[Safety evaluation on different ventilation strategies set for neonatal respiratory distress syndrome: a network Meta-analysis]

Objective: To evaluate the relative safety of different ventilation methods regarding mortality and rates of complication, on neonatal respiratory distress syndrome (NRDS). Methods: Network Meta-analysis was used to collect data on randomized controlled trials of pulmonary ventilation strategies in preterm infants with a mean gestational age of less than 32 weeks. Diagnostic criteria on NRDS were published in the PubMed, Cochrane, Web of Science, EBSCO, and Springer Link databases from January 1986 to June 2018. Revman 5.3 software was used to evaluate the quality of studies, based on the Cochrane quality assessment tool. Data were analyzed by Bayesian and frequency methods, using both Win BUGS 1.4.3 and STATA 13.0 software. Safety of different ventilation strategies for NRDS mortality and complications would include intraventricular hemorrhage (IVH), patent ductus arteriosus (PDA) and retinopathy of prematurity (ROP) and were evaluated. Counted data was displayed by OR and 95%CI. Results: A total of 31 RCTs were included in this paper, including 5 827 preterm infants and 11 ventilation strategies. There were no statistically significant differences appearing in 11 ventilation strategies on mortality, PDA or ROP. IVH results were reported in 28 studies. Compared with nasal intermittent positive pressure ventilation (NIPPV), both high- frequency oscillation ventilation (HFOV) (OR=3.33, 95%CI: 1.08-16.67, P<0.05) and synchronized intermittent mechanical ventilation (SIMV) (OR=8.22, 95%CI: 1.25-29.44, P<0.05) schemes seemed to have increased the risk of IVH in preterm infants with NRDS. NIPPV appeared the optimal ventilation strategy in the rankings of cumulative probability. Results on clustering showed that NIPPV was probably the best ventilation strategy for children with NRDS after considering the orders of IVH, PDA and ROP on mortality, respectively. However, HFOV, IMV, and SIMV did not seem to be the ideal ventilated strategies. Conclusions: Most of the clinical decision makers might prefer using NIPPV in the treatment of children with NRDS through mechanical ventilation systems to reduce both the incidence and death caused by IVH, PDA and ROP. It was not recommended to use HFOV, SIMV and IMV in treating NRDS with gestational less than 32 weeks. We suggested that larger numbers of multi-center RCTs ba carried out to make the above conclusions more convincing.