比較 Sacubitril/Valsartan 與單一血管張力素受體阻斷劑 (ARB) 於腹膜透析合併心衰竭患者之真實世界有效性與安全性：一項回溯性世代研究

研究目的：Sacubitril/Valsartan 作為血管張力素受體－腦啡肽酶抑制劑（Angiotensin Receptor-Neprilysin Inhibitor, ARNI），已廣泛應用於心臟衰竭之標準治療，然接受腹膜透析（PD）且合併心衰竭之患者，因其高風險與複雜之生理特徵，常被排除於大型隨機對照試驗之外。本研究旨在探討並比較 ARNI 與單一血管張力素受體阻斷劑（Angiotensin Receptor Blocker, ARB）在此特定族群中之長期臨床結局與安全性。研究方法及資料：本研究採用回溯性世代分析設計，資料擷取自 TriNetX US Collaborative Network 真實世界資料庫，收案期間為 2015 年 1 月 1 日至 2025 年 12 月 31 日。透過傾向分數配對平衡兩組基線特徵後，ARNI 組與 ARB 組各納入 714 名患者。針對各項終點進行時間至事件（time-to-event）分析，運用 Kaplan-Meier 存活分析與 Cox 比例風險模型計算風險比（Hazard Ratio, HR），並以 E-value 評估未量測干擾之穩健性。研究結果：經配對後，ARNI 與 ARB 兩組在腹膜透析技術失敗（PD failure, HR 0.90）、三重複合心血管不良事件（3-point Major Adverse Cardiovascular Events, MACE, HR 1.00）、全因死亡（HR 0.92）、嚴重高血鉀及低血壓等風險上，均未達統計顯著差異。唯一呈現顯著差異之結局為急性心衰竭事件（Acute Heart Failure, AHF）風險，ARNI 組顯著高於 ARB 組（HR 1.37, 95% CI 1.20–1.57, E-value 2.09）。此外，探索性分析顯示，可得之基線左心室射出分率（Left Ventricular Ejection Fraction, LVEF）在 ARNI 組顯著較低（33.9% vs. 39.8%, p = 0.0166）。結論與建議：本研究結果未支持 ARNI 在腹膜透析合併心衰竭患者中具備較佳之長期預後，惟兩組於嚴重高血鉀、低血壓等安全性指標上亦未見顯著差異，顯示 ARNI 之整體安全性與單一 ARB 相近。ARNI 組較高之 AHF 風險，極可能是受到真實世界中處方習慣與適應症混淆（confounding by indication）影響所致。本研究結論應以假說生成（hypothesis-generating）之方式進行學術詮釋，未來仍需進一步之研究設計加以驗證。

Objective: Sacubitril/Valsartan, an angiotensin receptor-neprilysin inhibitor (ARNI), is widely utilized as a standard therapy for heart failure; however, patients undergoing peritoneal dialysis (PD) with concurrent heart failure are frequently excluded from large-scale randomized controlled trials (RCTs). This study aimed to compare the long-term clinical outcomes and safety profiles of ARNI versus single-agent angiotensin receptor blockers (ARBs) in this specific population.Methods and Materials: A retrospective cohort study was conducted using the TriNetX US Collaborative Network database, including patients from January 1, 2015, to December 31, 2025. Following propensity score matching (PSM), 714 patients were included in both the ARNI and ARB groups. Time-to-event analyses were performed using Kaplan-Meier curves and Cox proportional-hazards models to estimate hazard ratios (HRs), and E-values were calculated to assess the robustness against unmeasured confounding.Results: After matching, there were no statistically significant differences between the two groups regarding the risks of PD failure (HR 0.90), 3-point major adverse cardiovascular events (MACE, HR 1.00), all-cause mortality (HR 0.92), severe hyperkalemia, and hypotension. The only outcome with a significant difference was the risk of acute heart failure events (AHF), which was significantly higher in the ARNI group (HR 1.37, 95% CI 1.20–1.57, E-value 2.09). Furthermore, exploratory analysis revealed that the available baseline left ventricular ejection fraction (LVEF) was significantly lower in the ARNI group (33.9% vs. 39.8%, p = 0.0166).Conclusion and Suggestions: The findings of this study do not support a better long-term prognosis for ARNI in PD patients with heart failure. However, no significant differences were observed in safety outcomes such as severe hyperkalemia and hypotension, indicating that the overall safety profile of ARNI is comparable to that of single-agent ARBs. The higher AHF risk observed in the ARNI group is likely influenced by confounding by indication in real-world clinical practice. Consequently, these results should be strictly interpreted as hypothesis-generating, warranting further prospective investigations.