Effect of acute kidney injury care bundles on patient prognosis: a systematic review and meta-analysis

Abstract

Background

Acute kidney injury (AKI) is a severe complication among hospitalized patients. This study aimed to investigate the effect of care bundles on the prognosis of AKI patients.

Methods

Electronic databases were searched from January 2012 to December 2023. Randomized controlled trials and cohort studies evaluating the effect of AKI care bundles were included. A meta-analysis using a random-effects model was conducted to explore the efficacy of the AKI care bundle.

Results

A total of 12 studies with 30,152 participants were included. Based on the random-effects model, the AKI care bundles significantly improved the AKI severity (RR: 0.77, 95% CI: 0.60–0.98, I² = 64%) and the need for renal replacement therapy (RR: 0.66, 95% CI: 0.46–0.94, I² = 14%). However, our study did not find a statistically significant impact of AKI care bundle on the incidence of AKI incidence (RR: 0.95, 95% CI: 0.81–1.13, I² = 87%), major adverse kidney events (RR: 1.06, 95% CI: 0.65–1.73), in-hospital mortality (RR: 0.93, 95% CI: 0.81–1.07, I² = 19%), and length of hospital stay (MD: -0.16, 95% CI: -0.80, 0.47).

Conclusion

This systematic review indicates that the implementation of the AKI bundle is a promising care model for AKI patients. There is a need for more high-quality prospective studies on AKI and patients at high risk of AKI to further determine feasible and standardized models of AKI bundle care.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12882-025-04349-z.

Introduction

Acute kidney injury (AKI), a heterogeneous clinical syndrome and significant challenge in modern medicine, is prevalent among hospitalized patients and demonstrates substantial mortality during the acute phase [1]. Its progression to chronic kidney disease and end-stage renal disease significantly increases the risk, accompanied by substantial growth in medical costs. Epidemiological data indicate that approximately 13.3 million people experience AKI annually and 1.7 million die of AKI, with a mortality rate as high as 50% among AKI patients requiring renal replacement therapy [2, 3]. A retrospective multi-center study involving hospitalized patients from China revealed an AKI incidence rate of 11.6% among hospitalized patients, with a nosocomial AKI of 9.1% and an in-hospital AKI mortality rate of 8.8% [4]. Therefore, early identification of AKI and effective treatment and management are crucial strategies to prevent the progression of adverse outcomes.

Notwithstanding the current absence of disease-modifying therapies for AKI, the Kidney Disease Improving Global Outcomes (KDIGO) consortium established evidence-based clinical practice guidelines in 2012, prioritizing three core management pillars: hemodynamic optimization through volume resuscitation, mitigation of nephrotoxic exposures, and stringent glycemic control. These preventive and management measures are referred to as “care bundles” [5]. Care bundles are defined as a structured approach to improving care processes and patient outcomes, which consists of a set of small, direct evidence-based practices, treatments, and/or interventions tailored to specific patient populations or groups and care environments. When collectively applied, care bundles significantly enhances care reliability and patient outcomes beyond the expected effects of individual implementations [6]. It is worth noting that there are certain differences in the bundle of care for Acute Kidney Injury (AKI), and the basic components and specific objectives are not yet clearly defined. Data from clinical study indicated that care bundles represent one of the most promising care models for preventing disease progression in AKI patients, which can not only reduce the AKI incidence but also achieve better renal outcomes following AKI [7]. Therefore, this study is designed to investigate the positive effect of AKI care bundles on patients through a systematic review, thereby providing a reference basis for optimizing clinical practices in AKI care.

Methods

Retrieval strategy

Following the PRISMA 2020 statement [8], a systematic retrieval was conducted in five Chinese and English electronic databases, including Pubmed, Web of Science, Embase, CNKI, and Wanfang Database. Since the KDIGO practice guidelines introduced the concept of AKI care bundles in 2012, the literature retrieval was conducted from January 1, 2012, to December 31, 2023. The retrieval strategy for English databases included the following keywords: “acute kidney injury” OR “acute kidney failure” OR “acute renal failure”; “bundle” OR “bundle care” OR “patient care bundles” OR “care bundle.” The same Chinese keywords were used for literature retrieval in Chinese databases. Additionally, target articles were obtained by reviewing relevant literature and references from included studies and related reviews.

