Background: Solid renal masses are commonly managed through surgical intervention, while conservative management is the preferred approach for simple renal cysts. However, the optimal management strategies for complex renal cysts in children remain unresolved.

Objective: To conduct a meta-analysis of the literature on the modified Bosniak (mBosniak) classification system to assess its diagnostic performance in pediatric complex renal cysts.

Data sources and study eligibility criteria: A predefined database search (January 1986-December 2023), used the following keywords: Bosniak classification, modified Bosniak, pediatric renal cysts, complex renal cyst, and pediatric renal tumor. Two independent reviewers extracted data on mBosniak categories and their associated pathology diagnoses. The inclusion criteria comprised (1) English-language publications, (2) involving the pediatric population, (3) using the mBosniak classification and (4) incorporating histopathological data.

Study appraisal and synthesis methods: The quality of studies was assessed using the QUADAS-2 tool. The analysis involved pooling data, performing subgroup analyses, and exploring potential publication bias.

Results: A total of 7 retrospective studies, which included 154 lesions and 368 scan readings, were included. The overall prevalence of intermediate/malignant lesions was 70% (257/368). In distinguishing benign from intermediate/malignant pathology using the mBosniak classification (I-II vs. III-IV), the overall pooled sensitivity, specificity, PPV, and NPV were 0.88 (227/257), 0.74 (82/111), 0.89 (227/256), and 0.73 (82/112), respectively. mBosniak III category showed a PPV of 0.78 (70/90) and mBosniak II category showed a NPV of 0.59 (36/61). From the perspective of pathological diagnoses, 0.24 (29/122) of the readings for intermediate-risk lesions inaccurately categorized the lesions as mBosniak I-II. The diagnostic performance of the classification varies among distinct subgroups of patients and lesion characteristics.

Limitations: We included studies with pathologically verified lesions only. In addition, aspects of the radiographic follow-up, such as the duration of follow-up, changes in cyst complexity or size over time, and the correlation between specific dynamics and malignancy, could not be evaluated based on the data provided in the included studies.

Conclusions and implications of key findings: The mBosniak classification is a less precise tool for stratifying the challenging cases of intermediate-risk lesions, and the relatively low predictive values of the middle categories (II-III) cast doubt on its utility. Clinicians should incorporate additional clinical-epidemiological data for a more accurate assessment of these lesions. Registration number CRD42023493223.

Keywords: Cysts; Kidney neoplasms; Pediatrics; Radiology; Renal carcinoma.

© 2025. The Author(s), under exclusive licence to International Pediatric Nephrology Association.

背景：

实体肾瘤通常通过手术干预进行管理，而单纯性肾囊肿则倾向于采取保守治疗。然而，对于儿童复杂性肾囊肿的最佳管理策略仍然没有定论。

目的：

对改良版Bosniak分类系统在儿科复杂性肾囊肿中的诊断性能进行文献荟萃分析。

数据来源和研究纳入标准：

预定义的数据库搜索（1986年1月至2023年12月），关键词包括：Bosniak分类、改良版Bosniak分类、儿科肾囊肿、复杂性肾囊肿以及儿科肾脏肿瘤。两位独立评审员提取了关于mBosniak分类及其相关病理诊断的数据。纳入标准包括（1）英文出版物；（2）涉及儿科人群；（3）使用mBosniak分类系统；（4）包含组织病理学数据。

研究评估和综合方法：

使用QUADAS-2工具对各研究的质量进行了评估。分析包括汇总数据、进行亚组分析以及探索潜在的出版偏差。

结果：

共纳入了7项回顾性研究，涉及154个病变和368次扫描读数。中等/恶性病变的整体发生率为70%（257/368）。在使用mBosniak分类系统（I-II类 vs III-IV类）区分良性与中度/恶性病理时，总体汇总的灵敏性、特异性、阳性预测值和阴性预测值分别为：0.88（227/257）、0.74（82/111）、0.89（227/256）和0.73（82/112）。mBosniak III类的阳性预测值为0.78（70/90），而mBosniak II类的阴性预测值为0.59（36/61）。从病理诊断的角度来看，中等风险病变中有0.24（29/122）次读数错误地将病变归类为mBosniak I-II。分类系统的诊断性能在不同患者亚组和病变特征之间存在差异。

局限性：

我们仅纳入了病理学验证的病变更相关研究。此外，放射学随访方面的内容（如随访时间、囊肿复杂度或大小的变化及特定动态与恶性肿瘤之间的关联）无法基于所包括的研究数据进行评估。

结论和关键发现的影响：

改良版Bosniak分类系统对具有中等风险的复杂性病例分层不够精确，其中间类别（II-III）相对较低的预测值对其实用性提出了质疑。临床医生应结合额外的流行病学数据以更准确地评估这些病变。注册号 CRD42023493223。

关键词：

囊肿；肾肿瘤；儿科；放射学；肾癌。

© 2025. The Author(s), under exclusive licence to International Pediatric Nephrology Association.