Position statement
Posted: Mar 4, 2022
Nicole Anderson, Michael Narvey, Canadian Paediatric Society, Fetus and Newborn Committee
Paediatr Child Health 2022 27(2):129 (Abstract)
“When will my baby come home?” is one of the most common questions asked by parents of preterm infants admitted to the neonatal intensive care unit (NICU). While the hospital course varies based on the gestational age at birth and the attainment of “physiological maturity”, the aim of this statement is to provide guidance for the safe discharge of infants born before 37 weeks. The discharge process should start at the time of admission to NICU, and with a plan for assessing physiological markers including thermoregulation, control of breathing, respiratory stability, and adequate weight gain as an indication of feeding skills. Importantly, the infant’s family unit is a crucial part of the care team and their involvement in the NICU will promote confidence, decrease anxiety, increase resilience, and help ensure a safe discharge environment.
Keywords: Apnea; Discharge; Late preterm; NICU; Psychosocial; Thermoregulation
This statement amalgamates and updates two previous Canadian Paediatric Society documents: ‘Going home: Facilitating discharge of the preterm infant’ and ‘Safe discharge of the late preterm infant’ [1][2].
A combined MEDLINE and Embase search was performed for studies from January 2014 up to May 2020, using identified keywords. Reference lists of publications were reviewed. The timeline was chosen to identify new literature published since previous CPS statements on this topic. A total of 178 references were retrieved, of which full text articles were reviewed. Levels of evidence and grades of recommendation were assigned in accordance with the Oxford Centre for Evidence-Based Medicine guideline [3] (Levels of evidence and grades of recommendation).
A preterm infant is defined as any infant born before the 37th week of gestation. The World Health Organization (WHO) defines an infant born between 33 to 37 weeks as moderate to late preterm [4]. Very preterm infants are those born between 28 to 32 6/7 weeks gestational age (GA). Extremely preterm infants are defined as any infant born less than 28 weeks GA. The preterm birth rate in Canada hovers around 8% of all live births, of which approximately 70% are born from 34 0/7 to 36 6/7 weeks, with the remaining 30% being born before 34 weeks GA [5][6].
The preterm infant’s journey may begin well before admission to the NICU, with a prolonged hospital admission for the mother, frequent perinatal ultrasounds, and close surveillance of the pregnancy. In some cases, the delivery of a preterm infant comes without warning. Regardless of scenario, admission to the NICU is a stressful and anxiety-provoking experience. Every infant admitted to the NICU experiences a unique hospital course that is often commensurate with their GA, with length of NICU admission being inversely correlated [7][8]. Complications such as sepsis, necrotizing enterocolitis, retinopathy of prematurity [9], or bronchopulmonary dysplasia (BPD) can individually or in combination contribute to length of stay. After recovering from such conditions, the safe discharge of a preterm infant is, at a minimum, dependent on the attainment of physiological maturity.
The weight and GA whereby an infant can maintain normothermia in an open cot is both infant-dependent and centre-specific. Though there is evidence supporting earlier weaning from the incubator with earlier discharge [10], one Cochrane review concluded that transfer to a cot at less than 1700 grams (versus greater than 1700 grams) did not lead to earlier discharge [11]. Studies on kangaroo care or skin-to skin-contact between caregiver and infant have been shown to promote normothermia at discharge, among other benefits [12][13].
In one observational study, all infants born ≤28 weeks GA had apnea of prematurity (AOP)[14]. The incidence of AOP followed an inverse relationship with GA, affecting 50% of those born at 30 weeks versus 7% born at 34 to 35 weeks. One systematic review in late preterm infants indicated an absolute risk of AOP of 0.87 % [15]. Caffeine (a methylxanthine) is one of the most commonly prescribed medications in the NICU and is the mainstay of pharmacological treatment for AOP [16]. Although there is variation as to when caffeine is discontinued, most centres will opt to do so between 32 and 37 weeks postmenstrual age (PMA). This timing is due to the fact that AOP is less common beyond 36 weeks GA [17][18]. The mean half-life of caffeine is approximately 100 hours but varies with GA at birth and chronological age [19]-[21]. Thus, a minimum 5- to 7-day period of monitoring in the NICU after caffeine discontinuation is suggested. A longer period of observation may be necessary for infants born at less than 29 weeks GA [17]. There is no current literature to support shorter durations of monitoring after discontinuation of caffeine. For infants with AOP who have not been treated with caffeine, an apneic-free period is more difficult to discern. The risk for recurrence of apneic events after a period of 3 versus 7 days from the last event shows minimal risk reduction. Therefore, practitioners may reasonably choose a minimum 3-day observation period and consider a longer period for infants with a more severe or prolonged course of AOP [22][23].
While it is preferable to discontinue caffeine in hospital, continuing caffeine at home may be considered as part of a shared decision between the family and health care team. Importantly, AOP is not an identified risk factor for sudden infant death syndrome (SIDS), and there is no evidence suggesting clear benefit of home monitoring in the context of AOP or SIDS prevention [24][25]. SIDS prevention strategies, including safe sleep, should be reviewed (and modelled) during the NICU stay [26][27]. A premature infant born at 32 weeks GA in the NICU should be placed supine during sleep to promote tolerance of this position before discharge [28].
