukodepleted red blood cells will be divided into four packs (quad‐packs) using closed techniques. Washed leukodepleted red blood cell quad‐packs will be supplied to the participating centres on a weekly basis. Washed, leukodepleted red cells split into quad packs using this process have a mean volume of 65 ± 6 mL/unit, a mean haematocrit of 54.8 ± 3.48. Each infant will receive their allocated red blood cell pack type for the first red blood cell transfusion and any subsequent red blood cell transfusions to first discharge home. The approach to transfusion will be standardised within each study centre, using a pre‐defined restrictive haemoglobin threshold used in the Transfusion of Prematures (TOP) study. The haemoglobin threshold will be derived from a capillary, arterial or venous blood haemoglobin concentration ([Hb]) and adjusted for whether or not the infant is receiving respiratory support (invasive ventilation via endotracheal tube (ETT), nasal continuous positive airway pressure (CPAP), nasal non‐invasive ventilation, nasal high flow, or supplemental oxygen) and by postnatal age. The protocol does not dictate sampling mode or how often [Hb] will be determined; this will be conducted according to local policies/clinician discretion. All [Hb] values derived from a complete blood examination (CBE) will be recorded to determine compliance with the transfusion threshold algorithm. Transfusion will be indicated whenever the [Hb] equals or falls below the threshold value and requires an active decision to transfuse to be made as soon as possible (at least CONDITION: Bronchopulmonary dysplasia;Intraventricular Haemorrhage;Retinopathy of Prematurity;Necrotizing Enterocolitis;Neonatal Mortality; ; Bronchopulmonary dysplasia ; Intraventricular Haemorrhage ; Retinopathy of Prematurity ; Necrotizing Enterocolitis ; Neonatal Mortality Eye ‐ Diseases / disorders of the eye Infection ‐ Other infectious diseases Inflammatory and Immune System ‐ Other inflammatory or immune system disorders Oral and Gastrointestinal ‐ Other diseases of the mouth, teeth, oesophagus, digestive system including liver and colon Reproductive Health and Childbirth ‐ Complications of newborn Respiratory ‐ Other respiratory disorders / diseases PRIMARY OUTCOME: A composite of mortality and/or major neonatal morbidities prior to first discharge home: ; 1. Death – defined as death from any cause.; 2. BPD – defined as a continued need for any form of respiratory support (supplemental oxygen and/or assisted ventilation) at 36 weeks post menstrual age (±2 days), or discharge home whichever occurs first. Based on a standardised diagnostic approach, a modified Walsh air reduction test, routinely performed in each of the participating centres. ; 3. ROP ‐ greater than grade 2 (uni‐ or bilateral) ; 4. NEC ‐ equal or greater then Bell’s stage =2 ; ; [Death or until primary hospital discharge] ; A Markov‐model will be built to estimate the long‐term costs (in Australian dollars) and health outcomes associated with RBC transfusion, using all data collected by ANZNN at 2‐3 years and then extrapolated over the subsequent 10 years to capture long term effects of common morbidities including asthma and neuro‐cognitive delay. Previous neonatal interventions have reported a sustained effect up to middle school age. 5% discounting per year will be applied to costs and outcomes, as per Australian government recommendations. The analysis will be performed using TreeAge Pro 2018 and R software. The model structure and analysis will follow best practice modelling guidelines from ISPOR. SECONDARY OUTCOME: all‐cause mortality after enrolment[time of death following enrolment] Bacterial, fungal or viral at greater than 48 hours of age (blood or cerebrospinal fluid culture or polymerase chain reaction positive and treatment with antibiotics with therapeutic intent) recorded from the subjects medical record[At death or at primary hospital discharge] Bronchopulmonary dysplasia – defined as a continued need for any form of respiratory support (supplemental oxygen and/or assisted ventilation) at 36 weeks post menstrual age (±2 days), or discharge home whichever occurs first. Based on a standardised diagnostic approach, a modified Walsh air reduction test, routinely performed in each of the participating centres. Results will be collected from the subjects medical record[At death or at primary hospital discharge] Home oxygen therapy recorded from the subjects medical record[At primary hospital discharge] Intraventricular haemorrhage greater than Grade 2 diagnosed by cranial ultrasound with results obtained from the subjects medical record.