利用週邊動脈阻塞的位置來預估血液透析病患的死亡

Abstract

背景背景知識: 腳踝手臂的血壓比值可以用來診斷週邊動脈阻塞疾病，和評估血液透析病患的死亡。然而，臨床上尚未有人提出週邊動脈的阻塞區域對透析病患未來死亡的影響。因此，這個研究是要來探討，週邊動脈阻塞的位置和血液透析病患的全因死亡與心血管性死亡之間的關聯。 研究設計: 回溯性，世代觀察研究。 受試者: 444位接受常規血液透析的台灣人，觀察期間為從2006/12至2013/06. 預測因子: 週邊動脈阻塞疾病主要的阻塞位置 感興趣的事件: 全因死亡，和心血管性死亡。 測量: 基本資料, 身體質量指標，心血管疾病病史，接受洗腎治療的時間，常規的血液透析觀察值，心臟胸廓比。 結果:受試者的平均年齡是61.6 ± 13.1歲，最長80個月的觀察期，平均觀察時間為51.5 ± 21.9個月. 如果病患的腳踝手臂的血壓比值小於0.9，則會被診斷為患有週邊動脈阻塞疾病。兩腳都有週邊動脈阻塞的比例為14.7 %，4.9％只有右腳，而只有左腳的是5.1%. 在整個觀察時間中，總共發生127件全因死亡案例，和93起心血管致死案例. 在利用Cox proportional hazards 回歸分析下，單因子分析顯示：週邊動脈阻塞的位置可以有效的預測血液透析病患的全因死亡與心血管性死亡。對於全因死亡的風險比值，右側的週邊動脈阻塞為3.03 (CI: 1.56 to 5.89)，左側為2.28 (CI: 1.14 to 4.58)，兩側皆有者為 4.10 (CI: 2.75 to 6.11) (參考值為沒有週邊動脈阻塞疾病). 對於心血管性死亡的風險比值，右側的週邊動脈阻塞為3.80 (CI: 1.86 to 7.74)，左側為1.62 (CI: 1.11 to 5.44)，兩側皆有者為3.94 (CI: 2.44 to 6.35) (參考值為沒有週邊動脈阻塞疾病). 經過多因子校正後， 右側的週邊動脈阻塞仍然可以有效預測全因和心血管性死亡。然而，左側的週邊動脈阻塞失去了它對死亡的預測能力。對於全因死亡的風險比值，右側的週邊動脈阻塞為3.26 (CI: 1.63 to 6.54)，左側為1.15 (CI: 0.54 to 2.44) 兩側皆有者為2.13 (CI: 1.24 to 3.68) (參考值為沒有週邊動脈阻塞疾病). 心血管性死亡的風險比值，右側的週邊動脈阻塞為4.02 (CI: 1.90 to 8.49)，左側為1.03 (CI: 0.43 to 2.44) 兩側皆有者為1.55 (CI: 0.80 to 3.42) (參考值為沒有週邊動脈阻塞疾病). 總結: 週邊動脈阻塞的位置可以有效預測血液透析病患的全因和心血管性死亡. 此外，右側週邊動脈的阻塞是血液透析病患死亡的獨立危險因子。

Background: The ankle-brachial blood pressure (BP) index (ABI) can indicate the presence of peripheral artery occlusive disease (PAOD) and predict mortality in patients on hemodialysis (HD). However, there is a lack of information on whether the predictive power for mortality of PAOD varies according to the dominant side of PAOD. This study was therefore designed to elucidate the associations between different locations of PAOD and all-cause and cardiovascular mortality in HD patients. Study Design: Retrospective observational cohort study. Setting & Participants: A cohort of 444 Taiwanese patients undergoing chronic HD was studied between December 2006 and June 2013. Predictor: The primary predictors were the occlusive locations of PAOD. Outcomes: All-cause mortality and fatal cardiovascular events. Measurements: Demographic data, body mass index, a history of cardiovascular diseases, HD vintage, routine laboratory data, and cardiothoracic ratio. Results: The mean age at entry was 61.6 ± 13.1 years and the longest follow-up time was 80 months, with a mean follow-up duration of 51.5 ± 21.9 months. PAOD was defined as an ABI below 0.9 as determined by using the oscillometric method. The frequency of PAOD was 14.7% in both legs, 4.9% in the right leg only, and 5.1% in the left leg only. During the study period, 127 all-cause mortality events and 93 fatal cardiovascular events occurred. On the basis of a Cox proportional hazards regression analysis, the location of PAOD was found to have significant predictive power for all-cause and cardiovascular mortality in the crude analysis. The hazard ration (HR) for all-cause mortality was 3.03 (confidence interval [CI]: 1.56 to 5.89) in those with PAOD on the right side, 2.28 (CI: 1.14 to 4.58) in those with PAOD on the left side, and 4.10 (CI: 2.75 to 6.11) in those with PAOD on both sides (all variables are compared to normal). The HR for cardiovascular mortality was 3.80 (CI: 1.86 to 7.74) in those with PAOD on the right side, 1.62 (CI: 1.11 to 5.44) in those with PAOD on the left side, and 3.94 (CI: 2.44 to 6.35) in those with PAOD on both sides (all variables are compared to normal). After adjustment for confounding variables, the right side PAOD continued to have significant predictive power for all-cause mortality and cardiovascular mortality. However, the left side PAOD appeared to lose its predictive power. The HR for all-cause mortality was 3.26 (CI: 1.63 to 6.54) in those with PAOD on the right side, 1.15 (CI: 0.54 to 2.44) in those with PAOD on the left side, and 2.13 (CI: 1.24 to 3.68) in those with PAOD on both sides (all variables are compared to normal). The HR for cardiovascular mortality was 4.02 (CI: 1.90 to 8.49) in those with PAOD on the right side, 1.03 (CI: 0.43 to 2.44) in those with PAOD on the left side, and 1.55 (CI: 0.80 to 3.42) in those with PAOD on both sides (all variables are compared to normal). Conclusions: PAOD location has significant predictive power for all-cause and cardiovascular mortality in HD patients. In addition, right side PAOD is an independent risk factor for the mortality of HD patients.