Objectives
This is a protocol for a Cochrane Review (intervention). The objectives are as follows:
To evaluate the benefits and harms of incorporating bioimpedance spectroscopy into the clinical assessment for estimating target weight in maintenance haemodialysis recipients compared with physical examination alone.
Background
Description of the condition
Chronic kidney disease (CKD) is an emerging public health problem worldwide [1]. The global prevalence of the disease was estimated at 850 million people in 2022 [2], with a higher prevalence in countries with a lower socio‐demographic index [3]. CKD rose from the 13th leading global cause of death in 2000 to the 10th in 2019, and it is projected to become the fifth cause of death by 2040 [4]. Its aetiology varies by region, with diabetes mellitus, glomerulonephritis, and polycystic kidney disease being the most common risk factors in high‐income countries and middle‐income countries. In contrast, in low‐income countries, recurrent acute kidney injury, post‐infection glomerulonephritis, and environmental exposures are more prevalent [5]. The diagnosis of CKD is based on structural (glomerular or interstitial fibrosis) or functional abnormalities (reduced glomerular filtration rate, albuminuria, or blood electrolyte disturbances) of the kidneys that persist for longer than three months [6]. In kidney failure, the most advanced stage of CKD, patients require kidney replacement therapy (KRT). Haemodialysis (HD) is the most frequently used KRT method [7]. Despite being a life‐sustaining therapy, HD is intermittent and lacks the sophisticated kidney apparatus to regulate body fluids. The volume of fluids to be removed by an HD session (ultrafiltration) is usually calculated as the difference between the individual's actual weight and the estimated target weight. The target weight is the weight at which there is no sign of volume overload (oedema, lung crackles, or hypertension) or depletion (hypotension, cramps) [8]. However, physical examination does not accurately measure volume status [9, 10]. The often resulting volume overload is associated with higher blood pressure [10, 11], more frequent occurrences of hospitalisations [12], poorer quality of life (QoL) [13], and shorter survival [14, 15] among maintenance HD patients [11, 16]. On the other hand, excessive volume removal during HD can induce hypotensive episodes [17], leading to loss of residual kidney function [18], cognitive decline [19], and vascular access thrombosis [20], all associated with shorter and poorer quality survival [21].
Description of the intervention and how it might work
The pitfalls of physical examination alone in determining target weight prompted the search for a technical, quantitative assessment of volume status in HD patients. Initially, vena cava echography and cardiac natriuretic peptides were evaluated [22, 23] and, more recently, bioimpedance techniques, such as bioimpedance spectroscopy, were proposed as an ancillary technology for target weight estimation [24, 25, 26, 27]. Bioelectrical impedance analysis utilises the electrical properties of tissues to estimate their various components. Bioimpedance spectroscopy is a bioelectrical impedance analysis modality that uses a broader range of frequencies, which allows better differentiation between intra‐ and extracellular water [28]. The bioimpedance spectroscopy device software performs biophysical modelling on the impedance meter data, fitting the spectral data to the Cole‐Cole model using non‐linear curve fitting [29]. This procedure generates resistance (R) and reactance (Xc, capacitance resistance of the cell membrane). Cole‐Cole model terms are then applied, and equations derived from the Hanai mixture theory account for the presence of nonconducting elements [30]. Assuming that lower frequency currents pass through the extracellular water only, while higher frequencies pass through both the extracellular and intracellular water, these compartments are individually estimated [28]. Although no widely accepted structured algorithm to integrate this information is available, the extracellular to intracellular water ratio, raw impedance data, or a fluid overload parameter determined by most devices may be a helpful adjunct in target weight estimation.
The assumptions required for bioimpedance spectroscopy include measurements performed before the HD session, with the subject in the supine position and arms and legs away from the body, electrodes placed on the ipsilateral upper and lower limbs, and avoiding the side of peripheral blood access. Bioimpedance spectroscopy can be less accurate in extremes of body composition (obesity or leanness) [31], situations of altered fluid distribution (e.g. critically ill patients), and severe oedema or ascites. It is also noteworthy that fluid overload may be due to inflammation, malnutrition or other causes when increasing ultrafiltration is not feasible or desirable. Bioimpedance spectroscopy is considered a safe and non‐invasive technique. However, some potential adverse effects or limitations have been reported, such as skin irritation on electrode sites and possible interference with implanted medical devices such as pacemakers or implantable cardioverter defibrillators. The safety of the measure in pregnant women is unclear [32].