Inclusion and exclusion criteria

Inclusion Criteria: (1) Studies published in peer-reviewed Chinese and English journals; (2) Study participants aged ≥ 18 years, with no restrictions on race or gender, and having AKI or at risk of AKI; (3) The intervention group received AKI care bundle intervention, and there was no specific limitation on components and targets of AKI care bundle; (4) The control group did not receive AKI care bundle intervention; (5) Provided prognosis outcomes following AKI care bundle intervention, such as AKI incidence, renal outcomes, mortality, and length of stay; (6) Prospective studies, including randomized controlled trials and cohort studies.

Exclusion Criteria: (1) Patients with advanced kidney disease, previous dialysis or kidney transplant recipients; (2) Pregnant women; (3) Studies without data available for analysis.

Literature screening and data extraction

Literature screening was conducted by two investigators independently (H.Y, G. LL), with a third investigator involved to resolve any potential discrepancies from the screening results (S. C). Data extraction was performed using a standardized data extraction form prepared by the research team. The extracted data included: basic data (first author, publication date, region), study design (type, sample size, follow-up time, study endpoints), and study population (average age, percentage of males, disease characteristics).

Quality evaluation of studies

The quality of the included studies was assessed using the Cochrane Collaboration Risk of Bias Tool [9], which evaluates randomization methods, allocation concealment, blinding, completeness of outcome data, selective reporting of results, and other sources of bias. Meanwhile, the Newcastle-Ottawa Scale (NOS) [10] was utilized to evaluate the quality of cohort studies, assessing study population selection, comparability, and exposure/outcome assessment, with a maximum score of 9, and studies with a score of 7 or above were considered high quality. Two investigators independently (H.Y, G. LL) conducted the quality assessment, and a third investigator was involved to resolve any potential discrepancies from the screening results (S. C).

Statistical analysis methods

The primary endpoints of this study included incidence of AKI, kidney replacement therapy, and incidence of moderate to severe AKI. The secondary outcomes of this analysis were incidence of major adverse kidney event (MAKE, defined as combination of mortality, need for kidney replacement therapy, and persistent renal dysfunction), in-hospital mortality, length of ICU stay, and length of hospital stay. The effect size for count data was expressed using relative risk (RR), while mean difference (MD) was used for measurement data, with 95% confidence interval (CI) utilized to estimate the range of effect sizes. Heterogeneity was assessed using the Q-test and I² statistics to determine the presence/absence and size of heterogeneity. Meanwhile, a random-effects model was used for meta-analysis due to differences in the detailed intervention protocols among the included studies. In case of significant heterogeneity between the included studies, a sensitivity analysis was further conducted to explore the potential sources of heterogeneity. Reman 5.3 was used for data analysis. Unless otherwise specified, the significance level was set at 0.05.

Results

Basic characteristics and quality assessment

After conducting a comprehensive search of electronic databases, a total of 8,392 articles were included in the literature review process. Following the established inclusion and exclusion criteria, the titles and abstracts were screened, leading to 837 articles proceeding to the full-text review stage. Ultimately, 12 eligible studies [11–22] were included in the meta-analysis. The literature screening process is depicted in Fig. 1.

Fig. 1.

Flowchart of literature screening

The publication dates of the included studies range from 2015 to 2023. The study locations were mainly in the United Kingdom, Germany, Brazil, Canada, Australia, and the Netherlands, with RCTs and cohort studies being the main study types. The included studies involved a total of 30,152 study participants (mean age 59.9–79 years), of which 14,159 received AKI care bundle interventions, and 9 studies applied the KDIGO care bundle model. Three studies specified the study patients included from the ICU, and the remaining studies included patients from different departments, such as the emergency, or patients who underwent cardiac surgery. The basic characteristics of the included studies are detailed in Table 1.

Table 1.

