Supplementary Material for: Agreement between Noninvasive Hemoglobin and Laboratory Hemoglobin Measurements in Neonates: A Systematic Review and Meta-Analysis

<b><i>Background and Objective:</i></b> The noninvasive hemoglobin (NHb) devices are recently evaluated as an alternative to laboratory hemoglobin (LHb) in neonates. This systematic review explores the diagnostic accuracy of NHb devices for neonatal hemoglobin measurement. <b><i>Methods:</i></b> Literature related to the comparison of NHb device with LHb in neonates was searched from Medline, PubMed Central, PubMed, Web of Science, Google Scholar, and Scopus databases after PROSPERO registration. The quality of included publications was assessed by QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies). The pooled correlation coefficient and bias (precision) in Bland-Altman difference plots were used for summary statistics using MetaXL 5.0 software. <b><i>Results:</i></b> A total of 1,477 paired NHb-LHb observations were analyzed from 1,047 neonates in 10 studies. Massimo radical-pulse co-oximetry (8 studies) and Mediscan-2000 (2 studies) were used for NHb estimation. The pooled correlation coefficient between NHb and LHb was <i>r</i> = 0.94 (95% CI: 0.83–0.98, <i>p</i> &lt; 0.001), and the pooled bias (precision) was −0.013 (1.4) gm/dL between NHb and LHb measurements in Bland-Altman analysis. NHb device had better precision in stable neonates (0.91gm/dL) over sick neonates (1.66 gm/dL). <b><i>Conclusions:</i></b> Hemoglobin measurement by NHb is excellently correlated with LHb measurement with a minimal average difference. It may be used as a screening tool for hemoglobin measurement in neonates to avoid frequent phlebotomy.

**背景与目的：** 无创血红蛋白（noninvasive hemoglobin, NHb）设备近年来被评估为新生儿实验室血红蛋白（laboratory hemoglobin, LHb）检测的替代方案。本系统综述旨在探讨无创血红蛋白设备用于新生儿血红蛋白检测的诊断准确性。

**方法：** 本研究于PROSPERO注册后，从Medline、PubMed Central、PubMed、Web of Science、Google Scholar及Scopus数据库中检索针对新生儿无创血红蛋白设备与实验室血红蛋白检测对比的相关文献。纳入文献的质量采用QUADAS-2（诊断准确性研究质量评价工具，Quality Assessment of Diagnostic Accuracy Studies）进行评估。采用MetaXL 5.0软件，通过Bland-Altman差值图计算合并相关系数与偏倚（精密度）作为汇总统计量。

**结果：** 本系统综述共纳入10项研究，涉及1047名新生儿，共计分析1477组无创血红蛋白-实验室血红蛋白配对检测数据。无创血红蛋白检测采用马西莫自由基脉冲血氧定量法（Massimo radical-pulse co-oximetry，共8项研究）及Mediscan-2000设备（共2项研究）。Bland-Altman分析结果显示，无创血红蛋白与实验室血红蛋白检测的合并相关系数为*r*=0.94（95%置信区间：0.83~0.98，*p*<0.001），合并偏倚（精密度）为-0.013（1.4）gm/dL。无创血红蛋白检测设备在病情稳定新生儿中的精密度（0.91gm/dL）优于患病新生儿（1.66gm/dL）。

**结论：** 无创血红蛋白检测与实验室血红蛋白检测具有极佳的相关性，且平均差值极小。该设备可作为新生儿血红蛋白检测的筛查工具，以避免频繁的静脉采血。