Why it is important to do this review
Clinical examination has low accuracy in estimating the target weight of HD recipients [10]. As a result, this population often suffers from fluid overload or depletion and its negative prognostic implications. Ancillary technologies like bioimpedance spectroscopy may contribute to more precise ultrafiltration settings during HD. However, previous studies evaluating bioimpedance spectroscopy‐guided target weight estimation amongst persons on maintenance HD found no effect on clinical outcomes such as survival [33, 34, 35, 36]. The results regarding hospitalisation [33, 35, 36] and complication [33, 35, 37, 38, 39, 40, 41] rates are heterogeneous. However, most studies that evaluated these clinical endpoints have short follow‐up times and are underpowered for the outcomes. Some previous studies have found a positive impact of bioimpedance spectroscopy‐guided target weight estimation on surrogate endpoints such as blood pressure control [24, 42] and arterial stiffness [34, 42], but these trials also include small samples and highly heterogeneous results. In addition, most of the studies used the fluid overload parameter loosely as a complement to clinical data to adjust target weight (pragmatic decision‐making) [34, 36, 37, 38, 39, 40, 41], and some of them have the fluid overload results included in an algorithm guiding the target weight estimation (structured decision‐making) [33, 35]. The frequency of assessment of fluid overload by bioimpedance spectroscopy is also highly variable among studies. Most studies used bioimpedance spectroscopy to reevaluate the target weight once a month [33, 37, 39], but this interval ranged from two weeks [38] to six months [36].
Three systematic reviews with meta‐analysis on technological adjuncts to guide fluid management in individuals undergoing dialysis have previously been published [43, 44, 45]. In addition to including non‐randomised studies [44], using different devices, such as other multifrequency impedance equipment [43, 44] or other types of technological adjuncts (ultrasound, blood volume monitoring, measurements of natriuretic peptides, and chest radiograph) [45], and including all modalities of dialysis (HD and peritoneal dialysis) [43, 44, 45], all three are outdated, with randomised controlled trials (RCTs) on the issue having recently been published [36, 37, 40].
To the best of our knowledge, we propose the first systematic review to focus exclusively on bioimpedance spectroscopy as an adjunct to volume assessment in HD recipients. Assuming it is feasible, a meta‐analysis aims to pool the results of the small RCTs to achieve enough power for a more accurate estimate of the effect of the bioimpedance spectroscopy‐guided intervention on meaningful clinical endpoints for the HD population.
Objectives
To evaluate the benefits and harms of incorporating bioimpedance spectroscopy into the clinical assessment for estimating target weight in maintenance haemodialysis recipients compared with physical examination alone.
Methods
Criteria for considering studies for this review
Types of studies
We will include all RCTs that compare bioimpedance spectroscopy‐guided target weight estimation (alone or combined with clinical data or another tool, with a pragmatic or structured decision flow including a pre‐designed algorithm) with target weight estimation based on clinical examination alone. Quasi‐RCTs will not be included. There will also be no restrictions regarding the year of publication, language, or report status.
Types of participants
We will include studies whose participants are adults (18 years or older) diagnosed with kidney failure and undergoing maintenance in‐centre HD, in facilities in or out of hospital, with sessions of at least three hours three times per week, for three months or more. Studies with participants without kidney failure or with kidney failure but treated in other settings or by different modalities of KRT (home HD, peritoneal dialysis, or kidney transplantation) or participants under 18 years of age will be excluded. In studies involving only a subset of eligible participants, individual data will be sourced from repositories or by directly contacting the study authors. Whenever individual data is not found, data will be included if the eligible subset represents the majority of the sample.
Types of interventions
We will include all studies that compare bioimpedance spectroscopy‐guided target weight estimation with target weight estimation based on clinical examination alone. The intervention group will consist of bioimpedance spectroscopy‐guided target weight estimation combined exclusively with clinical data or other tools that evaluate hydration/volaemic status, such as lung ultrasound or vena cava diameter assessed by ultrasound. The decision‐making process guiding using bioimpedance spectroscopy data for target weight estimation may be pragmatic (based on clinical impression alone) or structured (based on a pre‐designed algorithm). These co‐interventions will be allowed only for the intervention group. The comparison group should always rely on target weight estimation based on clinical examination alone, though the decision flow may be pragmatic or based on a pre‐designed algorithm. No other co‐intervention will be allowed.
Outcome measures
There will be no restriction of studies based on the reported outcomes. However, clinical endpoints are considered of the highest interest. Outcomes considered critical or important to end users, such as consumers of health services, health professionals and policymakers, will be selected for inclusion in the analyses. The Core Outcome Measures in Effectiveness Trials (COMET) [46] and the International Consortium for Health Outcomes Measurements (ICHOM) [47] were consulted. Two potential adverse effects will be included: intra‐dialytic complications and vascular access survival. The measurement taken closer to the specified time point will be included in the analysis. Studies that include the intervention and population of interest but lack the relevant outcomes will be searched in registries or published protocols to identify potential reporting bias.
