文献速递产科母胎医学

胎儿生长受限与小于胎龄儿作为新生儿发病率预测指标:应使用哪种生长曲线图?

Fetal growth restriction and small for gestational age as predictors of neonatal morbidity: which growth nomogram to use?
PubMed 文献 · Am J Obstet Gynecol · 2023-06-20

本研究为回顾性队列研究,纳入2013至2020年间在单一学术中心分娩的10045例非畸形单胎妊娠,比较4种常用生长曲线图与1种机构特定参考预测小于胎龄儿和新生儿复合结局的能力。结果显示各曲线图预测性能相近,估计胎儿体重<10百分位预测小于胎龄儿的阳性似然比为6.3至8.5,曲线下面积为0.62至0.67。

要点

  • 估计胎儿体重<10百分位者小于胎龄儿风险升高(OR 9.9至12.8)。
  • 估计胎儿体重<10和<3百分位者新生儿复合结局风险升高(OR 2.4至3.5和5.7至8.8)。
  • Hadlock、近期生长曲线图与本地参考预测性能相当。

核心结论

  • 在该大型队列中,Hadlock、近期胎儿生长曲线图与本地人群来源的胎儿生长参考在预测小于胎龄儿和新生儿复合结局方面表现相当。

研究设计与样本

  • 回顾性队列研究,纳入2013至2020年间在单一学术中心分娩、于≥20周妊娠接受超声检查的非畸形单胎妊娠,经随机选择限制为每例患者1次妊娠、每次妊娠1次于≥22周妊娠完成的超声,最终纳入10045例参与者。

主要发现

  • 在各曲线图中,被分类为<10百分位的胎儿比例从4.9%至9.7%不等。
  • 估计胎儿体重<10百分位者小于胎龄儿风险升高(OR 9.9 [95%CI 8.5-11.5]至12.8 [95%CI 10.9-15.0])。
  • 估计胎儿体重<10百分位和<3百分位者新生儿复合结局风险升高(OR分别为2.4 [95%CI 2.0-2.8]至3.5 [95%CI 2.9-4.3]和5.7 [95%CI 4.5-7.2]至8.8 [95%CI 6.6-11.8])。
  • 估计胎儿体重<10百分位预测小于胎龄儿的阳性似然比为6.3至8.5,曲线下面积为0.62至0.67;预测新生儿复合结局的阳性似然比为2.1至3.1,曲线下面积为0.55至0.57。
  • 限制于分娩前28天内超声时,<10百分位者小于胎龄儿风险升高(OR 16.7 [95%CI 12.6-22.3]至25.1 [95%CI 17.0-37.0]),预测改善(阳性似然比8.3-15.0;曲线下面积0.69-0.75);新生儿复合结局风险升高(OR 3.2 [95%CI 2.4-4.2]至5.2 [95%CI 3.8-7.2]),预测 marginally 改善(阳性似然比2.4-4.1;曲线下面积0.60-0.62)。