Basic characteristics of eligible studies

Study	Location	Study design	Patients setting	Mean age	Male%	Diabetes%	eGFR	Type of surgery	Use of biomarker	Type of care bundle	Outcomes
Kolhe,2015	UK	prospective cohort	adults patients from a tertiary care center	76.6	50	NA	NA	NA	NA	KDIGO care bundle	30-day mortality, 60-day mortality, length of stay
Kolhe,2016	UK	prospective cohort	adults patients from any location of the hospital, most came from the Emergency	76.2	49	24.5	NA	NA	NA	AKI bundle care derived from NCEPOD	in-hospital mortality, need for renal replacement therapy, length of stay
Meersch,2017	Germany	RCT	adults at high risk for AKI who underwent cardiac surgery with the use of cardiopulmonary bypass	68.4	72.1	24.3	89.85	cardiac surgery	NA	KDIGO care bundle	occurrence of AKI, moderate to severe AKI, major adverse kidney events, 30-day mortality, 60-day mortality, 90-day mortality, need for renal replacement therapy, length of stay
Göcze,2018	Germany	RCT	adults ICU patients with a risk of AKI who underwent non-cardiac surgery	64	73.55	NA	NA	non-cardiac surgery	urinary TIMP-2 × IGFBP7	KDIGO care bundle	occurrence of AKI, moderate to severe AKI, major adverse kidney events, in-hospital mortality, need for renal replacement therapy, length of stay
Schanz,2019	Germany	RCT	adults patients from the emergency department who with high risk of AKI	64.4	45	18		NA	urinary TIMP-2 × IGFBP7	KDIGO care bundle	moderate to severe AKI, in-hospital mortality, 60-day mortality, need for renal replacement therapy
Hodgson,2018	Brazil	RCT	adults patients without AKI from two general acute medical units	74.5	NA	NA	NA	NA	NA	KDIGO care bundle	occurrence of AKI, in-hospital mortality, length of stay
Engelman,2020	Canada	respective cohort	high-risk adults patients of AKI who underwent cardiac surgery	66.3	72.37	38.49	73.6	cardiac surgery	urinary IGFBP7 and TIMP2	KDIGO care bundle	occurrence of AKI, moderate to severe AKI, 30-day mortality, length of stay
Koeze,2020	Netherlands	RCT	consecutive patients admitted to the ICU	59.9	62.23	14.72	NA	NA	NA	STK bundle	occurrence of AKI, moderate to severe AKI, in-hospital mortality, need for renal replacement therapy, length of stay
Bourdeaux,2020	UK	prospective cohort	adults patients from two ICU units	63.3	64.31	NA	NA	NA	NA	KDIGO care bundle	occurrence of AKI, moderate to severe AKI, in-hospital mortality
Zarbock,2021	Europe	RCT	adults patients who underwent cardiac surgery	66.45	70.14	28.78	80.7	cardiac surgery	urinary TIMP-2 × IGFBP7	KDIGO care bundle	occurrence of AKI, moderate to severe AKI, major adverse kidney events, 30-day mortality, 60-day mortality, 90-day mortality, need for renal replacement therapy, length of stay
Kotwal,2023	Australian	prospective cohort	adults AKI patients from three hospitals	74	54	31	NA	NA	NA	STOP-AKI care bundle	in-hospital mortality, length of stay
Iwers,2023	Germany	RCT	adults AKI patients	79	53	18.5	47.5	NA	NA	KDIGO care bundle	occurrence of AKI, major adverse kidney events, in-hospital mortality, need for renal replacement therapy, length of stay

Note AKI, acute kidney injury; KDIGO, Kidney Disease Improving Global Outcomes; NCEPOD, National Confidential Enquiry into Patient Outcome and Death; RCT, randomized controlled trials

The quality assessment results of the included studies indicated that eligible studies were of modest high quality (Supplementary Fig. 1 and Supplementary Table 1). There may be potential bias in applying allocation concealment and blinding in RCTs.

