Association between invasive mechanical ventilation and acute kidney injury in critically ill patients: a protocol for a systematic review and meta-analysis of randomised controlled trials

Abstract

Introduction

Invasive mechanical ventilation (IMV) may contribute to the development and progression of acute kidney injury (AKI) in critically ill patients. Given the expanding evidence-base and the immense variation in the association described, particularly following the COVID-19 pandemic era, a rigorous synthesis is warranted. This systematic review and meta-analysis will evaluate this association and identify potential risk-mitigation strategies.

Methods and analysis

We will conduct a comprehensive search of databases, including Ovid MEDLINE, Ovid Embase, CINAHL, Scopus, Web of Science Core Collection and the Cochrane Library, using keywords to capture relevant literature on IMV and AKI. The search will include studies published from database inception through March 2026. Two reviewers will independently screen, select and extract studies using the web-based platform Covidence. Eligible studies will include randomised controlled trials reporting the occurrence of new-onset or worsening AKI following initiation of IMV in adult patients admitted to intensive care units. Studies in which IMV was initiated after the onset of AKI will be excluded. Study quality and risk of bias will be assessed using the Cochrane Risk-of-Bias 2 Tool. The primary outcome will be the proportion of critically ill patients who develop new-onset or worsening AKI after initiation of IMV. Secondary outcomes will include AKI trajectory, use of organ support and length of stay.

Ethics and dissemination

Ethics approval is not required, as this study will synthesise data from previously published studies and will not involve direct contact with human participants. The findings will be disseminated through presentations at academic conferences and publication in a peer-reviewed journal.

PROSPERO registration number

CRD420261303862.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The protocol was developed and will be reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols guidelines.

• Study screening, selection, data extraction and risk-of-bias assessment will be conducted independently by two reviewers.

• The literature search will be conducted using multiple electronic databases with a prespecified search strategy and predefined eligibility criteria.

• Prespecified subgroup and sensitivity analyses will be performed where sufficient data are available, and the certainty of evidence will be assessed using the Grading of Recommendations, Assessment, Development and Evaluation approach.

• Despite restricting inclusion to randomised controlled trials, substantial heterogeneity may exist in acute kidney injury definitions, invasive mechanical ventilation strategies, patient populations and outcome reporting across studies.

Introduction

The use of invasive mechanical ventilation (IMV) among patients with acute respiratory failure (ARF), including acute respiratory distress syndrome (ARDS), influences lung-kidney crosstalk in a bidirectional manner. Specific ventilator settings, particularly tidal volume (TV) and positive end-expiratory pressure (PEEP), can induce early haemodynamic alterations and neurohormonal activation that may adversely affect kidney function. In addition, injurious IMV can exacerbate these effects and promote biotrauma, triggering a systemic inflammatory response that may further compromise kidney function. Consequently, ventilator-associated acute kidney injury (AKI) may contribute substantially to morbidity, mortality and adverse clinical outcomes.¹

Over the past decade, the incidence of AKI among patients receiving IMV for ARF/ARDS has ranged from 25% to 60%, with AKI frequently developing within the first 1–3 days following initiation of IMV.^{2 3} A prior meta-analysis demonstrated that IMV is independently associated with a threefold increase in the odds of developing AKI, with this association remaining consistent across varying ventilator settings, suggesting that the intrinsic risk is related to IMV itself rather than to any specific ventilatory strategy.⁴ Nevertheless, existing evidence remains constrained by heterogeneity in study design, variability in AKI definitions and uncertainty regarding the temporal relationship between IMV initiation and AKI onset.⁴ In contrast, during and following the COVID-19 pandemic era, widespread use of IMV was accompanied by a high incidence of AKI treated with renal replacement therapy (RRT), potentially related to higher PEEP and conservative fluid management strategies, which may have further increased AKI risk.^{5 6} These findings introduce inconsistencies compared with data from the pre-COVID-19 era.

Patients who develop AKI experience higher mortality, longer durations of IMV and prolonged intensive care unit (ICU) stays compared with those who do not develop AKI.³ The coexistence of IMV and AKI markedly worsens outcomes, with mortality increasing fourfold to sixfold compared with either condition alone.⁷ Evidence suggests that lung-protective ventilation (LPV) strategies using low TV in the management of ARF/ARDS may mitigate ventilator-associated AKI.⁸ Moreover, emerging evidence indicates that extracorporeal organ support, particularly RRT, represents a promising adjunctive strategy to facilitate LPV and improve outcomes in patients with ventilator-associated AKI.^{9 10}

With the rapid expansion of contemporary evidence, an updated and rigorous synthesis is required. Accordingly, we aim to perform a systematic review and meta-analysis to evaluate the occurrence of AKI following the initiation of IMV in critically ill patients, synthesise current knowledge regarding its clinical impact and identify potential strategies for risk mitigation.

