Lung recruitment manoeuvres for reducing mortality and respiratory morbidity in mechanically ventilated neonates.

Background Preterm infants and neonates with respiratory conditions commonly require intubation and conventional mechanical ventilation (CMV) to maintain airway patency and support their respiration. Whilst this therapy is often lifesaving, it simultaneously carries the risk of lung injury. The use of lung recruitment manoeuvres (LRMs) has been found to reduce the incidence of lung injury, and improve oxygenation and lung compliance in ventilated adults. However, evidence pertaining to their use in neonates is limited, and there is no consensus of opinion as to whether LRMs are appropriate or effective in this population. Objectives To determine the effects of LRMs on mortality and respiratory outcomes in mechanically ventilated neonates, when compared to no recruitment (routine care). SEARCH METHODS: We used the standard search strategy of Cochrane Neonatal to search the Cochrane Central Register of Controlled Trials (CENTRAL 2020, Issue 4) in the Cochrane Library, MEDLINE via Ovid (1946 to 13 April 2020), and CINAHL via EBSCOhost (1989 to 13 April 2020). We also handsearched the reference lists of retrieved studies to source additional articles. SELECTION CRITERIA: We included randomised controlled trials (RCTs), quasi-RCTs and randomised cross-over studies that compared the effect of LRMs to no recruitment (routine care) in mechanically ventilated neonates. Data collection and analysis Two review authors independently assessed trial eligibility, extracted data and evaluated risk of bias in the included studies. When studies were sufficiently similar, we performed a meta-analysis using mean difference (MD) for continuous data and risk ratio (RR) for dichotomous data, with their respective 95% confidence intervals (CIs). We used the GRADE approach to assess the certainty of the evidence for key (clinically important) outcomes. Main results We included four studies (152 participants in total) in this review. Three of these studies, enrolling 56 participants, contributed data to our prespecified outcomes. Two studies enrolling 44 participants on CMV for respiratory distress syndrome compared a stepwise LRM with positive end-expiratory pressure (PEEP) to routine care. Meta-analysis demonstrated no evidence of a difference between the LRM and routine care on mortality by hospital discharge (RR 1.00, 95% CI 0.17 to 5.77; low-certainty evidence), incidence of bronchopulmonary dysplasia (RR 0.25, 95% CI 0.03 to 2.07; low-certainty evidence), duration of supplemental oxygen (MD -7.52 days, 95% CI -20.83 to 5.78; very low-certainty evidence), and duration of ventilatory support (MD -3.59 days, 95% CI -12.97 to 5.79; very low-certainty evidence). The certainty of the evidence for these outcomes was downgraded due to risk of bias, imprecision, and inconsistency. Whilst these studies contributed data to four of our primary outcomes, we were unable to identify any studies that reported our other primary outcomes: duration of continuous positive airway pressure therapy, duration of neonatal intensive care unit stay, and duration of hospital stay. The third study that contributed data to the review enrolled 12 participants on CMV for respiratory and non-respiratory causes, and compared two different LRMs applied after endotracheal tube suctioning to routine care. It was determined that both LRMs may slightly improve end-expiratory lung volume at 120 minutes' post-suctioning, when compared to routine care (incremental PEEP LRM versus routine care: MD -0.21, 95% CI -0.37 to -0.06; double PEEP LRM versus routine care: MD -0.18, 95% CI -0.35 to -0.02). It was also demonstrated that a double PEEP LRM may slightly reduce mean arterial pressure at 30 minutes' post-suctioning, when compared with routine care (MD -16.00, 95% CI -29.35 to -2.65). AUTHORS' CONCLUSIONS: There is insufficient evidence to guide the use of LRMs in mechanically ventilated neonates. Well-designed randomised trials with larger sample sizes are needed to further evaluate the potential benefits and risks of LRM application in this population.