許多大型世代研究指出血清中儲鐵蛋白(Ferritin)及鐵離子濃度較高者發生大腸病變或腸道相關惡性腫瘤風險有正向相關,而Transferrin則呈現負向相關。血液透析病患常因輸血而有血鐵質沉積,但其與腸道病變相關性的研究,目前尚無相關文獻報告。本篇研究欲評估血液透析病患血清中儲鐵蛋白(Ferritin)與發生腸道病變的風險。本研究收集自2004年9月至2011年12月於台灣北部某醫學中心進行2次以上血液透析治療之病患,並且排除在進行血液透析前已發生腸道病變者12名,我們總共收集740人(女性364人,男性376人),分別紀錄其年齡、性別、生化及Ferritin檢驗結果,追蹤時間之中位數為42.136.21個月。我們將初次Ferritin檢驗結果分為4組作為暴露因子,與病患於本院進行腸道病理檢驗作為事件發生,截止時間為最後一筆血液透析相關生化檢驗及Ferritin檢查開立時間,本篇研究設計為回溯性世代追蹤研究,欲觀察不同Ferritin濃度與發生腸道病變和管狀腺瘤的存活曲線關係,嘗試利用邏輯斯回歸(Logistic Regression)尋找腸道病變與管狀腺瘤的危險因子並使用Receiver operating characteristic curve(ROC)建立Ferritin濃度適當的Cut-off值。
Several cohort studies described that there is a positive association betweencorrelation of the level of iron and ferritin with colon lesions which includes cancer risks while for transferrin level there is a negative associationcorrelation of transferrin level with these risks. Normally, long-term blood transfusion causes hemochromatosis among hemodialysis patients. However, few studies presented that the risks of colon lesions are associated with ferritin level of hemodialysis patients. Our study aim is to investigate the correlation between the incidence of colon lesions and ferritin level among hemodialysis patients. We conducted a retrospective cohort study to analyze the relationship between ferritin level and colon lesions. From September of 2004 to November of 2011 in a medical center in Northern Taiwan, there were 752 hemodialysis patients in the case group and patients with obvious cancer and colon lesions were excluded before hemodialysis(n=12). In fFinally 740 hemodialysis patients were included in our study and the median of followed-up time was 42.1 months. Outcome of colon lesions was defined by biopsy. The definition of ferritin level since the first hemodialysis was then divided into 4 groups by interquartile range while the latest lab data was considered as end time. We also investigated risk factors associated with colon lesion at logistic regression model, especially tubular adenoma and estimated the cut off value of ferritin for tubular adenoma as usingReceiver operating characteristic curve.