Methods and analysis

Study design

We will perform a systematic review and meta-analysis to identify and evaluate the occurrence of AKI following the initiation of IMV among patients with critical illness, following guidelines from the Cochrane Collaboration and the Centre for Reviews and Dissemination. The reporting of this systematic review and meta-analysis protocol was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guidelines (online supplemental table 1).¹¹

Study registration

This systematic review and meta-analysis was registered in the International Prospective Register of Systematic Reviews (PROSPERO) on 19 March 2026 (CRD420261303862).

Search strategy

The initial search was conducted through March 2026, and data analysis will begin in November 2026. A research librarian (JYK) developed and will execute comprehensive searches in Ovid MEDLINE, Ovid Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Scopus, Web of Science Core Collection and Cochrane Library (via Wiley). To capture all relevant literature pertaining to the association between IMV and AKI in critically ill patients, relevant keywords and controlled vocabulary will be carefully selected. The search strategy will incorporate the following core concepts: (1) acute kidney injury, acute renal failure; (2) mechanical ventilation, ventilation strategy, TV, PEEP, mechanical power; and (3) critical care, critical illness, critically ill, intensive care and ICU (online supplemental table 2). The search will include studies from database inception. There will be no language limits applied. All identified records will be imported into Covidence (www.covidence.org) for title and abstract screening, full-text review and data extraction.

Trial eligibility for screening and selection

Two reviewers will independently screen study titles and abstracts, followed by full-text review to determine eligibility based on predefined inclusion and exclusion criteria using Covidence. Disagreements will be resolved through discussion; if the consensus cannot be reached, a third reviewer will adjudicate.

Inclusion criteria

Eligible studies will meet all the following criteria: (1) randomised controlled trials (RCTs) to provide a more structured and controlled study design; (2) adult patients (≥18 years) admitted to an ICU; and (3) studies reporting the occurrence of new-onset or worsening AKI after the initiation of IMV, using validated AKI definitions.

Exclusion criteria

We will exclude publications that are described as: (1) case-reports or case-series; (2) preclinical or experimental studies; (3) study protocols; and (4) studies reporting the use of IMV primarily after the onset of AKI.

Data extraction

Two reviewers will independently extract data using Covidence, with discrepancies resolved through discussion or consultation with a third reviewer as needed. The variables to be collected are summarised in table 1. IMV initiation will be considered the primary exposure of interest. Other factors, including ventilator strategies or intensity (TV, PEEP, plateau pressure and driving pressure), adjunctive respiratory interventions (neuromuscular blockade, prone positioning and inhaled nitric oxide or other pulmonary vasodilators) and extracorporeal organ support strategies, will be considered secondary exposures and evaluated separately where sufficient data are available.

Table 1. Summary of data to be extracted from eligible articles for systematic review.

Data extraction domain	Specific variables
Article publication	Journal, publication year, DOI.
Study design	Sites (single-centre, multicentre).
Population	Eligibility. Socio-demographic. Baseline characteristics. Type of ICU admission (sepsis, ARDS, post-operation). COVID-19 status. Measures of illness acuity.
Susceptibilities/exposures	AKI-related susceptibilities. AKI-related exposures (eg, hypotension, nephrotoxins, sepsis, surgery). ARF-related exposure (eg, type, cause). IMV-related exposures (eg, mode, TV, PEEP, plateau pressure, driving pressure, FiO2).
Primary outcomes	AKI (yes/no by any definition).
Secondary outcomes	Severity/duration/resolution of AKI. RRT (receipt, timing, dependence). MAKE at hospital discharge, 30 days and 90 days (mortality, doubling SCr, dialysis dependence). Mortality (ICU, in-hospital, long-term). Use of IMV strategies (LPV, neuromuscular blockade, prone positioning, inhaled nitric oxide/vasodilators). Extracorporeal organ support (ECMO, ECCO2R). Length of stay (ICU, hospital). Fluid balance (daily, cumulative). Use of diuretics for fluid management (type, cumulative dose).

AKI, acute kidney injury; ARDS, acute respiratory distress syndrome; ARF, acute respiratory failure; DOI, digital object identifier; ECCO₂R, extracorporeal carbon dioxide removal; ECMO, extracorporeal membrane oxygenation; FiO₂, fraction of inspired oxygen; ICU, intensive care unit; IMV, invasive mechanical ventilation; LPV, lung-protective ventilation; MAKE, major adverse kidney event; PEEP, positive end-expiratory pressure; RRT, renal replacement therapy; SCr, serum creatinine; TV, tidal volume.

