顯影劑暴露後發生急性及慢性透析之風險：系統性文獻回顧與統合分析和唯病例法研究

Risk of acute-temporary and chronic dialysis after radio-contrast medium exposure: a systematic review with meta-analysis and case-only studies

吳美儀(Mei-Yu Wu)

指導教授 ： 杜裕康

國立臺灣大學/公共衛生學院/流行病學與預防醫學研究所/博士(2022年)

https://doi.org/10.6342/NTU202200454

摘要

背景
末期腎臟病透析被認為是一項重大的公共衛生議題，顯影劑暴露後透析的風險尚不明確，本研究目標是透過統合分析的系統性回顧來評估顯影劑腎病變發生及透析的發生風險。具有慢性腎臟病及充血性心衰竭可能是顯影劑導致急性腎損傷的危險因子，本研究希望釐清高風險族群具慢性腎臟病及鬱血性心衰竭患者，在顯影劑暴露或冠狀動脈心導管介入後導致透析的風險。
方法
我們在PubMed、Medline、Embase 和 Cochrane Library電子數據庫檢索關鍵字，以確立相關研究，本研究調查了包括血管攝影後顯影劑腎病變及導致透析治療的發生率。經由完整的文獻回顧及數據萃取，接著使用隨機效應模型計算出顯影劑腎病變及血管造影後透析的總發生率，並使用次群組分析和統合迴歸分析以評估研究間發生率的異質性。同時，我們使用健康保險資料庫中的病例交叉研究設計來計算顯影劑後導致急慢性透析的風險，唯病例法研究設計的優點為不隨時間改變的干擾因子會自我校正，因此我們估計的相對風險可以避免干擾因子造成偏差，研究對象為慢性腎臟病及鬱血性心衰竭患者，以患者本身在6個月前的狀態作為對照，我們估計的暴露區間長度分別為ㄧ週和四週，使用條件式羅輯斯回歸模型來估計顯影劑暴露後短期的急性透析及慢性透析的風險。
結果
在2,243篇文章中，123 篇符合我們的納入標準，在篩選後被納入統合分析，血管攝影後顯影劑腎病變及需要透析的發生比率估計值分別為9.06%（95% 信賴區間：8.53%-9.58%；來自 120 項研究）和0.52%（95% 信賴區間：0.37%-0.70%；來自110 項研究）。動脈內注射顯影劑後顯影劑腎病變及需要透析的發生比率分別為9.60%（95% 信賴區間：9.0%-10.2%；來自106項研究）和0.6%（95% 信賴區間：0.40%-0.80%；分別來自 100 項研究），使用統合迴歸分析發現顯影劑暴露與顯影劑腎病變有關。
在病例交叉設計研究中，共有36,709名慢性腎臟病及鬱血性心衰竭患者在顯影劑暴露後接受了急慢性透析治療。進行放射顯影劑暴露後一週透析的勝算比為 4.49（95% 信賴區間：3.99-5.05）。急性(N=23,418) 和慢性透析 (N=13,291) 的勝算比在分別為 5.57 (95% 信賴區: 4.83-6.42) 和 2.37 (95% 信賴區間: 1.90-2.95)。晚期慢性腎臟病患者（N=12,030）接受放射顯影劑後需要透析的勝算比為 3.25（95% 信賴區: 2.53-4.19）；早期慢性腎臟病患者（N=24,679）接受放射顯影劑後需要透析的勝算比則為 4.85（95% 信賴區間: 4.24-5.54）。慢性腎臟病及充血性心衰竭患者冠狀動脈心導管介入術後需要透析的勝算比為 3.75 (95% 信賴區間: 2.57-5.48)。
結論
可藉此研究結果與接受需要顯影劑的診斷或冠狀動脈心導管介入術的患者說明顯影劑腎病變發生風險及可能導致透析的潛在風險。當充分考慮研究偏差後，患者進行急性或慢性透析的臨床風險顯著高，因此，我們需要針對高風險族群慢性腎臟病及充血性心衰竭患者在進行放射顯影劑暴露後避免透析風險的策略。

