目標：使用劑量與效應反應的統合分析之方法來檢視血糖值與牙周病風險之間的劑量與效 應關係。方法：本研究使用三種資料庫：Pubmed、Medline、Embase進行系統性文獻回顧以搜 尋血糖值與牙周病之間證據，最後進入統合分析部分之文獻為23篇。在統合分析部分，先是以 傳統統合分析方法探討高血糖為牙周病的危險因子，接著使用Greenland和Longnecker的廣義最 小平方法，進行劑量與效應反應的統合分析。結果：高血糖為牙周病的危險因子，而血糖值與 牙周病風險之間存在著劑量與效應非線性關係。隨著血糖值的上升，牙周病的風險增加；而當 血糖值上升至一個程度(血糖值約為160 mg/mL)時，牙周病風險會趨於平緩，藉由劑量與效應 反應的統合分析結果顯示並沒有一個血糖值範圍對於牙周病是安全的。結論：以劑量與效應反 應的統合分析探討血糖值與牙周病間的關係，發現血糖值與牙周病間確實存在著劑量與效應 的關聯性，對於糖尿病患的照護亦應特別重視口腔保健並定期追蹤及檢查。

Objectives: The aim of the study was to undertake a systematic review and dose response meta-analysis to evaluate the relationship between glucose levels and the risk of periodontal disease. Methods: We first carried out an electronic search of three databases: Pubmed, Medline, Embase, to examine the relation between glucose and periodontal disease. In total, 517 studies were identified. After screening titles and abstracts, we retrieved the full texts of 147 articles, with 23 articles included in the meta-analysis. We first undertook a traditional meta-analysis on the risk of developing periodontal disease between subjects with high or low glucose levels. The generalized least squares method proposed by Greenland and Longnecker was then used for the dose-response meta-analysis. Results: People with a high glucose level were more likely to have suffered periodontal diseases than those with a low glucose level (odds ratio 2.61, Confidence interval: 2.00 to 3.41). The dose-response meta-analysis showed that glucose level had a nonlinear relation with the risk of periodontal disease. When glucose level reached 160 mg/dl, the risk of developing periodontal disease gradually reached a plateau. Our meta-analysis suggested that there is no safe level of glucose for periodontal disease. Conclusions: Our dose-response metaanalysis found a nonlinear relation between glucose level and an increased risk of periodontal disease. Diabetic patients should pay attention to oral health care and receive regular oral examination.