Evidence reviews for respiratory support: Specialist neonatal respiratory care for babies born preterm

Excerpt

The type of care a preterm baby receives within the first few hours of life can have a significant impact on their long-term outcome. The phrase “golden hour,” first used in trauma patients, has been adopted to refer to neonatal care at this crucial time.

Early delivery room respiratory support in preterm infants has been extensively investigated and may make a significant contribution to reducing the risk of long-term lung damage, other morbidities and even death. One of the difficult choices in the current era is to determine whether or not to intubate a preterm baby in order to give surfactant very soon after birth. Many babies can be supported by non-invasive methods of delivering oxygen, such as continuous positive airways pressure (CPAP), which avoid intubation. There is evidence that surfactant can be administered to these non-intubated babies using less invasive administration techniques that may reduce the risk of morbidity associated with intubation. At present it is not clear which is the best strategy.

This review aims to explore which delivery room respiratory support techniques are likely to give optimal disease-free survival in preterm infants. We have compared early invasive intubation and surfactant administration techniques, with less invasive surfactant administration techniques and non-invasive respiratory support techniques.

Respiratory distress syndrome (RDS) in preterm babies is caused by a deficiency of lung surfactant. The risk of RDS increases with decreasing gestational age, and is almost inevitable in babies born at less than 28 weeks gestation. Without surfactant the lungs become stiff and the alveoli collapse at end-expiration, and untreated RDS is a major cause of morbidity and mortality in preterm infants.

Surfactant is a naturally produced surface-active lipoprotein complex mixed with proteins, which reduces the surface tension at the alveolar liquid surface. Surfactant allows alveoli to stay open in expiration and substantially reduces the work of breathing. It also reduces shearing forces on immature alveolar membrane, preventing membrane rupture and protein leak into the alveolar space with resulting lung damage. RDS in preterm babies can be prevented by administration of exogenous animal derived surfactant therapy, and this substantially reduces mortality and respiratory morbidity for this population, including improved survival without bronchopulmonary dysplasia (BPD) at 28 days.

However, the optimal dose (including the use of single or multiple administration) and mode of administration of surfactant remains controversial and may make a significant contribution to the chances of long term lung damage, other morbidities or death. The various techniques of administration can be grouped into three categories:

1. conventional endotracheal intubation (where the baby is intubated, surfactant is administered and the baby then continues on mechanical ventilation)

2. endotracheal intubation and surfactant administration followed by immediate extubation (also called Intubate, Surfactant, Rapid Extubation, and known as InSuRE or ISX)

3. surfactant administration without endotracheal intubation via a thin endotracheal catheter during spontaneous breathing or with continuous positive airways pressure (CPAP) (known as Minimally Invasive Surfactant Therapy, MIST or Less Invasive Surfactant Administration, LISA, or Avoidance of Mechanical Ventilation, AMV).

This review aims to explore which administration technique and dosing regimen is likely to give optimal outcomes in preterm infants.

Low flow oxygen is frequently used in neonatal units, as an integral part of respiratory support in preterm babies. The goal of oxygen therapy is to achieve adequate delivery of oxygen to the tissues without causing oxygen toxicity.

In addition to delivering oxygen via a ventilator or CPAP circuit, there are several different methods of low-flow oxygen administration: head box oxygen, incubator, facemask, nasal prongs, nasal cannula and nasopharyngeal catheter. Oxygen can be delivered humidified or non-humidified, and the method of titration can be automated or manual. It is important to know, the efficacy, potential risks, and the impact on lung function of the different methods when used in preterm babies.

The aim of this review is to determine the optimal type of oxygen delivered, method of administration and method of titration in preterm babies requiring respiratory support.

The lungs of preterm babies are structurally immature, deficient of surfactant and not supported by a rigid chest wall. They are therefore highly susceptible to injury from the different types of respiratory support available for use in this population.

Whereas pressure support ventilation (PSV) and continuous positive airway pressure (CPAP) have been used in neonatology for many years, newer modes of ventilation such as volume targeted ventilation (VTV) and, more recently, heated humidified high-flow nasal cannula (HHHFNC) have become popular. This review will look at the evidence available to assess the effectiveness of the different types of assisted ventilation techniques in preterm babies.

Inhaled nitric oxide is a potent, selective pulmonary vasodilator. It has a clearly defined role in the management of pulmonary hypertension in term babies, where it has been shown to improve oxygenation and reduce the need for extracorporeal membrane oxygenation.

