文献速递小儿神经外科与脑积水神经重症

去骨瓣减压术后脑室腹腔分流与颅骨修补时序的网络荟萃分析

Optimal timing of ventriculoperitoneal shunt insertion relative to cranioplasty post-decompressive craniectomy: a frequentist network meta-analysis.
PubMed 文献 · Acta Neurochir (Wien) · 2026-05-12

结果显示同期CP-VPS较CP后VPS总体并发症风险显著更高(RR 2.07;95% CI 1.16-3.71;P<0.05),再手术、感染及硬膜内出血风险各时序策略间无显著差异。证据支持分期重建策略,即先行CP恢复颅内顺应性及脑脊液动力学,VPS留待持续性或进展性脑积水时进行。

要点

  • 同期CP-VPS较CP后VPS总体并发症风险显著更高(RR 2.07;95% CI 1.16-3.71;P<0.05)。
  • 再手术、感染及硬膜内出血风险在三种时序策略间无显著差异。
  • 证据支持先行CP、VPS留待持续性或进展性脑积水时进行的分期策略。

核心结论

  • 同期CP-VPS较CP后VPS总体并发症风险显著更高(RR 2.07;95% CI 1.16-3.71;P<0.05)。
  • 分期策略(VPS后CP或CP后VPS)在感染、再手术及出血风险上总体相当。
  • 证据支持先行CP恢复颅内顺应性及脑脊液动力学、VPS留待持续性或进展性脑积水的分期重建模式。

研究设计与样本

  • 291例(54.7%)为CP前VPS,156例(29.3%)为同期CP-VPS,85例(16.0%)为CP后VPS。
  • 采用Mantel-Haenszel法合并,设计-处理交互模型及节点拆分评估传递性与一致性,CINeMA分级证据质量。

主要发现

  • 同期CP-VPS较CP后VPS总体并发症风险显著更高(RR 2.07;95% CI 1.16-3.71;P<0.05)。
  • 再手术风险68/386(17.6%)、感染风险62/532(11.7%)、硬膜内出血风险22/509(2.8%),各时序策略间无显著差异。
  • 次要结局报告较少,组间无统计学显著差异。
  • 研究间异质性与网络不一致性低,主要比较(同期 vs CP后VPS)证据质量评为高。

关键数值与推荐

  • 同期CP-VPS较CP后VPS总体并发症风险显著更高(RR 2.07;95% CI 1.16-3.71;P<0.05)。
  • 再手术(68/386,17.6%)、感染(62/532,11.7%)、硬膜内出血(22/509,2.8%)风险各时序策略间无显著差异。
  • 主要比较(同期 vs CP后VPS)证据质量CINeMA评级为高。

临床怎么用

  • 去骨瓣减压术后需VPS者,优先考虑分期策略,先行CP恢复颅内顺应性及脑脊液动力学。
  • 避免在CP同期常规置入VPS,因同期CP-VPS总体并发症风险显著高于CP后VPS。
  • VPS留待持续性或进展性脑积水时进行,而非CP时常规置入。

局限性

  • 仅纳入观察性研究,存在选择偏倚等混杂可能。
  • 次要结局报告不频繁,相关比较把握度有限。
  • 样本量532例,各时序亚组样本不均(CP后VPS仅85例)。

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

OBJECTIVE: Optimal timing of ventriculoperitoneal shunting (VPS) relative to cranioplasty (CP) after decompressive craniectomy remains controversial. VPS may be performed before, during, or after CP, but no consensus exists on which sequence minimizes complications. We performed a systematic review and network meta-analysis to compare the safety profiles of pre-, simultaneous, and post-CP shunting. METHODS: We searched PubMed/MEDLINE, Embase, and CENTRAL from inception to October 2025. Eligible studies directly compared at least two timing strategies in patients undergoing both CP and VPS. Primary outcomes included overall complications, reoperation, infection, and intradural bleeding. Secondary outcomes included shunt-related and other procedure-specific adverse events. We used a frequentist random-effects network meta-analysis with Mantel-Haenszel pooling. Transitivity and consistency were assessed using design-by-treatment interaction modeling and node-splitting. Certainty of evidence was graded using CINeMA. RESULTS: Ten observational studies met inclusion criteria, comprising 532 patients: 291 (54.7%) pre-CP VPS, 156 (29.3%) simultaneous CP-VPS, and 85 (16.0%) post-CP VPS. Simultaneous CP-VPS was associated with a significantly higher overall complication risk compared with post-CP VPS (RR 2.07; 95% CI 1.16-3.71; P < 0.05). Risk of reoperation (68/386, 17.6%), infection (62/532, 11.7%), and intradural bleeding (22/509, 2.8%) did not differ significantly across timing strategies. Secondary outcomes were infrequently reported and showed no statistically significant differences between groups. Inter-study heterogeneity and network inconsistency were low. Certainty of evidence for the main comparison (simultaneous versus post-CP VPS) was rated high. CONCLUSION: Simultaneous CP-VPS carried a higher composite complication risk than post-CP VPS, whereas staged strategies (VPS followed by CP or CP followed by VPS) demonstrated broadly comparable risk of infection, reoperation, and hemorrhage. These findings support a staged reconstruction paradigm in which CP is performed first to restore intracranial compliance and CSF hydrodynamics, with VPS reserved for persistent or progressive hydrocephalus rather than being routinely inserted at the time of CP.

来源:Acta Neurochir (Wien),查看原文。版权归原作者所有。

引用

  1. Patel S, Nischal SA, Kale KM, et al. Optimal timing of ventriculoperitoneal shunt insertion relative to cranioplasty post-decompressive craniectomy: a frequentist network meta-analysis. Acta Neurochir (Wien). 2026. PMID: 42118179. doi:10.1007/s00701-026-06885-5

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。