Frontiers in endocrinology · 2025
Risk prediction model for cervical lymph node metastasis of papillary thyroid microcarcinoma: a systematic review and meta-analysis.
СтатьяСистематический обзорМетаанализНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
He Xiaoli, Zhang Qiang, Yang Kaiju, Li Jian, Luo Mingzhu, Liu Ruihan, Yang Xi
Аннотация
BACKGROUND: A growing number of risk prediction models for cervical lymph node metastasis (CLNM) in papillary thyroid microcarcinoma (PTMC) have been developed, but their performance and methodological rigor remain unclear. This study systematically reviews these models to evaluate their predictive performance and critically appraise their risk of bias.
METHODS: We conducted a systematic search of seven databases up to July 29, 2025. The methodological quality of the included studies was assessed using PROBAST. Model performance, measured by the area under the curve (AUC), was pooled using a random-effects meta-analysis.
RESULTS: A total of 15 studies, comprising 24 predictive models, were included. The pooled AUC was 0.794 (95% CI: 0.769-0.820), but with substantial heterogeneity (I 2 = 89.6%). Subgroup analysis revealed a performance drop from the training set (pooled AUC, 0.812) to the validation set (pooled AUC, 0.774). The PROBAST assessment revealed that 12 of the 15 studies (80%) were critically at a high risk of bias, primarily due to flaws in participant selection.
CONCLUSION: Although existing CLNM prediction models for PTMC show moderate to good discrimination on average, their clinical utility is severely limited by widespread methodological weaknesses and a high risk of bias. The current evidence is not robust enough to recommend any specific model for routine clinical use, and future research must prioritize methodological rigor and independent external validation.
METHODS: We conducted a systematic search of seven databases up to July 29, 2025. The methodological quality of the included studies was assessed using PROBAST. Model performance, measured by the area under the curve (AUC), was pooled using a random-effects meta-analysis.
RESULTS: A total of 15 studies, comprising 24 predictive models, were included. The pooled AUC was 0.794 (95% CI: 0.769-0.820), but with substantial heterogeneity (I 2 = 89.6%). Subgroup analysis revealed a performance drop from the training set (pooled AUC, 0.812) to the validation set (pooled AUC, 0.774). The PROBAST assessment revealed that 12 of the 15 studies (80%) were critically at a high risk of bias, primarily due to flaws in participant selection.
CONCLUSION: Although existing CLNM prediction models for PTMC show moderate to good discrimination on average, their clinical utility is severely limited by widespread methodological weaknesses and a high risk of bias. The current evidence is not robust enough to recommend any specific model for routine clinical use, and future research must prioritize methodological rigor and independent external validation.