Abstract: Mendelian randomization（MR）is an analytical approach that utilizes genetic variants as instrumental variables to assess causal relationships between exposures and outcomes. Its core principle is to leverage the random allocation of genetic variants during gamete formation，effectively mitigating confounding factors，reverse causation，and measurement errors inherent in traditional observational studies，thereby enhancing the reliability of causal inference. Periodontal disease，as a multifactorial-mediated chronic inflammatory condition，relies predominantly on epidemiological observational studies to identify risk factors（e.g.，obesity，metabolic disorders），yet such methods struggle to establish definitive causal links. Furthermore，evidence regarding the impact of periodontal disease on systemic conditions（e.g.，cardiovascular diseases，diabetes）exhibits heterogeneity，and the reproducibility of conclusions remains incompletely established. Consequently，MR provides a novel causal inference paradigm for periodontal research，demonstrating unique advantages，particularly in elucidating bidirectional associations between periodontal diseases and systemic disorders. This review summarizes the advances in MR applications in periodontology，encompassing studies on pathogenic mechanisms，risk factor analyses，and systemic implications of periodontal diseases. Nevertheless，MR findings still require validation through triangulation of multiple evidence sources（e.g.，experimental verification，prospective cohort studies）to strengthen robustness of conclusions. Future research should optimize instrumental variable selection，expand sample sizes，and integrate multi-omics data to deepen mechanistic exploration，thereby providing a theoretical foundation for precision prevention/control and interdisciplinary intervention of periodontal diseases.

Key words: Mendelian randomization, periodontitis, causal inference, systemic diseases, genetics

摘要： 孟德尔随机化（Mendelian randomization，MR）是一种将遗传变异作为工具变量，评估暴露因素与结局间因果关系的分析方法；其核心原理是利用遗传变异在配子形成时的随机分配特性，有效规避传统观察性研究中混杂因素、反向因果及测量误差的干扰，从而增强因果推断的可靠性。牙周病作为多因素介导的慢性炎症性疾病，其风险因素（如肥胖、代谢异常等）多依赖流行病学观察性研究识别，但此类方法难以确证因果关联。此外，关于牙周病对全身性疾病（如心血管疾病、糖尿病等）影响的证据存在异质性，结论的可重复性尚未完全确立。因此，MR为牙周病研究提供了新的因果推断范式，尤其在解析牙周病与系统性疾病的双向关联中展现出独特优势。文章就MR在牙周病学领域的应用进展做一综述，涵盖牙周病致病机制、风险因素解析及全身性影响的相关研究。然而，MR结果仍需通过多证据三角验证（如实验验证、前瞻性队列研究），以增强结论稳健性。未来研究需优化工具变量选择、扩大样本规模，并结合多组学数据深化机制探索，从而为牙周病的精准防控及跨学科干预提供理论支撑。

关键词: 孟德尔随机化, 牙周炎, 因果推断, 全身疾病, 遗传学