Incidence of AKI

Eight studies reported the impact of AKI bundle care on the incidence of AKI. A total of 12,595 individuals completed AKI bundle care, with 2,145 developing AKI, while 2,129 individuals in the control group developed AKI. Meta-analysis results based on the random-effects model indicated that the impact of AKI bundle care on the occurrence of AKI was not statistically significant (RR: 0.95, 95% CI: 0.81–1.13, I² = 87%), as shown in Fig. 2. Additionally, sensitivity analysis performed by sequentially excluding study data did not reveal any significant sources of heterogeneity, and the direction of the results did not change, suggesting that the heterogeneity among the included studies was relatively stable.

Fig. 2.

Effect of care bundles on incidence of AKI

Renal replacement therapy

Seven studies reported the application of renal replacement therapy after AKI care bundle interventions, with 64 out of 2,774 patients in the observation group receiving renal replacement therapy, compared to 128 patients in the control group. The results of the meta-analysis based on the random-effects model showed that AKI care bundles could effectively reduce the application of renal replacement therapy, with a pooled effect size of 0.66 (95% CI: 0.46–0.94, I² = 14%), as shown in Fig. 3.

Fig. 3.

Effect of AKI care bundles on renal replacement therapy

Incidence of moderate to severe AKI

Seven studies reported the effect of AKI care bundle interventions on the incidence of moderate to severe AKI. A total of 4,666 patients completed AKI care bundles, of which 556 developed moderate to severe AKI, and 649 patients in the control group developed moderate to severe AKI. The results of the meta-analysis based on the random-effects model showed that the implementation of AKI care bundles could reduce the risk of developing moderate to severe AKI, with a pooled effect size of 0.77 (95% CI: 0.60–0.98, I² = 64%), as shown in Fig. 4. Sensitivity analysis excluding one study [18] reduced the heterogeneity to 49%, with a pooled effect size of 0.68 (95% CI: 0.51–0.91).

Fig. 4.

Effect of care bundles on incidence of moderate to severe AKI

MAKE

Four studies reported the outcomes of MAKE following the implementation of AKI bundle care. Meta-analysis results based on the random-effects model indicated that the impact of AKI bundle care on MAKE was not statistically significant (RR: 1.06, 95% CI: 0.65–1.73, I² = 0%, Supplementary Fig. 2).

In-hospital mortality

Eight studies reported the in-hospital mortality after AKI care bundle interventions, with a total of 12,534 patients receiving AKI care bundles and 13,455 patients in the control group, of which 472 and 924 patients died, respectively. The meta-analysis results showed that AKI care bundles had no statistically significant effect on the risk of in-hospital mortality, with a pooled effect size of 0.93 (95% CI: 0.81–1.07, I² = 19%), as shown in Fig. 5. Furthermore, no statistically significant effect of AKI bundle care on mortality at different measurement time points was observed, with the following results: 30-day mortality (OR: 0.90, 95% CI: 0.74–1.09), 60-day mortality (RR: 0.89, 95% CI: 0.74–1.07), and 90-day mortality (RR: 1.17, 95% CI: 0.57–2.41). See Supplementary Fig. 3.

Fig. 5.

Effect of AKI care bundles on in-hospital mortality

Length of ICU stay and length of hospital stay

A meta-analysis based on the random-effects model was conducted to evaluate the impact of AKI bundle care on ICU and hospital length of stay. The results indicated that AKI bundle care did not significantly affect ICU stay (MD: 0.32, 95% CI: -0.33, 0.97) or hospital stay (MD: -0.16, 95% CI: -0.80, 0.47), as illustrated in Supplementary Fig. 4. Additionally, sensitivity analysis did not identify any clear sources of heterogeneity.

Discussion

This systematic review consolidates contemporary evidence evaluating the clinical efficacy of AKI care bundle implementation in improving patient-centered outcomes, employing rigorous methodology to appraise intervention protocols and prognostic trajectories. Additionally, our study provided an updated analysis, which suggests that the implementation of AKI care bundles can reduce the risk of moderate to severe AKI and decrease the application of renal replacement therapy, but the positive effects on mortality were not statistically significant. Our findings were consistent with previous studies that AKI care bundles can significantly reduce AKI severity and may have the potential to improve renal outcomes [23, 24], suggesting that the AKI care bundle represents a promising and effective care model for AKI patients or patients with a high risk of AKI. However, the incidence risk of AKI in our study was not consistent with that of Zhang and colleagues [24]. Notably, the positive effect of the AKI care bundle on AKI incidence was only found in a small sample size of other countries’ subgroup, but was not observed in the European countries subgroup.