关键数值与推荐

  • 比较了Hadlock、国际胎儿和新生儿生长联盟为21世纪制定的标准、Eunice Kennedy Shriver国家儿童健康与人类发展研究所统一标准、世界卫生组织胎儿生长曲线图以及1种机构特定参考。
  • 估计胎儿体重<10百分位和<3百分位与10至90百分位参考组比较,用于预测小于胎龄儿和新生儿复合结局(围产期死亡、缺氧缺血性脑病或惊厥、呼吸系统疾病、脑室内出血、坏死性小肠结肠炎、需新生儿重症监护室收治的高胆红素血症或低血糖、早产儿视网膜病变)。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Fetal growth nomograms were developed to screen for fetal growth restriction and guide clinical care to improve perinatal outcomes; however, existing literature remains inconclusive regarding which nomogram is the gold standard. OBJECTIVE: This study aimed to compare the ability of 4 commonly used nomograms (Hadlock, International Fetal and Newborn Growth Consortium for the 21st Century, Eunice Kennedy Shriver National Institute of Child Health and Human Development-unified standard, and World Health Organization fetal growth charts) and 1 institution-specific reference to predict small for gestational age and poor neonatal outcomes. STUDY DESIGN: This was a retrospective cohort study of all nonanomalous singleton pregnancies undergoing ultrasound at ≥20 weeks of gestation between 2013 and 2020 and delivering at a single academic center. Using random selection methods, the study sample was restricted to 1 pregnancy per patient and 1 ultrasound per pregnancy completed at ≥22 weeks of gestation. Fetal biometry data were used to calculate estimated fetal weight and percentiles according to the aforementioned 5 nomograms. Maternal and neonatal data were extracted from electronic medical records. Logistic regression was used to estimate the association between estimated fetal weight of <10th and <3rd percentiles compared with estimated fetal weight of 10th to 90th percentile as the reference group for small for gestational age and the neonatal composite outcomes (perinatal mortality, hypoxic-ischemic encephalopathy or seizures, respiratory morbidity, intraventricular hemorrhage, necrotizing enterocolitis, hyperbilirubinemia or hypoglycemia requiring neonatal intensive care unit admission, and retinopathy of prematurity). Receiver operating characteristic curve contrast estimation (primary analysis) and test characteristics were calculated for all nomograms and the prediction of small for gestational age and the neonatal composite outcomes. We restricted the sample to ultrasounds performed within 28 days of delivery; moreover, similar analyses were completed to assess the prediction of small for gestational age and neonatal composite outcomes. RESULTS: Among 10,045 participants, the proportion of fetuses classified as <10th percentile varied across nomograms from 4.9% to 9.7%. Fetuses with an estimated fetal weight of <10th percentile had an increased risk of small for gestational age (odds ratio, 9.9 [95% confidence interval, 8.5-11.5] to 12.8 [95% confidence interval, 10.9-15.0]). In addition, the estimated fetal weight of <10th and <3rd percentile was associated with increased risk of the neonatal composite outcome (odds ratio, 2.4 [95% confidence interval, 2.0-2.8] to 3.5 [95% confidence interval, 2.9-4.3] and 5.7 [95% confidence interval, 4.5-7.2] to 8.8 [95% confidence interval, 6.6-11.8], respectively). The prediction of small for gestational age with an estimated fetal weight of <10th percentile had a positive likelihood ratio of 6.3 to 8.5 and an area under the curve of 0.62 to 0.67. Similarly, the prediction of the neonatal composite outcome with an estimated fetal weight of <10th percentile had a positive likelihood ratio of 2.1 to 3.1 and an area under the curve of 0.55 to 0.57. When analyses were restricted to ultrasound within 4 weeks of delivery, among fetuses with an estimated fetal weight of <10th percentile, the risk of small for gestational age increased across all nomograms (odds ratio, 16.7 [95% confidence interval, 12.6-22.3] to 25.1 [95% confidence interval, 17.0-37.0]), and prediction improved (positive likelihood ratio, 8.3-15.0; area under the curve, 0.69-0.75). Similarly, the risk of neonatal composite outcome increased (odds ratio, 3.2 [95% confidence interval, 2.4-4.2] to 5.2 [95% confidence interval, 3.8-7.2]), and prediction marginally improved (positive likelihood ratio, 2.4-4.1; area under the curve, 0.60-0.62). Importantly, the risk of both being small for gestational age and having the neonatal composite outcome further increased (odds ratio, 21.4 [95% confidence interval, 13.6-33.6] to 28.7 (95% confidence interval, 18.6-44.3]), and the prediction of concurrent small for gestational age and neonatal composite outcome greatly improved (positive likelihood ratio, 6.0-10.0; area under the curve, 0.80-0.83). CONCLUSION: In this large cohort, Hadlock, recent fetal growth nomograms, and a local population-derived fetal growth reference performed comparably in the prediction of small for gestational age and neonatal composite outcomes.

来源:Am J Obstet Gynecol,查看原文。版权归原作者所有。

引用

  1. Leon-Martinez D, Lundsberg LS, Culhane J, et al. Fetal growth restriction and small for gestational age as predictors of neonatal morbidity: which growth nomogram to use? Am J Obstet Gynecol. 2023. PMID: 37348779. doi:10.1016/j.ajog.2023.06.035

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。