The impact of transfers from neonatal intensive care to paediatric intensive care

Infants receiving care from neonatal intensive care unit (NICU) can develop chronic problems with resource implications [1]. They may be transferred to a paediatric intensive care unit (PICU) for ongoing care never having been home. There is concern that such infants may take up a large amount of PICU resource, but there is no evidence base to confirm or otherwise such a suggestion. As a consequence, we assessed the impact of such transfers in terms of PICU length and cost of stay. We also tested the hypothesis that infants born at less than 28 weeks of gestational age, are most likely to develop chronic problems and would, therefore, have particularly long stays.

We studied, over a 10-year period (2009–2019), all infants admitted to two regional PICUs from their local or other NICUs without ever being discharged home. Infants were not included if they were transferred to PICU because of a lack of NICU cot capacity.

One ten-bedded PICU and the other 16 bedded PICU receive referrals from South London and the wider region, both have an average number of 600 admissions per year [2]. The national average unit cost per bed day in PICU is £3,748 and in NICU is £1,218 [3]. In 2019–2020, there were 620 admissions to one PICU and 4,504 bed days with a total income of £7,661,854.

Data were collected from the PICAnet database and the Electronic Patient Record (EPR) system, paper notes were requested for those admitted prior to the introduction of EPR (2012).

Differences between those infants born extremely premature and the rest of the NICU transfers were assessed for statistical significance using the Mann Whitney U test. Statistical analysis was performed using SPSS software version 25.

During the 10-year period, there were 89 transfers from NICUs to the two PICUs and approximately 12,000 admissions overall to the two PICUs. Twenty-two infants were excluded from the analysis as five had missing data and the other 17 were transferred because of lack of NICU cot capacity or required neurosurgery and were transferred to PICU post-operatively for ongoing care.

The 67 infants included in the analysis had a median gestational age at birth of 34.7 (IQR 27.1–38.8, range 23.4–42.1) weeks and birth weight of 2.21 (IQR 0.88–3.27, range 0.53–4.80) kg. Their corrected gestational age on admission was 43.6 (IQR 39.1–52.3, range 33.4–70.6) weeks. Forty infants were transferred from the local NICUs and the rest from other NICUs. Forty of the infants were born prematurely. Nineteen infants were born at less than 28 weeks of gestation and transferred to PICU at a corrected gestational age of 47.3 (IQR 43.9–58.0, range 33.8–70.6) weeks.

Forty-nine infants were receiving respiratory support on admission: 27 were ventilated, 16 were receiving non-invasive ventilation and six were receiving nasal cannula oxygen. Nine infants had a tracheostomy prior to admission. Fifty-three infants (79%) survived to discharge.

The median length of stay was 12 (IQR 4–41) days, range 1–470 days. The median cost of PICU stay was £23,800 (IQR £8,800–83,600, range £1,205–1,034,000) per baby and the total number of bed days was 3,099 days. The total cost of care for infants transferred from NICUs was £6,457,955. Infants born at less than 28 weeks of gestation had a greater median length of stay (32, IQR 10–93, days) than the more mature born infants (median 7.5 (IQR 4–26 days)) (p=0.003). The median cost of care of the very prematurely born infants was £68,200 (IQR £18,000–204,600) per baby.

We have demonstrated that infants transferred from NICUs for ongoing care are a small fraction of the PICU population, but some, particularly those born before 28 weeks of gestation, had a prolonged stay. We did not include the impact of the 17 infants transferred to PICU because of lack of neonatal cot capacity. The cost of a bed day in PICU is three times that of a NICU cot day (the national average unit cost per bed day in PICU is £3,748 and the national average unit cost per bed day in NICU is £1,218 [3]). Such a transfer then is an expensive option and our data argues for increased neonatal cot capacity. A strength of this study is that it is the first study to quantify the impact of transfers from NICUs to PICUs. Furthermore, we collected data from two PICUs. A limitation is that this a retrospective study, but we analysed prospectively, electronically collected data for the majority of the infants.

In conclusion, some infants transferred from NICUs had prolonged stays which need to be considered when determining bed capacity.