請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/70861
標題: | 探討糖尿病前期是否增加罹患肺結核風險:世代研究 Association between prediabetes and tuberculosis: a cohort study |
作者: | Tsun-Hao Ko 柯尊皓 |
指導教授: | 林先和(Hsien-Ho Lin) |
關鍵字: | 結核病,空腹血糖異常,糖尿病前期,縱貫性世代追蹤研究,劑量反應關係, TB,impaired fasting glucose,intermediate hyperglycemia,longitudinal cohort design,dose response relationship, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 背景
第二型糖尿病是結核病的危險因子之一。近年來由於人口老化以及生活習慣改變,第二型糖尿病的盛行率逐年上升,此現象在開發中國家更為明顯。因此,第二型糖尿病在開發中國家的結核病防治扮演關鍵的角色。儘管糖尿病在過去研究中已經確認為結核病的危險因子,鮮少研究探討糖尿病前期對結核病防治可能產生的影響。本研究的目的為探討糖尿病前期與結核病之間是否存在相關性。 方法 研究對象來自新北市社區性健康檢查的受試者(n=124,455),空腹血糖大於等於126 mg/dl或者有降血糖藥物使用紀錄定義為糖尿病,空腹血糖大於等於100md/dl且小於126 mg/dl則定義為糖尿病前期。新發生結核病個案是利用台灣疾病管制署之結核病通報系統定義。統計分析方面,利用inverse probability weighting校正過之Kaplan-Meier 存活曲線觀察正常血糖、糖尿病前期以及糖尿病受試者結核病發生之差異。並且利用Cox 比例風險模式探討糖尿病前期與糖尿病相較於正常血糖發生結核病的風險比值。最後透過spline回歸分析觀察空腹血糖值與結核病風險的劑量反應關係。 結果 在119,352名受試者中有27,404名糖尿病前期(22.96%)及10,943名糖尿病患者(9.17%)。經過平均7.2年的追蹤,總共有322名結核病新發個案。在Cox比例風險模式中,控制其他共變項後,相較於正常血糖受試者,糖尿病前期的風險比值為0.73 (95% CI 0.545-0.965),糖尿病的風險比值為1.48 (95% CI 1.105-1.978)。在spline迴歸分析中,空腹血糖及結核病風險呈現U型曲線關係,在空腹血糖為110 mg/dl時罹患結核病風險最低。最後針對潛在干擾因子、暴露變項的定義以及排除條件進行敏感度分析都呈現一致性結果。 結論 糖尿病前期的受試者罹患結核病的風險低於正常血糖者27%,然而未來仍需要更多研究了解糖尿病前期與罹患結核病風險之間的相關性。 Introduction Tuberculosis remains one of the top ten causes of death globally, and more than 10 million incident cases were reported annually in the past decade. Type 2 diabetes (DM) is a notable risk factor for active tuberculosis (TB). Prediabetes is a metabolic condition that the blood sugar level is higher than normal but not yet to develop DM. Although DM has been shown as one of risk factors for TB, much less is known about the impact of prediabetes on TB control (nature history). The aim of this study is to investigate the relationship between prediabetes and risk of active TB. Methods Study population were from the community-based health screening service in northern Taiwan. Baseline fasting plasma glucose (FPG) and prescription of hypoglycemic agents were used to define DM and prediabetes (measured as impaired fasting glucose). Incident cases of TB were identified by linking to the National Tuberculosis Registry. Kaplan-Meier curves were conducted to compare TB-free survival by DM status. Cox regression analysis was used to estimate the hazards ratio for prediabetes and DM compared to normoglycemia. Spline regression was used to investigate the dose-response relationship between FPG level and risk of active TB. Results Of 119,352 participants, 27,404 (22.96%) were prediabetic, and 10,943 (9.17%) had DM. After an average 7.2 years of follow-up, 322 tuberculosis cases occurred. Compared to normoglycemic individuals, the adjusted hazards ratio (aHR) was 0.73 (95% confidence interval (CI) 0.55-0.97) for prediabetes subjects and 1.48 (95% CI 1.11-1.98) for DM patients. In the spline regression, a U-shape association was found between FPG level and the TB risk among people without diabetes, and the risk of TB was the lowest at FPG level of around 110 mg/dl. Sensitivity analyses revealed that the lower risk of TB among prediabetics could not be explained other potential confounders, misclassification of blood glucose level, and selection bias. Conclusion Prediabetic subjects have a 27% lower risk of active TB compared to normoglycemic subjects. More studies are needed to understand the relationship between prediabetes and TB. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/70861 |
DOI: | 10.6342/NTU201802184 |
全文授權: | 有償授權 |
顯示於系所單位: | 流行病學與預防醫學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-1.pdf 目前未授權公開取用 | 1.72 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。