關鍵字

顯影劑腎病變 ； 統合分析 ； 病例交叉設計 ； 自我對照研究設計 ； 透析風險

並列摘要

Background
Dialysis for end stage kidney disease is considered a major public health challenge. The risk of dialysis following contrast exposure is unclear. We aimed to examine the overall risk of contrast induced nephropathy and the need of dialysis based on a systematic review with random-effects meta-analysis. Pre-existing chronic kidney disease (CKD) and congestive heart failure (CHF) may be independent risk factors for contrast-induced acute kidney injury. We aimed to investigate dialysis risk in patients with CKD and CHF after radio-contrast medium exposure or coronary catheterization.
Methods
We searched the electronic database including PubMed, Medline, Embase, and Cochrane Library from inception to 31 Oct, 2020 with predetermined search term to identify relevant studies. Observational studies investigating the association between contrast induced nephropathy after angiography and the need of dialysis were included, and summary risks were estimated. Two independent reviewers extracted the data, followed with random effects model to calculate the overall pooled incidence of contrast induced nephropathy and the need of dialysis after angiography. Subgroup-analysis and meta-regression were performed to assess heterogeneity of incidence across studies.
To investigate the dialysis risk in patients with CKD and CHF, we used case-crossover design in the Health Insurance Database to identify incident dialysis patients with CKD and CHF. Patients themselves in 6-months ago serve as their own controls. The advantage of this method is that confounding factors which do not vary with time are adjusted for implicitly. This prevents selection bias in the control group, such as healthy volunteer bias and confounding bias. Conditional logistic regression model was used to estimate the risk of dialysis shortly after radio-contrast medium exposure.
Results
Of 2,243 identified articles, 123 met our inclusion criteria were included in the meta-analysis after screening. Pooled effect estimates had the following summary incidence proportion for contrast induced nephropathy after angiography: 9.06% (95% Confidence Interval (CI): 8.53%-9.58%; derived from 120 studies) and 0.52% (95% CI: 0.37%-0.70%; derived from 110 studies) for the need of dialysis, respectively. The stratified summary incidence proportion of contrast induced nephropathy after contrast administration via intra-arterial route: 9.60% (95% CI: 9.0%-10.2%; derived from 106 studies) and 0.6% (95% CI: 0.40%-0.80%; derived from 100 studies) for the need of dialysis, respectively. Our meta-regressions found that the contrast medium exposure was associated with contrast-induced nephropathy.
In the cross-over design study, totally 36,709 patients with CKD and CHF underwent dialysis after radio-contrast medium exposure. At 1 week, the odds ratio (OR) for dialysis was 4.49 (95% CI: 3.99-5.05). The ORs for acute-temporary (N=23,418) and chronic dialysis (N=13,291) were 5.57 (95% CI: 4.83-6.42) and 2.37 (95% CI: 1.90-2.95) after radio-contrast medium exposure, respectively. The ORs for dialysis after radio-contrast medium exposure in advanced CKD patients (N=12,030) were 3.25 (95% CI: 2.53-4.19) and 4.85 (95% CI: 4.24-5.54) in early CKD patients (N=24,679). The ORs for dialysis after coronary catheterization in patients with CKD and CHF was 3.75 (95% CI: 2.57-5.48).
Conclusion
The potential risk of dialysis needs to be communicated to patients undergoing diagnostic and /or interventional procedures requiring contrast medium. The clinical risk for acute-temporary or chronic dialysis was significantly high when the bias was fully considered. We need strategies to reduce the subsequent risk of dialysis after radio-contrast medium exposure, especially in patients with CKD and CHF.

並列關鍵字

contrast induced nephropathy ； meta-analysis ； case-crossover design ； self-controlled design ； risk of dialysis

參考文獻

1. Couser, W.G., et al., The contribution of chronic kidney disease to the global burden of major noncommunicable diseases. Kidney Int, 2011. 80(12): p. 1258-70.

2. Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet, 2016. 388(10053): p. 1459-1544.

3. Mehta, R.L., et al., International Society of Nephrology's 0by25 initiative for acute kidney injury (zero preventable deaths by 2025): a human rights case for nephrology. Lancet, 2015. 385(9987): p. 2616-43.

4. Hoste, E.A.J., et al., Global epidemiology and outcomes of acute kidney injury. Nat Rev Nephrol, 2018. 14(10): p. 607-625.

5. Kellum, J.A., et al., Recovery after Acute Kidney Injury. Am J Respir Crit Care Med, 2017. 195(6): p. 784-791.