Scholarlink

Background <p>Probable meningitis, operationally defined by meningitis-consistent cerebrospinal fluid (CSF) abnormalities in the absence of CSF culture confirmation, is a clinically important complication of early-onset sepsis (EOS) in very preterm and/or very low birth weight (VLBW) infants.</p> Methods <p>In this retrospective multicenter cohort study, we included infants with gestational age &lt; 32&#xa0;weeks and/or birth weight &lt; 1500&#xa0;g diagnosed with EOS and admitted to three tertiary hospitals in northern China between January 2023 and December 2024. All infants underwent lumbar puncture. The primary outcome was probable meningitis, operationally defined as meningitis-consistent CSF abnormalities without CSF culture confirmation, specifically culture-negative CSF with pleocytosis, neutrophil predominance, hypoglycorrhachia, and elevated protein. Candidate predictors available at birth or within the first 24&#xa0;h were prespecified. Least absolute shrinkage and selection operator (LASSO) regression with cross-validation was used for predictor selection, followed by multivariable logistic regression to develop a nomogram. Model discrimination, calibration, and clinical utility were evaluated using the C-index/AUC, Brier score, calibration plots, and decision-curve analysis with bootstrap internal validation.</p> Results <p>Among 359 eligible infants, 88 met the study definition of probable meningitis. LASSO identified 17 candidate predictors, and the final model retained four variables: mode of delivery (cesarean delivery vs vaginal delivery), blood lactate, C-reactive protein, and procalcitonin. The nomogram demonstrated good discrimination (apparent C-index/AUC 0.861; 95% CI 0.819–0.904 [DeLong]; optimism-corrected AUC 0.855), acceptable overall performance (Brier score 0.124), and good calibration (optimism-corrected intercept − 0.035; slope 0.954).</p> Conclusions <p>This nomogram showed good discrimination and calibration for predicting probable meningitis in very preterm and/or VLBW infants with clinical EOS. It may support early risk stratification and prioritization of lumbar puncture when clinically feasible, but external validation is required before wider clinical use.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A nomogram to predict probable meningitis in very preterm and/or very low birth weight infants with early-onset sepsis: a multicenter retrospective study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月1日

查看全文
  • Xiaowei Sun,
  • Rui Jing,
  • Zihan Zhuang,
  • Jialin Wen,
  • Wenying Meng,
  • Yang Li

摘要

Background

Probable meningitis, operationally defined by meningitis-consistent cerebrospinal fluid (CSF) abnormalities in the absence of CSF culture confirmation, is a clinically important complication of early-onset sepsis (EOS) in very preterm and/or very low birth weight (VLBW) infants.

Methods

In this retrospective multicenter cohort study, we included infants with gestational age < 32 weeks and/or birth weight < 1500 g diagnosed with EOS and admitted to three tertiary hospitals in northern China between January 2023 and December 2024. All infants underwent lumbar puncture. The primary outcome was probable meningitis, operationally defined as meningitis-consistent CSF abnormalities without CSF culture confirmation, specifically culture-negative CSF with pleocytosis, neutrophil predominance, hypoglycorrhachia, and elevated protein. Candidate predictors available at birth or within the first 24 h were prespecified. Least absolute shrinkage and selection operator (LASSO) regression with cross-validation was used for predictor selection, followed by multivariable logistic regression to develop a nomogram. Model discrimination, calibration, and clinical utility were evaluated using the C-index/AUC, Brier score, calibration plots, and decision-curve analysis with bootstrap internal validation.

Results

Among 359 eligible infants, 88 met the study definition of probable meningitis. LASSO identified 17 candidate predictors, and the final model retained four variables: mode of delivery (cesarean delivery vs vaginal delivery), blood lactate, C-reactive protein, and procalcitonin. The nomogram demonstrated good discrimination (apparent C-index/AUC 0.861; 95% CI 0.819–0.904 [DeLong]; optimism-corrected AUC 0.855), acceptable overall performance (Brier score 0.124), and good calibration (optimism-corrected intercept − 0.035; slope 0.954).

Conclusions

This nomogram showed good discrimination and calibration for predicting probable meningitis in very preterm and/or VLBW infants with clinical EOS. It may support early risk stratification and prioritization of lumbar puncture when clinically feasible, but external validation is required before wider clinical use.