Primary outcome

The primary outcome will be the pooled proportion of critically ill patients who develop new-onset or worsening AKI after initiation of IMV. Initiation of IMV will be considered time zero. AKI will be defined according to the definition used in each eligible study and, where feasible, standardised to the Kidney Disease: Improving Global Outcomes (KDIGO) definition of AKI. New-onset AKI will refer to AKI occurring after initiation of IMV in patients who did not have AKI at or before IMV initiation. Worsening AKI will refer to progression of pre-existing AKI after initiation of IMV in patients who already had AKI at or before IMV initiation, based on the AKI criteria or staging reported by the individual study. Patients with pre-existing AKI at the time of IMV initiation will not be considered to have new-onset AKI.

Secondary outcomes

Secondary outcomes will include: (1) severity of AKI; (2) receipt of RRT including the timing of initiation after IMV initiation, where reported; (3) major adverse kidney events at hospital discharge, 30 days and 90 days, including mortality, RRT use, doubling of serum creatinine, estimated glomerular filtration rate loss and RRT dependence at follow-up; (4) RRT dependence at hospital discharge or other reported follow-up time points among survivors; (5) AKI resolution or recovery, including duration of AKI, according to the definitions and time points reported by individual studies; (6) IMV-related outcomes, including time to successful weaning, failure of IMV weaning attempts as defined by individual studies and use of tracheostomy; (7) lengths of ICU and hospital stay; (8) use of IMV strategies and adjunctive respiratory interventions, including LPV TV 6 mL/kg), neuromuscular blockade, prone positioning, inhaled nitric oxide or other inhaled pulmonary vasodilators, extracorporeal membrane oxygenation (ECMO) and extracorporeal carbon dioxide removal (ECCO₂R); (9) fluid balance (daily and cumulative); and (10) pharmacological strategies for fluid management, including loop diuretics, thiazide diuretics, acetazolamide, spironolactone and cumulative dose, where reported.

Subgroup analysis

Subgroup analyses will include the proportion of patients with AKI according to the following characteristics: (1) new-onset vs worsening AKI; (2) ARF/ARDS versus non-ARF/ARDS; (3) COVID-19 versus non-COVID-19; (4) specific IMV parameters, including mode (controlled vs assisted), TV (mL/kg ideal body weight), peak PEEP, peak plateau pressure, driving pressure and fraction of inspired oxygen (peak or averaged); and (5) duration of follow-up after IMV initiation (time-at-risk). Subgroup analyses will be undertaken only when enough studies and adequate between-study variation are available.

Evaluation of study quality

Two reviewers will independently assess the risk of bias of all included RCTs using the Cochrane Risk-of-Bias 2 (RoB 2) tool.¹² Any disagreements will be resolved through discussion or, if necessary, adjudication by a third reviewer.

The RoB 2 tool evaluates five domains of bias: (1) bias arising from the randomisation process; (2) bias due to deviations from intended interventions; (3) bias due to missing outcome data; (4) bias in measurement of the outcome; and (5) bias in selection of the reported result.¹² Each domain will be assessed using signalling questions, with judgments categorised as ‘low risk of bias’, ‘some concerns’ or ‘high risk of bias’.

Data analysis and synthesis

Study design and patient characteristics will be summarised narratively. The primary analysis will be descriptive and will estimate the pooled proportion of critically ill patients who develop new-onset or worsening AKI after initiation of IMV, with corresponding 95% CIs. For secondary outcomes, such as the severity of AKI, pooled proportions will also be reported where appropriate.

Meta-analyses will be conducted only when studies are considered sufficiently clinically and methodologically comparable with respect to the study population, exposure or intervention, comparator, outcome definition and duration of follow-up. Where appropriate, pooled estimates will be calculated using a random-effects model, with between-study variance estimated using the restricted maximum likelihood method. CIs for pooled estimates will be calculated using the Hartung-Knapp adjustment, where appropriate. For comparative analyses of secondary outcomes or subgroup analyses, dichotomous outcomes will be summarised using risk ratios or ORs with corresponding 95% CIs, as appropriate to the available data. Continuous outcomes will be summarised using weighted mean differences with 95% CIs when outcomes are measured using the same scale. When outcomes are reported as medians and IQRs, means and SDs will be estimated using validated mean-variance conversion methods.¹³

For multiarm trials, relevant intervention groups will be combined where appropriate to avoid double-counting participants in pairwise meta-analyses. When intervention groups cannot be combined, the shared comparator group will be divided across the relevant comparisons.

