運動降低末期腎臟病的風險- 五十萬人的世代研究

Abstract

背景與目標 自從1995年全民健康保險開辦以來，台灣的末期腎臟病，包括透析的發生率與盛行率都在快速增加，台灣的透析盛行率是世界第一位。過去針對一般民眾以運動來預防末期腎臟病或透析的研究並不多。本研究評估不同運動量與運動強度，對於預防末期腎臟病的效果。 方法 計畫一：本研究世代總共有543,667名參加健康檢查的參與者，他們在1996-2017年間，接受多次的健康檢查，追蹤時間的中位數為13年，在檢查時參與者會填寫一份關於其疾病史和生活方式的自填式問卷，其中包括詳細的運動資料，根據每週運動量，我們將參與者分為五組不同的運動量MET (metabolic equivalent of task)-hour。同時蒐集參與者的生化健檢資料，透過串聯全國重大傷病檔的檔案，找到2,520位接受長期透析或換腎的參與者。本研究以多變量COX比率風險模式來計算不同運動量與運動強度對於末期腎臟病的風險，同時也對其他共變項包括教育程度、臨床檢查和生活習慣等變項進行調整。 計畫二：本研究建立了三個不同的慢性透析預測模型來進行比較，模型1：腎衰竭風險模型 (Kidney failure risk equation, KFRE) (納入：年齡、性別、腎絲球過濾率和蛋白尿)；模型2：模型1 + 疾病史 + 生活方式等變項；模型3：完整模型+所有重要的臨床健康檢查變項。本研究也會依COX比率風險模式的結果來發展末期腎臟病的預測模式，以C-index來比較不同模式的預測能力。 結果 計畫一：有達到中度運動量以上，也就是每週約150分鐘以上運動量者，其末期腎臟病風險減少了12% (Hazard ratio：0.88, 95% confidence interval: 0.80, 0.98)，此風險已校正基線的腎絲球過濾率及蛋白尿在內的不同共變項。我們也觀察到不同運動量與末期腎臟病風險之間有劑量效應關係。與不運動的人相比，平均每增加30分鐘的運動，末期腎臟病風險減少5% (HR：0.95, 95% CI: 0.92, 0.98)，另外，在相同運動量中，與中等強度相比，高強度的運動降低末期腎臟病風險達35% (HR：0.65, 95% CI: 0.52, 0.81)。每週約150分鐘以上的運動量降低末期腎臟病風險的效果在男性、60歲以下以及糖尿病或高血脂症的參與者中更為明顯。 計畫二：在末期腎臟病預測模式中，除了年齡、性別、腎絲球過濾率及蛋白尿之外，再加入生活習慣及疾病史等因子可以顯著提高Harrell’s concordance index預測值(C 統計量)，從 0.91 提高到 0.94 (95%, CI: 0.94, 0.95)。在全模式的模式三中，更進一步將 C 統計量提高到 0.95 (95%, CI: 0.95, 0.96)。若參與者在全模式中的分數為33以上，在10 年會有達到3%的透析風險。而在本研究世代中，這個風險族群者有超過一半最後發展為需接受透析治療(敏感性為：0.53，95% CI: 0.51, 0.55)，這個敏感性比慢性腎臟病第三期以上還要更高(敏感性為：0.48，95% CI: 0.46, 0.50)。 結論 計畫一：每週約150分鐘以上的運動量能夠顯著降低末期腎臟病的。這個結果表示，那些有心血管疾病風險的人，包括慢性腎臟病者、糖尿病者、高血壓或高血脂者，都應該從事更多的運動來減少他們之後得到末期腎臟病的風險。 計畫二：我們所發展的末期腎臟病/透析預測模式，可以應用在一般沒有慢性腎臟病者或是有慢性腎臟病1~5期者來使用，本研究指出，納入生活習慣及疾病史等資料，可以顯著提高末期腎臟病的預測能力。

Background and Objectives The incidence and prevalence of end-stage renal disease (ESRD)/kidney dialysis have increased significantly in Taiwan since the initiation of the National Health Insurance in 1995. As a result, Taiwan has the world's highest dialysis prevalence per million population. Limited studies have been conducted on the effect of physical activity (PA) in preventing ESRD and dialysis in the Asian population. This prospective study examines the impact of amounts and the intensity of PA engagement per week on the risk of ESRD/dialysis in later life. Methods Project (1): The study cohort comprising 543,667 individuals participated in a standard health examination program in Taiwan. Participants received multiple medical screenings between 1996 and 2017, with a median of 13 years. At each visit, participants completed a self-administered questionnaire on their medical history and lifestyle risk factors, including detailed physical activity engagement, which was later converted into five amounts of MET (metabolic equivalent of task)-hour categories. Specimens for clinical tests were also collected. By linking ID data to the national registry for Catastrophic Illness Patients, we identified a total of 2,520 participants under long-term dialysis or kidney transplantation. A multivariate Cox proportional hazards model was used to assess the hazard ratio (HR) of different levels and intensity of physical activity engagement, adjusting for education, clinical, and lifestyle covariates. Project (2): We built three models for comparison: model 1: kidney failure risk equation (KFRE) model (age, sex, estimated glomerular filtration rate, and proteinuria); model 2: model 1 + medical history + lifestyle risk factors; and model 3: full model + all significant biochemical factors. We used the Cox proportional hazards model to develop points-based models and applied the C statistic to compare the predictive ability of different models. Results Project (1): The fully active group had a 12% lower hazard of ESRD compared with the no reported LTPA group (HR: 0.88, 95% CI: 0.80, 0.98) adjusting for covariates including baseline estimated glomerular filtration rate and proteinuria. We observed a dose-response relationship between LTPA and the risk of ESRD. For each additional 30 minutes of physical activity, the risk of end-stage renal disease was reduced by 5% (HR: 0.95, 95% CI: 0.92, 0.98). Within the exact amounts of LTPA, vigorous-intensity PA carried a 35% lower HR for ESRD than moderate-intensity (HR: 0.65, 95% CI: 0.52, 0.81). The effect was more substantial among men, younger participants, and participants with diabetes or hyperlipidemia. Project (2): Adding lifestyle factors to the basic model significantly improved the C statistic from 0.91 to 0.94 (95% CI: 0.94, 0.95). The full model improved the C statistic to 0.95 (95% CI: 0.95, 0.96). With a cut-off score of 33 of a 10-year ESRD risk (ESRD risk > 3%), this full model detected over half of the individuals progressing to ESRD (sensitivity: 0.53, 95% CI: 0.51, 0.55), which was higher than the sensitivity of cohort participants by the criteria with CKD stage 3 or higher (sensitivity: 0.48, 95% CI: 0.46, 0.50). Conclusions Project (1): More than 150 minutes of LTPA per week had a significant effect on lowering the ESRD risk. This finding suggested that no reported LTPA patients with cardio-vascular risks should engage more LTPA to lower their risk of ESRD. Project (2): Our ESRD prediction model, including medical history and lifestyle factors, significantly improved the predictive ability for long-term dialysis in a large cohort with or without chronic kidney diseases.