【文章】在英国患有肾衰竭并伴衰弱或多病的老年人中，为反应性管理或肾脏透析做准备（Prepare for Kidney Care）：一项双臂、优效性、平行组、开放标签、实用性、3期、个体随机对照试验
完整摘要 · 中文译文
背景：对于患有肾衰竭的老年人（无论衰弱和共病情况如何，年龄≥80岁；或年龄65-79岁但存在衰弱或多病共存），保守性肾脏管理与透析的比较有效性尚不清楚。我们旨在评估准备接受反应性管理（保守性肾脏管理的一种规范化方案）与准备接受肾脏透析的相对有效性。 方法：Prepare for Kidney Care 是一项双臂、优效性、平行组、开放标签、实用性、3期、个体随机对照试验，在英格兰、威尔士、苏格兰和北爱尔兰的31个肾脏中心开展。符合条件的患者估算肾小球滤过率低于15 mL/min/1.73 m2，且经同意前12个月内至少另一次非急性肾损伤所致的读数确认，并且年龄为80岁或以上，或年龄65-79岁且WHO体能状态评分至少为3或Davies合并症评分至少为2。符合条件的参与者按1:1随机分配至准备接受反应性管理——一种以患者为中心、侧重于减缓疾病进展、积极症状管理和维持生活质量而不使用透析或肾移植的方法——或准备接受肾脏透析。随机化按中心分层并按年龄最小化；集中化系统确保在随机化进行前分配隐藏。主要结局是质量调整生命年（QALYs）的平均差异，结合患者报告的EQ-5D-5L量表所得寿命和健康相关生活质量，包括从随机化至2025年8月31日的所有随访。主要结局在完整分析集中按意向治疗分析，完整分析集包括所有随机分配的参与者，存活参与者缺失的EQ-5D-5L值采用插补。危害（住院和透析通路并发症）和不良事件（死亡和住院，按原因） 在全分析集中进行评估。该试验已在ISRCTN注册，注册号为17133653（已完成）。 研究结果：2017年7月1日至2024年8月31日期间，在英国31个肾脏中心共筛查了6120例患者，其中445例（7%）符合条件并同意随机分组。445例参与者中，228例（51%）被随机分配至准备接受反应性管理组，217例（49%）被随机分配至准备接受肾脏透析组。中位年龄为82岁（IQR 80至86），445例参与者中女性137例（31%）。平均随访时间4.5年（SD 2.1）后，准备接受反应性管理组的平均QALYs为1.354（95% CI 1.201至1.508），准备接受肾脏透析组为1.472（1.288至1.655）（校正平均差-0.147 [95% CI -0.350至0.056]，p=0.16）。具体危害结局未预先设定，但两组住院率相似，且准备接受反应性管理组的透析通路并发症显著更低（校正发病率比0.41 [95% CI 0.19至0.88]），基于两组共36例参与者的74起事件。准备接受反应性管理组发生667起严重不良事件（228例参与者中157例[69%]死亡，共510次住院），准备接受肾脏透析组发生578起（217例参与者中136例[63%]死亡，共442次住院）。这些严重不良事件均未达到加速报告的标准。 解读：在患有肾衰竭且合并衰弱或多种疾病的老年人中，我们未发现证据表明，在平均4.5年的随访期内，准备接受反应性管理或准备接受肾脏透析在质量调整生命年（QALYs）方面孰优孰劣。本试验为临床决策提供了新的比较信息。 资助：美国国家卫生研究院卫生技术评估项目。
完整摘要 · 英文原文
BACKGROUND: The comparative effectiveness of conservative kidney management and dialysis in older people (aged ≥80 years regardless of frailty and comorbidity or aged 65-79 years if frailty or multimorbidity are present) living with kidney failure is unknown. We aimed to assess the relative effectiveness of preparing for responsive management-a protocolised version of conservative kidney management-and preparing for renal dialysis. METHODS: Prepare for Kidney Care was a two-arm, superiority, parallel-group, open-label, pragmatic, phase 3, individually randomised controlled trial conducted in 31 kidney centres in England, Wales, Scotland, and Northern Ireland. Eligible patients had an estimated glomerular filtration rate below 15 mL/min per 1·73 m2, confirmed by at least one additional reading in the 12 months preceding consent that was not due to acute kidney injury, and were either aged 80 years or older, or aged 65-79 years with a WHO performance status of at least 3 or a Davies comorbidity score of at least 2. Eligible participants were randomly assigned (1:1) to prepare for responsive management-a patient-centred approach focused on slowing disease progression, active symptom management, and preserving quality of life without the use of dialysis or kidney transplantation-or to prepare for renal dialysis. Randomisation was stratified by centre and minimised by age; a centralised system ensured allocation concealment until randomisation had taken place. The primary outcome was the mean difference in quality-adjusted life-years (QALYs), combining length and health-related quality of life from the patient-reported EQ-5D-5L scale, including all follow-up from randomisation to Aug 31, 2025. The primary outcome was analysed by intention to treat in the full analysis set, which included all randomly assigned participants, with missing EQ-5D-5L values for living participants imputed. Harms (hospital admissions and dialysis access complications) and adverse events (deaths and hospital admissions, by cause) were assessed in the full analysis set. The trial was registered with ISRCTN, 17133653 (completed). FINDINGS: From July 1, 2017, to Aug 31, 2024, 6120 patients were screened across 31 kidney centres in the UK, 445 (7%) of whom were eligible and agreed to be randomly assigned. 228 (51%) of 445 participants were randomly assigned to prepare for responsive management and 217 (49%) of 445 participants were randomly assigned to prepare for renal dialysis. Median age was 82 years (IQR 80 to 86) and 137 (31%) of 445 participants were female. After a mean follow-up duration of 4·5 years (SD 2·1), mean QALYs were 1·354 (95% CI 1·201 to 1·508) in the prepare for responsive management group and 1·472 (1·288 to 1·655) in the prepare for renal dialysis group (adjusted mean difference -0·147 [95% CI -0·350 to 0·056] p=0·16). Specific harms outcomes were not prespecified, but hospital admission rates were similar in the two groups and dialysis access complications were significantly lower in the prepare for responsive management group (adjusted incident rate ratio 0·41 [95% CI 0·19 to 0·88]), based on 74 events in 36 participants across both groups. There were 667 serious adverse events in the prepare for responsive management group (157 [69%] of 228 participants died, with 510 hospital admissions) and 578 in the prepare for renal dialysis group (136 [63%] of 217 participants died, with 442 hospital admissions). None of these serious adverse events met the criteria for expedited reporting. INTERPRETATION: Among older people living with kidney failure and frailty or multimorbidity, we found no evidence that either preparing for responsive management or preparing for renal dialysis was superior in terms of QALYs over a mean follow-up duration of 4·5 years. This trial provides novel comparative information for clinical decision making. FUNDING: National Institute for Health Research Health Technology Assessment programme.