Statistical heterogeneity will be assessed through visual inspection of forest plots, Cochran’s Q (χ²) test and the I² statistic. Substantial heterogeneity will be considered at a Q test p value<0.10 or an I² value >50%. Where substantial clinical or methodological heterogeneity precludes meaningful quantitative synthesis, pooled estimates will not be reported and findings will instead be summarised narratively. Where appropriate, clinically relevant subgroup and sensitivity analyses will be performed to explore potential sources of heterogeneity.

Meta-regression analyses of study-level IMV-related variables, including TV, PEEP and duration of follow-up after IMV initiation, will be considered exploratory and performed only where sufficient studies and data are available. These analyses will be interpreted cautiously because study-level associations may be influenced by differences in underlying illness severity and other study characteristics.

Publication bias will be assessed using funnel plots when >10 studies are available for a given outcome. Additional methods, such as Egger’s regression test, will be applied when appropriate to evaluate potential small-study effects. All statistical analyses will be conducted using Stata V.17 (StataCorp, College Station, Texas, USA). The overall certainty of evidence for each outcome will be assessed using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) framework.

Patient and public involvement

This systematic review and meta-analysis has engaged DD, as a patient partner and collaborator. DD is a survivor of critical illness, where he received therapy with continuous RRT. He has been an engaged advocate for healthcare quality and safety, having served several roles in the health system, including Choosing Wisely Canada, the Health Quality Council of Alberta and the Critical Care Strategic Clinical Network, Alberta Health Services, Alberta, Canada. DD participated in the 31st Acute Disease Quality Initiative consensus meeting focused on endpoints for RCTs in AKI and currently serves as the Patient Partner on the international committee updating the KDIGO Clinical Practice Guidelines for AKI.

Discussion

IMV is a cornerstone in the management of ARF/ARDS, however, it can significantly influence lung-kidney interactions in a bidirectional manner. These effects are primarily mediated through haemodynamic alterations and neurohormonal activation. In addition, injurious IMV may trigger a systemic inflammatory milieu that contributes to the development of IMV-associated AKI.¹⁴ Several strategies, most notably LPV and conservative fluid management, have been developed to mitigate these deleterious consequences. Nevertheless, the rapid expansion of contemporary evidence in the post-COVID-19 era, characterised by widespread IMV use, has highlighted a persistent risk of AKI and increased RRT utilisation despite adherence to LPV strategies. Moreover, extracorporeal organ support has emerged as a promising adjunctive approach to facilitate LPV and potentially improve outcomes in this vulnerable population. Collectively, these evolving insights underscore the need to synthesise current evidence to inform clinical practice and guide future guideline development.

The potential role of ventilatory and extracorporeal strategies in reducing AKI risk will necessitate cautious interpretation. Although LPV and other respiratory or extracorporeal strategies may be associated with kidney outcomes, therapies such as RRT, ECMO, ECCO₂R are often initiated in patients with greater illness severity and are influenced by clinical treatment selection. Therefore, associations between these therapies and AKI outcomes may reflect underlying illness severity or indications for treatment rather than a direct protective or harmful effect of the intervention. Unless a relevant strategy was specifically randomised and kidney outcomes were prespecified or appropriately assessed, findings involving these therapies will be considered exploratory and hypothesis-generating rather than evidence of causality. Accordingly, this review will distinguish evidence derived from randomised comparisons of specific interventions from associations involving therapies that were not the primary randomised intervention.

This systematic review and meta-analysis will include an extensive literature search to comprehensively summarise and synthesise current knowledge regarding the association between IMV and AKI among critically ill patients. Studies will be rigorously screened and selected, with standardised data extraction and risk of bias assessment performed to ensure methodological quality. Prespecified subgroup analyses and assessments of publication bias are planned, and the certainty of evidence will be evaluated using the GRADE approach, thereby strengthening the interpretability and applicability of the findings.

Although the review is restricted to RCTs, substantial heterogeneity is anticipated due to variations in AKI definitions, IMV strategies across healthcare systems and differences in outcome reporting. Despite these limitations, the findings of this systematic review and meta-analysis are expected to have important clinical and research implications. A clearer understanding of the relationship between IMV and AKI may inform kidney-protective ventilation strategies, enhance risk stratification and support multidisciplinary approaches to organ protection in the ICU. Furthermore, identifying key knowledge gaps and methodological limitations may help refine future study designs and promote more consistent reporting of kidney-related outcomes in ventilated patients. Ultimately, this work aims to contribute to evidence-based critical care practices that balance effective respiratory support with the prevention of IMV-associated AKI.

Ethics and dissemination

Ethics approval is not required for this study, as all data included in this evaluation are already published and our study will not directly involve human participants. Findings will be disseminated through academic conferences and published in a peer-reviewed journal.