[Lancet发表述评]：采取集束化措施预防急性肾损伤 | 中国病理生理学会危重病医学专业委员会

上篇下篇

[Lancet发表述评]：采取集束化措施预防急性肾损伤

2025年12月29日 ⁄ 研究点评, 进展交流 ⁄ 已关闭评论

Comment

Acute kidney injury (AKI) is among the most common organ failures observed in the hospital and can irrevocably alter the trajectory of a patient's short-term and long-term health.1, 2, 3, 4 The lack of pharmacological treatments for AKI has generated broad interest in applying supportive care strategies that reduce the condition's development and complications.5, 6 Multi-faceted, or so-called bundled-care interventions, have shown promise for reducing the incidence of AKI in single-centre studies, particularly in the perioperative setting, where preventive treatments are more easily timed than with AKI that occurs during acute illness .7, 8

In The Lancet Alexander Zarbock and colleagues⁹ report the results of a randomised clinical trial (BigpAK-2) done at 34 hospitals across Europe involving 1180 patients (396 [33·6%] women and 784 [66·4%] men; median age in the intervention group 72·0 [IQR 63·0–78·0] years and 71·0 [63·0–77·0] years in the control group; ethnicity was unreported) undergoing major surgery at risk for AKI. At risk was defined by the presence of an established risk factor for AKI (age ≥75 years, vasopressors or mechanical ventilation use, pre-existing stage 3 chronic kidney disease, and intraoperative use of radiocontrast) and increased urine TIMP-2 × IGFBP7, biomarkers associated with developing stage 2–3 AKI.¹⁰ The intervention included good practice concepts adapted from the 2012 Kidney Disease Improving Global Outcomes AKI guidelines.¹¹ These elements focused on (1) optimising haemodynamics by serially assessing fluid responsiveness along with the use of pressors and ionotropes to target a mean arterial pressure of at least 65 mm Hg and cardiac index of at least 2·5 mL/min per m²; (2) minimising nephrotoxin exposures; and (3) strict glycaemic control (100–150 mg/dL). The primary endpoint was moderate or severe AKI, defined as oliguria (<0·5 mg/kg per h for ≥12 h), a doubling or increase in serum creatinine to at least 4 mg/dL, or renal replacement therapy (RRT) within 72 h of surgery.¹¹ Secondary outcomes included AKI lasting at least 48 h, major adverse kidney events until day 90 (MAKE₉₀; a composite of death, RRT, or persistent doubling of serum creatinine),¹² and changes in urine TIMP-2 × IGFBP7 concentrations.

Complete adoption of the bundle occurred in 268 (46·9%) participants in the intervention group and 353 (62·7%) participants when excluding strict glycaemic control. Despite this finding, the authors observed a reduction in the primary endpoint occurring in 84 (14.4%) patients in the intervention group compared with 131 patients (22·3%) in the control group (odds ratio 0·57 [95% CI 0·40–0·79]; p=0·0002). Differences in the primary endpoint were most often triggered by the urine output criteria; although, reassuringly, significant reductions in AKI using serum creatinine criteria were also observed (6·3% vs10·0%; p=0·016), along with non-significant reductions in persistent AKI. No difference in MAKE₉₀ was observed, nor in TIMP-2 × IGFBP7 concentrations.¹⁰ Analyses remained robust to sensitivity analyses and across subgroups, with no differences in prespecified adverse events. Exploratory analyses suggested that avoidance of hypotension and discontinuation of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers were most strongly associated with the primary outcome.

The results from this study are thought-provoking, and the investigators should be commended. The size, broad surgical case-mix, multicentre design, and enrichment with patients at high risk were major undertakings that add robustness and generalisability to the findings. The attention to developing and implementing a specific set of care elements to reflect more general optimal supportive care recommendations was also meticulous and laudable. Limitations included moderate protocol adoption and the exclusion of stage 4 chronic kidney disease, a population at increased risk for AKI and its complications but potentially less likely to respond to the intervention than those with earlier stages of chronic kidney disease. The data in support of some of the elements, such as tight glycaemic control targets, have also not been confirmed in other trials, and this bundle element was not implemented in many participants randomly assigned to the intervention.

The findings from BigpAK-2 also raise important questions for potential adopters. Despite more frequent use of passive leg raises and cardiac index assessments in the intervention group, other differences in treatments were moderate. A small but non-significant difference in fluid administration was observed, with more fluid administered in the intervention group, along with a small increase in patients receiving inotropic therapy. There was also no difference in the proportion of patients receiving vasopressors, nor in the vasopressor dose. Thus, although avoiding sustained hypotension and ensuring adequate volume resuscitation were probably instrumental in reducing AKI risk, how and which components of the bundle triggered the individualised responses to achieve these goals are less clear. BigpAK-2 also only enrolled patients who had a urinary catheter and central line after surgery and required technologies to estimate cardiac index whose application and availability in the postoperative setting varies widely by region and case mix. Lastly, an additional biomarker test was applied to identify patients for enrolment. As the intervention did not appear to increase harm, adopters will need to weigh the benefit of measurement against the added cost. Although it is possible to implement the intervention in all patients who meet inclusion criteria without biomarker measurement, the lower biomarker population (about half of those who met all other inclusion criteria) probably has a different risk profile and lower event rate. Assessing whether the intervention can benefit patients at high risk identified using other readily available clinical enrichment strategies might facilitate more widespread adoption.

What does this mean for clinical care? AKI remains a challenging and resource-intensive condition without novel therapies. The results from BigpAK-2 reinforce that excellent supportive care, when applied to a high-risk surgical population, can reduce the risk for AKI and that improving the process of care strategies to prevent this common condition cannot be overlooked in an area in need of effective treatments. Yet, maximising the intervention's effect will depend on ensuring that supportive strategies are both practical and efficient. In settings where components of the intervention are not easily adapted today (eg, lack of biomarkers or no plan for cardiac output monitoring or central line placement), working to identify the most effective elements, the most feasible ways to implement them, and additional pragmatic approaches to identify which patients can benefit will expand the ability to improve the outcomes long overdue for this population.