Supplementary Material for: <b>NET</b>work <b>M</b>eta-analysis Of <b>T</b>rials of <b>I</b>nitial <b>O</b>xygen in preterm <b>N</b>ewborns (NETMOTION): A Protocol for Systematic Review and Individual Participant Data Network Meta-Analysis of Preterm Infants &lt;32 Weeks’ Gestation Randomized to Initial Oxygen Concentration for Resuscitation

<b><i>Background:</i></b> Internationally recognized guidelines recommend the judicious use of low oxygen (21–30%), titrated to peripheral oxygen saturation targets, for the initiation of resuscitation of very and extremely preterm infants (&lt;32 weeks’ gestation). However, despite more than 10 randomized controlled trials on this question, the ideal initial oxygen concentration for this group of vulnerable infants remains uncertain. <b><i>Aims:</i></b> This study aims to assess the effect of various initial oxygen concentrations on (1) all-cause mortality, chronic lung disease, intraventricular hemorrhage, and retinopathy of prematurity; and (2) reaching the prescribed oxygen saturation targets by 5 min after birth, in preterm infants requiring resuscitation. <b><i>Methods:</i></b> We will conduct a systematic review and network meta-analysis using individual participant data. Studies of preterm infants &lt;32 weeks’ gestation, randomized to initial oxygen concentration, will be included. We will systematically search medical databases and trial registries for eligible studies (published or unpublished). Records will be screened by two independent reviewers, with conflicts resolved by the inclusion of a third reviewer. Identified initial oxygen concentrations will be grouped into the following nodes: low (≤30%), intermediate (60%), and high (≥90%) oxygen. A two-step random-effects contrast-based network meta-regression will be calculated to compare and rank different oxygen concentrations. Analyses will be intention-to-treat, with the primary outcome of all-cause mortality. <b><i>Discussion:</i></b> This is the first individual participant data network meta-analysis of initial oxygen concentrations for the resuscitation of preterm infants. This novel approach may address long-standing uncertainty regarding optimal oxygen supplementation practice for the resuscitation of preterm infants &lt;32 weeks’ gestation.背景：国际公认指南建议审慎使用低浓度氧（21%~30%），并根据外周血氧饱和度目标进行滴定，用于极早早产儿（妊娠<32周）的初始复苏。然而，尽管已有超过10项针对该问题的随机对照试验 (randomized controlled trials)，但这类脆弱早产儿的理想初始氧浓度仍不明确。研究目的：本研究旨在评估不同初始氧浓度对需复苏的早产儿的两项结局的影响：(1) 全因死亡率、慢性肺疾病、脑室内出血以及早产儿视网膜病；(2) 出生后5分钟内达到预设的血氧饱和度目标。研究方法：本研究将采用个体参与者数据 (individual participant data) 开展系统评价 (systematic review) 与网状meta分析 (network meta-analysis)。纳入妊娠<32周的早产儿被随机分配至不同初始氧浓度组的相关研究。我们将系统检索医学数据库与试验注册库，以获取符合条件的已发表或未发表研究。由两名独立评审员筛选文献，若存在分歧则引入第三名评审员协助解决。将明确的初始氧浓度划分为以下组别：低浓度（≤30%）、中浓度（60%）与高浓度（≥90%）。将采用基于对比的两步随机效应网状meta回归模型，以比较并排序不同氧浓度的疗效。分析将遵循意向性治疗 (intention-to-treat) 原则，主要结局为全因死亡率。讨论：本研究是首个针对极早早产儿（妊娠<32周）复苏时初始氧浓度的个体参与者数据网状meta分析。这种新颖的研究方法或可解决长期以来关于极早早产儿复苏时最佳氧补充策略的不确定性。