Respiratory stability is achieved when an infant no longer requires ventilatory support and (preferably) is not receiving supplemental oxygenation. Preterm infants often require ventilatory support, with appropriate de-escalation during their NICU stay. A retrospective study of 224 infants <27 weeks GA found a correlation between earlier attempts at extubation and shorter lengths of stay [29]. Clinicians should also consider that reintubation (within 14 days of first extubation) may be associated with adverse outcomes [30].
BPD is clinically diagnosed when supplemental oxygen or mechanical ventilation is provided beyond 36 weeks GA, although this definition is constantly evolving [31][32]. Oxygen saturation targets in the NICU remain inconclusive, but most authors suggest a target of 90% to 95% [32]-[34]. Diuretics, steroids (inhaled or systemic), and home oxygen may be part of the discharge plan for infants with BPD [35]. For appropriate infants and families, home oxygen is preferable to a prolonged NICU admission. Arranging for home oxygen support may help to increase growth, decrease mean pulmonary artery pressure, and (when used at night) improve sleep duration [32].
BPD-associated pulmonary hypertension (PH) is an important diagnosis because it carries significant mortality [36]. The prevalence of PH is proportional to BPD severity, with up to 39% of infants with this condition being classified as having severe BPD [37]. Screening for PH in high-risk infants, along with evaluation of comorbidities including intermittent hypoxemia, aspiration, gastroesophageal reflux, structural airway abnormalities, pulmonary artery or vein stenosis, and left ventricular dysfunction, should be pursued in the NICU as well as followed up post-discharge [38].
To minimize further risk in infants with compromised pulmonary reserve, health care practitioners should promote a smoke-free home environment and infection prevention strategies, such as handwashing. Moreover, adherence to the routine immunization schedule, along with respiratory syncytial virus (RSV) prophylaxis when criteria are met, is essential [39].
The establishment of oral feeding is a significant milestone during an NICU admission. Establishing breastfeeding should be encouraged when that is the mother’s goal. In preterm infants, barriers to oral feeding may include an immature suck and swallow mechanism, invasive ventilatory support, cardiorespiratory instability, and ‘nil per os’ (NPO) status, all of which can make feeding skills difficult to attain. Organized non-nutritive sucking has been observed in infants around 28 to 29 weeks PMA, with subsequent attainment of exclusive breastfeeding at 32 to 38 weeks [40][41]. One recent study showed the median PMA for establishment of full oral feeds was 37.1 weeks and 34.7 weeks for extremely preterm and very preterm infants, respectively [42].
There has been a recent shift to infant-driven and standardized feeding algorithms for preterm babies [43][44]. Introduction of the ‘safe individualized nipple-feeding competence’ (or SINC) approach to feeding extremely preterm infants on non-invasive respiratory support has allowed earlier development of feeding skills than the traditional approach of waiting until respiratory support has been discontinued [45]. Importantly, however, SINC implementation requires a standardized and educated team. For infants born <32 weeks, SINC has been shown to achieve earlier attainment of first breast or nipple feeds [45]. One recent randomized controlled trial (RCT) comparing SINC with “regular care” showed an effect on weight gain but not on transition to full oral feeds or length of hospitalization [45]. Hence, there is a need for evidence-based tools that assess preterm infant readiness to initiate oral feeds [46].
When caring for preterm babies, practitioners should provide the tools necessary for mothers to provide as much human milk as possible. Initiation relies on lactogenesis, a transitional process that is altered by a premature birth [47]. Also, breast milk supply can be negatively affected by concerns about adequacy of amount, difficulties with expression, and physical separation from the infant [48]. Promotion of breastfeeding should begin with maternal obstetrical care and continue from point of admission throughout the NICU stay. Studies have shown a significant association between initiation of early pumping and first oral feeding at breast and breast milk feeding at discharge [49]-[51]. Skin-to-skin care has been shown to improve breastfeeding efficacy and duration post-discharge [52]. Galactogogues for breast milk production, such as metoclopramide and domperidone, increase serum prolactin by inhibition of dopamine [48]. Evidence in this area is limited by heterogeneity in trial design and outcome measures. However, domperidone appears to be more efficacious in increasing breast milk production [47][53].
Mothers who plan to breastfeed their infant post-discharge can begin to introduce solid foods at 4 to 6 months to meet iron requirements, while continuing to breastfeed. There should be appropriate supports in place (e.g., referral to a lactation consultant) to support breastfeeding past discharge [54]. Factors associated with shorter duration of breastfeeding post-NICU include lower maternal socioeconomic status (SES), lower maternal education, and maternal smoking. Mothers with these risk factors require additional support during the discharge planning process [52].
Reflux is reported in approximately one-in-ten preterm infants [55]. Routine pharmacological intervention treatment of reflux is not recommended [56]. However, as per the guidelines developed by the European Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) with the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN), when a pharmacological is needed, proton pump inhibitors (PPIs) are the treatment of first line [57]. Given the likelihood of prolonged use in the outpatient setting and the side effect profile, practitioners should review the ongoing need for a PPI at the time of discharge [56][58].
