The ductus arteriosus (DA) is a large channel found normally in all mammalian foetuses, connecting the main pulmonary trunk with the left-sided descending aorta. During fetal life, it diverts most of the combined ventricular output away from the lungs toward the placenta. In full-term newborns, the DA is functionally open in all newborns and its constriction and closure, within 24 to 48 hours after delivery, are part of the normal process of postnatal adaptation. In extremely preterm infants, ie, those born before 29 weeks of gestation, the DA frequently fails to close and this results in a condition called patent ductus arteriosus (PDA). In these patients, PDA has been associated with increased mortality and morbidity in the neonatal period, and neonatal morbidities may in turn be associated with later deficits in cognitive functioning. Inhibition of prostaglandin synthesis, through inhibition of the cyclo-oxygenase (COX) enzymes, results in ductal constriction. However, treatments with COX inhibitors, as ibuprofen or indomethacin, aiming at closing the PDA have been associated with numerous adverse effects and failed to demonstrate significant clinical benefits. Prophylactic use of indomethacin or ibuprofen has been shown to reduce subsequent development of PDA but not to reduce mortality or morbidity. Further, investigations of early cardiac ultrasound-targeted treatment of a large PDA using indomethacin and ibuprofen do not show efficacy on primary outcomes of neonatal death or morbidity. Although this evidence base shows that prophylactic treatment by indomethacin or ibuprofen cannot be recommended, evidence from observational studies suggests that early (during the first 3 days of life) conservative treatment should not be recommended either. Two observational studies indicated that ignoring a PDA during the first days of life is associated with increase in mortality or morbidity. Our interpretation of these results is that screening echocardiography of large ductus arteriosus and early targeted treatment using echocardiography reduce pulmonary haemorrhage, but that the use of drugs with severe adverse effects may cancel expected beneficial effects on the occurrence of death or morbidity.
The use of a medication with an effect on PDA and that has fewer adverse effects is therefore desirable. Early treatment of PDA with paracetamol (acetaminophen) has been proposed as an alternative to COX inhibitors. According to a recent meta-analysis, use of paracetamol has a comparable effectiveness to close a PDA, and fewer harmful effects. However, only few extremely preterm infants were included in this meta-analysis. Therefore, efficacy and safety of paracetamol for PDA treatment in this population required further studies.
The ongoing pan-European TREOCAPA phase III study (NCT04459117) is a multicentre, double-blind, randomised, placebo-controlled superiority trial that assesses prophylactic use of paracetamol to improve survival without severe neonatal morbidity until discharge from hospital in infants of 23-28 weeks of gestational age.12 The trial has been powered to identify an absolute difference of 10% in the primary outcome, survival without severe neonatal morbidity (defined as any of: bronchopulmonary dysplasia grade 3 according to National Institute of Health criteria, necrotising enterocolitis stage 2 or 3 using Bell's criteria, intraventricular haemorrhage grade III or IV using the classification of Papile et al, or any evidence of cystic periventricular leukomalacia), corresponding to an increase from 50% to 60% in the rate between groups at hospital discharge. It aims to recruit 398 infants born at 27-28 weeks of gestation, and 396 infants born at 23-26 weeks of gestation. The first patients were recruited in October 2020 and the recruitment period is anticipated to last until April 2024 at the latest.
The TREOCAPA phase III trial is ambitious and innovative in its scope, including more than 40 hospitals in 16 countries. However, long-term follow-up was not included in the TREOCAPA protocol. The follow-up of infants enrolled in clinical trials enables to measure the long-term effectiveness and safety of interventions performed in the neonatal period. In particular for exposition to paracetamol in the neonatal period, data from randomized studies are very limited, while long-term adverse effects of paracetamol, including notably behavioral problems, have been raised. There is hence a need to understand longer term outcomes for the infants enrolled in this study, as these longer-term endpoints correspond to important patient-valued outcomes and are critical to determining the comparative effectiveness of interventions. The working hypothesis, if the intervention has a positive impact on short-term endpoints, is that longer term outcomes will be better in the intervention group, reflecting reduced morbidity at discharge home from the neonatal hospitalisation.
As long-term follow-up was not planned by the TREOCAPA phase III protocol, TREOCAPA-LT study will use an existing European research infrastructure, the RECAP Preterm platform to follow-up patients enrolled in the TREOCAPA cohort at 2 years of corrected age. This platform federates 23 European longitudinal observational cohorts of children born very preterm. The platform includes expertise, tools and infrastructure to follow patients enrolled in trials. By integrating trial follow-up in this platform, it is also possible to compare patients enrolled in the TREOCAPA trial with those from population-based samples, allowing assessment of transportability of trial results.
Follow-up at 2 years of age is recommended as the focus for the first phase of long-term outcome monitoring and has become the de facto age which is used in many published and ongoing clinical trials. Furthermore, many hospital services provide follow-up for their patients until at least 2 years of age, and research contact with the families at this point provides an opportunity to maintain contact and demonstrate to patients the ongoing importance of their involvement in the study.
The TREOCAPA-LT primary hypothesis is that there will be improved cognitive outcome at 2 years of corrected age in children born at less than 29 weeks of gestational age who were treated with paracetamol during the first 5 days of life in the TREOCAPA trial.
Secondary hypotheses are that, among children treated with paracetamol/acetaminophen during the first 5 days of life:
1. There will be reduced moderate or severe neurodevelopmental impairment, defined as the presence of at least one of the following: PARCA-R non-verbal cognitive score less than two standard deviations below the mean (score \<70) moderate-severe motor impairment, unilateral or bilateral deafness or blindness.
2. There will be reduced need for secondary hospitalisations following the initial neonatal hospitalisation.
3. There will be reduced long term health care utilisation costs associated with extremely preterm birth.
Secondary hypotheses also relate to the use of the RECAP Preterm platform of observational VPT cohorts in this study. These are that:
1. As a trial population, the children born VPT included in the TREOCAPA-LT study will differ from those included in population-based observational cohorts in their perinatal characteristics and possibly in their health and developmental outcomes.
2. That these differences in population characteristics can affect the transportability of the trial results to other populations of VPT infants.
