Department of Neurosurgery, Lenox Hill Hospital/Northwell Health, New York, NY, USA.
Department of Medicine, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil.
Neurosurg Rev. 2024 Oct 3;47(1):722. doi: 10.1007/s10143-024-02952-6.
Meta-analyses are highly valued in medical science, yet accurately reporting complications in neurosurgical studies remains challenging. Examples include inconsistencies in defining and classifying complications and variations in reporting methods. This lack of reproducibility and comparability, along with other issues related to biases, hinders the ability of meta-analyses to yield significant advancements. This systematic review investigated the challenges and limitations inherent in meta-analyses of complications in neurosurgery. Based on the identified challenges and our group's experience, we developed a practical checklist to mitigate and avoid common errors in meta-analyses of complications in neurosurgery.We searched PubMed, Embase, and Web of Science for studies addressing challenges in assessing complications in neurosurgery. The main findings were qualitatively synthesized to identify common challenges and limitations. The proposed checklist was developed using a modified Delphi technique. Eleven studies were included, uncovering heterogeneity and a lack of standardization regarding the classification of complications in neurosurgery across various authors and institutions. They suggested solutions such as implementing a more uniform classification system. Additionally, the NeuroComp Meta-Analysis Checklist was developed, comprising 23 items divided into 5 domains, with a practical approach and suggestions on how to deal with the challenges when meta-analyzing.We identified numerous challenges and concerns when assessing complications in the neurosurgical field. The NeuroComp Meta-Analysis Checklist incorporated methodologies and approaches we utilized in several previously published meta-analyses. While we acknowledge that the proposal cannot solve all the issues involved in comparing and meta-analyzing complications in neurosurgery, it has the potential to enhance the informativeness of future meta-analyses and help authors mitigate common errors. Ultimately, this tool has the potential to contribute to the advancement of accumulating real-world evidence in neurosurgical science.
Meta 分析在医学科学中具有很高的价值，但准确报告神经外科研究中的并发症仍然具有挑战性。例如，并发症的定义和分类不一致，以及报告方法的变化。这种缺乏可重复性和可比性，以及与偏见相关的其他问题，阻碍了荟萃分析取得重大进展的能力。本系统评价调查了神经外科并发症荟萃分析中固有的挑战和局限性。基于确定的挑战和我们小组的经验，我们开发了一份实用的清单，以减轻和避免神经外科并发症荟萃分析中的常见错误。我们在 PubMed、Embase 和 Web of Science 中搜索了评估神经外科并发症的挑战的研究。主要发现进行了定性综合，以确定常见的挑战和局限性。使用改良 Delphi 技术制定了拟议的清单。共纳入 11 项研究，发现不同作者和机构在神经外科并发症分类方面存在异质性和缺乏标准化。他们提出了一些解决方案，例如实施更统一的分类系统。此外，开发了神经并发症荟萃分析清单，包含 23 个项目，分为 5 个领域，具有实用的方法和处理荟萃分析中挑战的建议。我们在评估神经外科领域的并发症时发现了许多挑战和关注点。神经并发症荟萃分析清单纳入了我们在之前发表的几项荟萃分析中使用的方法和方法。虽然我们承认该提案不能解决神经外科并发症比较和荟萃分析中涉及的所有问题，但它有可能提高未来荟萃分析的信息量，并帮助作者减轻常见错误。最终，该工具有可能为神经外科科学中积累真实世界证据的进展做出贡献。
