How infants die in neonatal intensive care units – a European perspective

How decisions and processes should take place in neonatal end of life situations is a matter of constant concern since the emergence of modern neonatal intensive care in many high-income countries. Every time, a redirection of care is being weighted in an infant’s best interest, a considerable amount of stress and responsibility is imposed on parents and on care givers teams and this happens nearly daily in many NICUs. However, environments that differ in terms of available resources, of culture and religions, of laws and public policies, of personal opinions and emotions result in differing decisions, argumentations and outcomes in medically otherwise comparable situations. This leads to the perception that a single baby in a determined critical condition may be offered radically different orientations of care, depending on where and eventually by whom he is cared for. This may reflect the inevitable and necessary counterpart to a perfect but inhuman standardization of care. But comparing opinions, facts and argumentations over the world in the most common circumstances encountered in neonatal intensive care may help find common grounds and define the better, or at the least harmful ways to answer ethical dilemmas. Most of the information available on the decision-making processes in neonatal critical situations is on the Western European and the North American experience. In this issue of Acta Paediatrica, C. Fajardo et al. (1) report the first description of how babies die in a selected sample of 8 NICUs in Hispanic Latin America. The study carries the limitations which are usually associated with retrospective observations of patients’ charts, especially on subjective issues such as the motives and the degree of parental involvement in decision-making. Also, the study is restricted to the patients who died, and how far its results are representative of the wider scene may be questioned, especially for extremely low gestational age infants who seemingly are underrepresented in the sample. However, the authors report a clear-cut difference with what is predominantly known in other countries. In 42% of the cases, the patients died while full care was still provided; less than half of these cases had do not resuscitate orders. Moreover, in no patient care had been withdrawn, at any time, in any participating centre: those who died in the absence of mechanical ventilation had never been ventilated. Withholding of intensive care was applied exclusively in the delivery room, to 17% of the patients. There are apparently contradictory findings in the study, possibly related to methodological issues, such as the 21% rate of decisions justified by future quality of life concerns, which contrasts with the absence of cases of intensive care withdrawal. Also, areas for possible progress have been identified, in the participation of parents in decision-making (15%) and in low rates (14%) of use of sedatives and narcotics. The findings in this study re-open the question of whether withholding life support in at-risk newborn infants is different from withdrawing it, or more precisely, whether withholding resuscitation or acute life support at birth is different from withdrawing prolonged life support care later, in the NICU. The figures reported by Fajardo et al. (1) are different from what is generally assumed to occur internationally and in Europe. A recent cross-cultural comparison based on a similar method of chart review and using standardized definitions suggests that between 69% and 93% of all infant deaths in NICUs occurs after intensive support had been Invited Commentary for Fajardo et al, End of life, death and dying in neonatal intensive care units in Latin America, pages 609–613. Acta Paediatrica ISSN 0803–5253