Abstract：
BACKGROUND: The SafeBoosC-III trial investigated the effect of cerebral oximetry-guided treatment in the first 72 h after birth on mortality and severe brain injury diagnosed by cranial ultrasound in extremely preterm infants (EPIs). This ancillary study evaluated the effect of cerebral oximetry on global brain injury as assessed by magnetic resonance imaging (MRI) at term equivalent age (TEA).
METHODS: MRI scans were obtained between 36 and 44.9 weeks PMA. The Kidokoro score was independently evaluated by two blinded assessors. The intervention effect was assessed using the nonparametric Wilcoxon rank sum test for median difference and 95% Hodges-Lehmann (HL) confidence intervals (CIs). The intraclass correlation coefficient (ICC) was used to assess the agreement between the assessors.
RESULTS: A total of 210 patients from 8 centers were included, of whom 121 underwent MRI at TEA (75.6% of alive patients): 57 in the cerebral oximetry group and 64 in the usual care group. There was an excellent correlation between the assessors for the Kidokoro score (ICC agreement: 0.93, 95% CI: 0.91-0.95). The results showed no significant differences between the cerebral oximetry group (median 2, interquartile range [IQR]: 1-4) and the usual care group (median 3, IQR: 1-4; median difference -1 to 0, 95% HLCI: -1 to 0; p value 0.1196).
CONCLUSIONS: In EPI, the use of cerebral oximetry-guided treatment did not lead to significant alterations in brain injury, as determined by MRI at TEA. The strong correlation between the assessors highlights the potential of the Kidokoro score in multicenter trials.
METHODS: MRI scans were obtained between 36 and 44.9 weeks PMA. The Kidokoro score was independently evaluated by two blinded assessors. The intervention effect was assessed using the nonparametric Wilcoxon rank sum test for median difference and 95% Hodges-Lehmann (HL) confidence intervals (CIs). The intraclass correlation coefficient (ICC) was used to assess the agreement between the assessors.
RESULTS: A total of 210 patients from 8 centers were included, of whom 121 underwent MRI at TEA (75.6% of alive patients): 57 in the cerebral oximetry group and 64 in the usual care group. There was an excellent correlation between the assessors for the Kidokoro score (ICC agreement: 0.93, 95% CI: 0.91-0.95). The results showed no significant differences between the cerebral oximetry group (median 2, interquartile range [IQR]: 1-4) and the usual care group (median 3, IQR: 1-4; median difference -1 to 0, 95% HLCI: -1 to 0; p value 0.1196).
CONCLUSIONS: In EPI, the use of cerebral oximetry-guided treatment did not lead to significant alterations in brain injury, as determined by MRI at TEA. The strong correlation between the assessors highlights the potential of the Kidokoro score in multicenter trials.
摘要：
背景：SafeBoosC-III试验研究了出生后第一个72小时脑血氧饱和度引导治疗对死亡率和重度脑损伤的影响，通过颅脑超声诊断为早产儿（EPIs）。这项辅助研究评估了在足月等效年龄（TEA）通过磁共振成像（MRI）评估的脑血氧饱和度对整体脑损伤的影响。
方法：在36至44.9周的PMA之间进行MRI扫描。Kidokoro评分由两名盲评估员独立评估。使用非参数Wilcoxon秩和检验中值差异和95％Hodges-Lehmann（HL）置信区间（CI）评估干预效果。使用组内相关系数(ICC)来评估评估者之间的一致性。
结果：共纳入来自8个中心的210例患者，其中121例接受了TEAMRI检查(75.6%的存活患者):57例在脑血氧饱和度测定组,64例在常规护理组.Kidokoro评分的评估者之间存在极好的相关性（ICC一致性：0.93，95％CI：0.91-0.95）。结果显示，脑血氧饱和度测定组（中位数2，四分位距[IQR]：1-4）和常规护理组（中位数3，IQR：1-4；中位数差异-1至0，95％HLCI：-1至0；p值0.1196）之间没有显着差异。
结论：在EPI中，使用脑血氧饱和度指导治疗并没有导致脑损伤的显著改变，在TEA通过MRI确定。评估者之间的强相关性突出了Kidokoro评分在多中心试验中的潜力。
方法：在36至44.9周的PMA之间进行MRI扫描。Kidokoro评分由两名盲评估员独立评估。使用非参数Wilcoxon秩和检验中值差异和95％Hodges-Lehmann（HL）置信区间（CI）评估干预效果。使用组内相关系数(ICC)来评估评估者之间的一致性。
结果：共纳入来自8个中心的210例患者，其中121例接受了TEAMRI检查(75.6%的存活患者):57例在脑血氧饱和度测定组,64例在常规护理组.Kidokoro评分的评估者之间存在极好的相关性（ICC一致性：0.93，95％CI：0.91-0.95）。结果显示，脑血氧饱和度测定组（中位数2，四分位距[IQR]：1-4）和常规护理组（中位数3，IQR：1-4；中位数差异-1至0，95％HLCI：-1至0；p值0.1196）之间没有显着差异。
结论：在EPI中，使用脑血氧饱和度指导治疗并没有导致脑损伤的显著改变，在TEA通过MRI确定。评估者之间的强相关性突出了Kidokoro评分在多中心试验中的潜力。