Education on how to administer medications and supplements, including iron, vitamin D and, if appropriate, calcium and/or phosphate, should be provided to caregivers. Guidance on monitoring and length of treatment should be shared with the family’s health care provider (HCP). Iron supplementation should continue for 6 to 12 months chronological age (depending on birth weight) for infants who are breastfed exclusively to 6 months of age [59]. For infants requiring additives to human milk or formula feeding, caregivers should be taught and given opportunities to prepare feeds before NICU discharge [60]. Use of fortified human milk is usually limited to hospital settings, such that alternatives needed to meet the infant’s nutritional requirements must be anticipated and planned before discharge. Inaccurate mixing of feeds is a common mistake and can lead to electrolyte disturbance and poor growth. After discharge, consultation with an outpatient dietitian and the infant’s primary HCP should guide the duration of treatment with milk additives.
There is no standard minimum weight at which an infant can be discharged, but timing is guided by weight criteria for infant car seats. Families who anticipate going home before their infant weighs 2.2 kg (5 lbs) must make sure they have a car seat with a ~1.8 kg (4 lb) minimum weight.
As an infant nears the physiological maturity necessary for safe discharge, caregivers and the infant’s environment must be adequately prepared for life at home. While there is an abundance of literature on the negative psychosocial effects of having a preterm infant in the NICU, it is essential to consider each family’s sources of support, including cultural networks and care practices [58][61]-[68].
Family readiness should be assessed throughout the NICU admission. Family-centred or integrated care frameworks (or an equivalent centre-specific initiative) provide a basis for NICU assessment and support. Families should be both involved with and informed of their infant’s care, and collaborative care helps increase resilience and readiness for homecoming and decreases risk for hospital readmission [69]-[74]. Preparation also improves neonatal outcomes [75][76]. Data from locations with an early discharge program (often complemented by outreach and home support) have clearly demonstrated that collaboration between caregivers and medical teams means safer transitions home [77][78].
Typical newborn care, as outlined in ‘Facilitating discharge from hospital of the healthy term infant’ [79], is important for context. The following steps are always part of discharge education:
Late preterm infants may appear to be mature but still lack key elements of physiological maturity that facilitate safe discharge home. Before discharge, late preterm newborns must demonstrate respiratory and temperature stability, adequate feeding skills, and euglycemia. Deficits in these areas can interact to put infants at risk [80]. For example, a late preterm infant may become hypothermic and feed poorly, which can lead to lethargy that prompts an unnecessary work-up for sepsis and hospitalization.
Late preterm infants are at risk for problems post-discharge and require timely follow-up by community care providers. Those with stable weight and bilirubin levels may be discharged in the first week post-birth, provided there is a follow-up appointment within 72 h of discharge. Bilirubin levels in this population peak later, remain elevated for longer, and achieve higher peak values compared with term infants [81][82]. A clinician should arrange for follow-up within 24 to 48 h from discharge with appropriate public health services.
The psychosocial effects of having a preterm baby often linger, with many families requiring ongoing support at home [83]-[85]. In some birthing centres, home-based support programs follow preterm infants post-discharge. This population is frequently re-hospitalized during infancy, requiring ongoing health care resources. Home-based discharge teams can help mitigate such needs [72][86]-[90]. The positive outcomes of these programs include improved breastfeeding, parent-infant interactions, and neurodevelopmental outcomes [72]. When an infant is being discharged into a remote community, liaising directly with the child’s primary care provider in-community is recommended. There may be a role for supportive web-based or telemedicine programs [91][92].
The following checklist is based on recommendations reached by consensus and drawn from several evidence-based resources, with a general evidence level of 2 [3]. For a printable version of this checklist, click here.
Within the first 72 h of admission and before delivery, if possible:
Before discharge:
At discharge:
This position statement was reviewed by the Community Paediatrics and Nutrition and Gastroenterology Committees of the Canadian Paediatric Society. It has also been reviewed by the College of Family Physicians of Canada.
CANADIAN PAEDIATRIC SOCIETY FETUS AND NEWBORN COMMITTEE (2020-2021)
Members: Gabriel Altit MD, Nicole Anderson MD (Resident Member), Heidi Budden MD (Board Representative), Leonora Hendson MD (past member), Souvik Mitra MD, Michael R. Narvey MD (Chair), Eugene Ng MD, Nicole Radziminski MD, Vibhuti Shah MD (past member)
Liaisons: Radha Chari MD (The Society of Obstetricians and Gynaecologists of Canada), James Cummings MD (Committee on Fetus and Newborn, American Academy of Pediatrics), William Ehman MD (College of Family Physicians of Canada), Danica Hamilton RN (Canadian Association of Neonatal Nurses), Chloë Joynt MD (CPS Neonatal-Perinatal Medicine Section Executive), Chantal Nelson PhD (Public Health Agency of Canada)
Principal authors: Nicole Anderson MD, Michael Narvey MD
Disclaimer: The recommendations in this position statement do not indicate an exclusive course of treatment or procedure to be followed. Variations, taking into account individual circumstances, may be appropriate. Internet addresses are current at time of publication.
Last updated: Feb 8, 2024