[At death or at primary hospital discharge] Length of hospital stay from birth to first discharge home recorded from subjects medical record[At primary discharge home] Maximal grade ROP to 3 months’ corrected age recorded from subjects medical record[At three months corrected age] Modelled cost‐effectiveness analysis ‐ Given the morbidity from ROP, NEC, BPD, IVH has a longer‐term impact on visual impairment, gastro‐intestinal dysfunction, and asthma; a modelled cost‐effectiveness analysis is necessary. Resource use and long‐term outcomes beyond the first hospital discharge will be obtained from a number of administrative datasets including the 2‐3‐year neurodevelopmental follow‐up of the ANZNN, National Perinatal Data Collection, MBS and PBS. Linkage will be through AIHW or Services Australia (formerly Department of Human Services). Data linkage of state‐based admission data for consenting individuals will enable the per participant calculation of health system resource use and costs up to 2 years after the primary hospital discharge. ; [At primary hospital discharge] INCLUSION CRITERIA: Infants born born less than 28 weeks’ gestation. Require one or more red blood cell transfusion ; [At three years of age] necrotising enterocolitis greater to or equal to Bells Stage 2 ‐ data obtained from subjects medical record[At death or at primary hospital discharge] Patent ductus arteriosus requiring treatment (pharmacological and/or surgical) recorded from the subjects medical record[At death or at primary hospital discharge] Porencephalic cysts, periventricular leukomalacia and any intracranial haemorrhage diagnosed by cranial ultrasound and record from the subjects medical record[At death or at primary hospital discharge] Postnatal steroids for broncho‐pulmonary dysplasia recorded from subjects medical record[At death or at primary hospital discharge] Retinal ablation or medical treatment (anti‐VEGF) to 3 months’ corrected age recorded from the subjects medical record[at 3 months corrected age] Retinopathy of Prematurity greater than grade 2 ‐ diagnosed by formal ophthalmology examination and result obtained from subjects medical record [At death of at to primary hospital discharge] Severity of BPD (Grade 1, 2 or 3) recorded from the subjects medical record[AT death or at primary hospital discharge] Spontaneous intestinal perforation (not necrotising enterocolitis associated) recorded from the subjects medical record[At death or primary hospital discharge] Total hours of high frequency oscillatory ventilation recorded from the subjects medical record [At death or at primary hospital discharge] Total hours of invasive ventilatory support (via endotracheal tube), nasal continuous positive pressure ventilation, nasal/non‐invasive ventilation, nasal high flow, and any respiratory support recorded from the subjects medical record.[At death or at primary hospital discharge] Total hours of parenteral nutrition recorded from the subjects medical record[At death or at primary hospital discharge] Weight, length and head circumference z‐score at first discharge home recorded from the subjects medical record[At primary hospital discharge] Within trial cost‐effectiveness ‐ The outcome for the analysis will be the additional cost per case of major neonatal morbidity/death avoided (i.e. primary end point) at hospital discharge. The total volume of major categories of resource use (e.g. red blood cell packs, days in each level of care and total length of hospitalisation) will be collected. Resource use will be valued using the most relevant unit pricing. For example, by multiplying the National Weighted Activity Unit (NWAU) for the infant’s diagnosis related group (DRG) relating to their initial hospitalisation, by the Net Efficient Price for the given reference year. Red blood cell packs from the Red Cross will be valued at market prices where available. Volumes of resource use and costs will be tabulated by allocated group. Cost data are likely to be right skewed meaning logarithm transformation of cost data will be considered for analysis. Mean costs with standard deviations and total costs for each group will be reported in Australian dollars for the most recent reference year. Linear mixed models will be used to compare resource use and costs between the two groups. ; An incremental cost effectiveness ratio (ICER) will be reported for the cost per neonatal morbidity/death avoided of the washed red blood cell group compared with the unwashed red blood cell group. Results will be plotted on a cost‐effectiveness plane. Bootstrapping will be used to estimate a distribution around costs and outcomes and to calculate confidence intervals around the ICER. A cost‐effectiveness acceptability curve will be plotted providing the probability of the intervention being cost‐effective given a decision maker’s willingness to pay for preterm infant morbidity/death avoided. One‐way sensitivity analyses will be conducted on key variables, for example, cost of washed red blood cells, gestational age, plurality, and study centre.