Inter-hospital transports of critically ill children

Abstract

There is an increasing need for inter-hospital transports of seriously ill children as a result of centralization of intensive care for children to dedicated pediatric intensive care units (PICU). This thesis has explored the short-and long-term outcomes of critically ill children after transport by a specialized pediatric transport team to a PICU. In order to investigate the effects of high altitude on transported children, a novel method was introduced to improve the performance of a device for regional oxygen saturation (rSO2) monitoring working in a transport environment.

In Study I, outcomes of critically ill children acutely transported to PICU by a specialized pediatric transport team were retrospectively compared to a control group of children acutely admitted to the same PICU but through other routes. Transported children were younger, sicker and stayed longer in PICU and their use of PICU specific therapies was higher. The transport per se did not increase the risk of death irrespective of transport distance. Age and risk score related differences in short-and long-term survival among transported patients were evaluated in Study II. Median follow-up time for survivors was 4.4 years. Survival in neonatal patients was high after discharge from intensive care, and patients with a Predicted Death Rate (PDR) > 50% showed no mortality after the 30-day follow up. In contrast, there was clinically significant late mortality for the whole cohort, especially in those transported multiple times. In Study III we analyzed the impact of an arterial blood gas sample (i.e. the PaO2/FiO2 ratio) on the Pediatric Index of Mortality2 (PIM2) score and its derived probability of death (%). The PIM2 and probability of death only became more accurate if PaO2/FiO2 was available for the respiratory admission group, and when a rather high severity of illness was expected. The aim of Study IV was to investigate the applicability of near-infrared spectroscopy (NIRS) data during transport. The ability to distinguish between real and artefact-related events increased considerably by removal of zero values and “floor-effect values” and then using a filtering technique on the NIRS signal to reduce noise in the signal without loss of the original signal structure. In Study V, rSO2 with NIRS registration from cerebral (rSO2-C) and splanchnic (rSO2-A) areas during air ambulance transports of critically ill children was investigated in relation to the effect of altitude ≥ 5000 feet. Both rSO2-C and rSO2-A decreased significantly at altitude ≥ 5000 feet compared to baseline in a majority of patients. In most patients rSO2-A decreased more than rSO2-C ≥ 5000 feet as expressed by the rSO2-C/rSO2-A ratio, which was > 1 in 67% of patients at baseline and > 1 in 77% of patients at altitude ≥ 5000 feet.

In conclusion, this thesis has addressed various aspects of children in need of transport to pediatric intensive care. Acute pediatric inter-hospital transports can be performed without increasing the mortality risk regardless of transport distance if performed by a specialized team. There is a notable late mortality after the 30-day follow up for the transported group as a whole. An arterial blood gas sample for the PIM2 score is only needed when the patient has respiratory reason for admission and a rather high severity of illness is expected. Reliable NIRS data can be obtained during transport when cleared for artefacts and smoothed by a noise-reduction algorithm. Both rSO2-C and rSO2-A decreased as an effect of altitude ≥ 5000 feet, however rSO2-C was better preserved than rSO2-A.