The risk of moderate to severe AKI is lower than AKI incidence following the intervention of the AKI care bundle reported in our study. There are several potential explanations. The increased awareness and attention to AKI result in a higher detection rate of AKI, which may be a potential reason why the correlation between AKI care bundles and a reduced risk of AKI was not observed in the study [25]. At the same time, most of the included studies were conducted in ICU settings, the findings of which also indicated a decreased risk of moderate to severe AKI after the application of AKI care bundles [13, 14, 20]. Besides, Tsui et al. [26] conducted an educational study on AKI care bundles for primary care physicians in the UK, which effectively reduced the number of ICU admissions (P < 0.001) and renal replacement therapy in the ICU (1.8%-0%). The above evidence may suggest that the application of AKI care bundles in specific situations could exhibit a significant positive effect on high-risk patients. Additionally, no statistically significant improvement on mortality was found following intervention with AKI care bundle, indicating that the care bundle may improve renal endpoints without altering survival in populations with multi-organ failure. We have strengthened this discussion by proposing two mechanistic hypotheses: First, mortality in hospitalized AKI patients predominantly reflects extra-renal comorbidities rather than isolated kidney injury [27, 28]. Second, delayed bundle initiation and poor compliance may have attenuated survival benefits [29, 30]. Although our study found that care bundle had a tendency to reduce the overall AKI incidence, we did not find this difference to be statistically significant. This result, although somewhat different from previous findings, is noteworthy because our analysis included a larger sample size and the results of the sensitivity analysis showed that our findings were robust. Although our findings are inconsistent with two previous analyses [24, 31], the confidence intervals for the results of these two studies are very close to 1, and it is uncertain whether the results are fully robust.

The records of “Sepsis-6” and the application of sepsis care bundles provide evident reasons for applying care bundles. Results from a US multicenter observational study show that after introducing sepsis bundle care, patient mortality was halved (from 21.2 to 2.8%), and there was a clear rise in compliance (from 4.9 to 73.4%) [32]. In the case of AKI care bundles, compliance and completion rates are relatively low. A prospective cohort study conducted in eight Grade-III medical centers in Asia showed that compliance with AKI care bundles increased from 13 to 54% after 4 cycles of quality improvement and education, which was significantly associated with a reduction in mortality risk [33]. Despite previous studies suggesting that interruptive alerts could improve compliance with AKI care bundles [11, 34], no positive effect of this intervention was observed through a systematic review exploring the effect of interruptive alerts [35]. In contrast, the hefty management burden and the large number of checks may lead to “bundle or alter fatigue,” which could be a potential reason for this result [36].

However, this study also comes with the following limitations. Firstly, due to a lack of data in the included studies, further detailed analyses of complications, side effects, and mild AKI of care bundles were not performed. Additionally, from a study design perspective, these studies were mainly conducted on European populations, with patients primarily being those in emergency/critical care, and only one study included non-ICU patients [19], which may affect the generalizability of the study outcomes underscoring a more diverse population is needed in the future study. Lastly, there is a lack of standardized basic components for AKI care bundles, leading to high variability in the components and compliance of AKI care bundles across different studies, making it difficult to draw conclusions on the overall effectiveness of applying AKI care bundles.

In conclusion, the findings of this systematic review indicate that the application of AKI care bundles can lower the risk of moderate to severe AKI and reduce the need for renal replacement therapy. However, no significant effects were observed on in-hospital mortality rate. Given the limitations of this study, more high-quality prospective studies on AKI and high-risk AKI patients are needed to further establish feasible and standardized care bundle models.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Huang Y and Guan LL conceived of the study. Sun C participated in its design and data analysis and statistics. All authors helped to draft and revise the manuscript. All authors read and approved the final manuscript.

Funding

This study did not receive any funding in any form.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

An ethics statement and consent to participate are not applicable because this study is based exclusively on published literature.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.