However, its role in preterm babies requiring invasive respiratory support is less well defined. Recently, a number of studies have been published looking at the effect of inhaled nitric oxide on the incidence of bronchopulmonary dysplasia (BPD) in preterm babies and its potential role in hypoxic respiratory failure. This review aims to determine the effectiveness of inhaled nitric oxide in preterm babies requiring invasive ventilation, both in hypoxic respiratory failure and in the prevention of bronchopulmonary dysplasia

In the NMA for the outcome of mortality prior to discharge and BPD at 36 weeks PMA there was no evidence to suggest a difference between between CPAP, NIPPV, BiPAP/SiPAP, or Hi Flow. Similarly, pairwise analyses did not identify any meaningful differences between non-invasive ventilation techniques.

The committee acknowledged two existing non-UK economic evaluations comparing CPAP with NIPPV and Hi Flow, respectively. However, these analyses did not include all non-invasive ventilation techniques of interest.

Given the lack of differences in the clinical effectiveness between non-invasive ventilation techniques and the lack of existing economic evidence the committee considered it important to compare the costs of the techniques to aid considerations of cost effectiveness. Generally the NHS Reference Costs (DHSC, 2018) is the recommended source of unit cost data that should be used to aid considerations of cost effectiveness. However, the committee explained that the neonatal activity payments are based on the level of activity (that is, intensive care, high dependency and special care) rather than procedures. As a result, costings of non-invasive ventilation techniques were undertaken to aid considerations of cost effectiveness and included equipment acquisition costs, maintenanece costs and consumable costs.

The results of conventional pairwise meta-analyses of direct evidence alone do not help to fully inform which invasive and non-invasive ventilation technique is most effective in preterm babies requiring respiratory support.

Each pairwise comparison does not fully inform the choice between the different treatments and having a series of discrete pairwise comparisons can be incoherent and difficult to interpret.

In addition, direct comparisons of treatments of clinical interest are not fully available, for all comparisons.

To overcome these issues, a Bayesian network meta-analysis (NMA) was performed. Advantages of performing this type of analysis are as follows.

1. It allows the synthesis of evidence on multiple treatments compared directly and indirectly without breaking randomisation. If treatment A has never been compared to treatment B in a head to head trial, but these two interventions have been compared to a common comparator, then an indirect treatment comparison can be derived using the relative effects of the two treatments versus the common comparator. Indirect estimates can be calculated whenever there is a path linking two treatments through a set of common comparators. All the randomised evidence is considered simultaneously within the same model.

2. For every intervention in a connected network, a relative effect estimate (with its 95% credible intervals, CrIs) between any two interventions can be estimated. These estimates provide a useful clinical summary of the results and facilitate the formation of recommendations based on all relevant evidence, whilst appropriately accounting for uncertainty. Ranks of interventions may also be calculated.

3. Estimates from the NMA can be used to directly parameterise treatment effectiveness in cost-effectiveness modelling of multiple treatments.

Conventional fixed effect meta-analysis assumes that the relative effect of one treatment compared to another is the same across an entire set of trials. In a random effects model, it is assumed that the relative effects are different in each trial but that they are from a single common distribution and that this distribution is common across all sets of trials.

NMA requires an additional assumption over conventional meta-analysis. The additional assumption is that intervention A has the same effect on people in trials of intervention A compared to intervention B as it does for people in trials of intervention A versus intervention C and so on. Thus, in an NMA, the assumption is that intervention A has the same effect across trials of A versus B, A versus C and so on.

The terms indirect treatment comparisons, mixed treatment comparisons and NMA are used interchangeably. We use the term NMA as the network consists of both indirect treatment comparisons (some trials have a common comparator and some do not) and mixed treatment comparisons (with at least one closed loop, combination of direct and indirect evidence).

The purpose of this analysis was to assess the consistency assumption in the network meta-analysis (NMA) model used to estimate the comparative effectiveness of non-invasive ventilation techniques in specialist neonatal respiratory care (SNRC) for babies born preterm. The outcomes included in this analysis were 1) mortality prior to discharge and 2) bronchopulmonary dysplasia (BPD) at 36 weeks postmenstrual age (PMA).

The purpose of this analysis was to assess the consistency assumption in the network meta-analysis (NMA) model used to estimate the comparative effectiveness of specialist neonatal respiratory care (SNRC) interventions for babies born preterm. The outcomes included in this analysis were 1) mortality prior to discharge and 2) Bronchopulmonary Dysplasia (BPD) at 36 weeks PMA.