Critical outcomes
| Table 1. Critical outcomes (measured at a short time (six months) or a long time (one year or more) | |
| Outcome domain | Outcome measures |
| Survival | Time until death (in months) (continuous variable) |
| Quality of life | Standardised score (physical role limitations, bodily pain and vitality domains and physical and mental component summary score) of the Short Form‐36 Health Survey Questionnaire (SF‐36) (recommended by ICHOM). The scores of the Kidney Disease Quality of Life (KDQoL) instrument may be used as a second choice (continuous variable) |
| Hospitalisation | Number of hospital admissions per unit of time (year) due to any cause (continuous variable) |
Important outcomes
| Table 2. Important outcomes (measured at a short time (6 months) or a long time (1 year or more) | |
| Outcome domain | Outcome measures |
| Intra‐dialytic complications | Number of episodes of intra‐dialytic hypotension, dizziness, or cramps per month of follow‐up (continuous variable) |
| Acute fluid overload or CV events | Number of hospitalisations per year related to cardiovascular or cerebrovascular events or episodes of acute dyspnoea relieved by HD ultrafiltration or associated with pulmonary congestion detected by image exam (continuous variable) |
| Blood pressure | Systolic blood pressure (mmHg) at the last follow‐up (continuous variable) |
| Vascular access survival | Time from confection until loss of vascular access (arteriovenous fistula or grafts) (continuous variable) |
Search methods for identification of studies
Electronic searches
The Cochrane Kidney and Transplant Information Specialist will search the following databases for RCTs without language, publication year, or publication status restrictions.
Cochrane Kidney and Transplant Specialised Register via MeerKat (a software application built on Microsoft Access used to manage the database). MeerKat is a study‐based register used to store bibliographic and study details, link multiple reports to a single study, link studies to reviews, and track the progress of reviews.
Cochrane Central Register of Controlled Trials (CENTRAL) via the Cochrane Library (latest issue)
Ovid MEDLINE(R) ALL (from 1946)
Embase.com records as part of the search of CENTRAL, as described in How CENTRAL is created (https://www.cochranelibrary.com/central/central-creation)
US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (https://www.clinicaltrials.gov) as part of the search of CENTRAL
World Health Organization International Clinical Trials Registry Platform (https://trialsearch.who.int) as part of the search of CENTRAL.
The Information Specialist has designed search strategies for MEDLINE, Embase, and CENTRAL. Please see Supplementary material 1 for the search terms used. MEDLINE searches combine the subject strategy adaptations with the sensitivity and precision maximising search strategy designed by Cochrane for identifying RCTs (as described in the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0) and updated by us to account for new MeSH study types and wider use of machine indexing by the National Library of Medicine [48].
Searching other resources
We will search the reference lists of review articles and relevant studies, clinical practice guidelines, and the grey literature sources OpenGrey and ProQuest Dissertations & Theses (e.g. abstracts, dissertations, and theses).
We will check the reference lists of included studies and any relevant systematic reviews identified for further references to relevant studies.
We will check the included studies for retractions and errata via the Retraction Watch Database and report the search dates in the review (https://retractiondatabase.org).
We will search Epistemonikos for related systematic reviews (https://www.epistemonikos.org).
We may contact the original authors or funders of included studies for clarification and further data if study reports are unclear.
Data collection and analysis
Selection of studies
The search strategy described will be used to obtain titles and abstracts of studies that may be relevant to the review. The titles and abstracts will be screened independently by two review authors (RKN, ERK), who will discard studies that are not applicable; however, studies and reviews that might include relevant data or information on studies will be retained initially. Two review authors will independently assess the retrieved abstracts and, where necessary, the full text of these studies to determine which studies satisfy the inclusion criteria. Disagreements will be resolved in consultation with a third review author (NF).
Data extraction and management
Two review authors (RKN, GP) will perform data extraction independently using a custom‐built data system (Covidence) previously piloted. Disagreements will be resolved in consultation with a third review author (EK). Studies reported in non‐English language journals will be translated before assessment. Where more than one publication of one study exists, reports will be grouped together, and the publication with the most complete data will be used in the analyses. Where relevant outcomes are only published in earlier versions, this data will be used. Any discrepancies between published versions will be highlighted.
The extraction form will initially identify the extractor's name, the extraction date, and each report from which data was extracted. The reports will be classified as included (whether they fulfil inclusion criteria) or excluded (describing the reasons for exclusion).
The following information will be collected.
Study method
Details about the design of the RCT (e.g. parallel, factorial, cross‐over), whether single or multicentre (including the number of recruiting centres), recruitment and sampling procedures, dates of first and last enrolment, length of participant follow‐up, the methods for random sequence generation and allocation concealment, who was masked, how missing data was handled, which unit of analysis was employed, the statistical methods used and covariates included in the model, the likelihood of biases, sources of funding or support for the study, potential conflicts of interest of authors.
Participants
Setting, region or countries where participants were recruited, eligibility and diagnosis criteria, and characteristics of participants at baseline (age, sex, comorbidity, dialysis vintage, and other relevant data).
Interventions
Intervention group
Timing regarding HD and frequency of assessment, criteria used to evaluate the quality of the bioimpedance spectroscopy data and length of intervention. The software versions linked to the devices and the number and types of electrodes used will be extracted from all included studies. The protocol for using bioimpedance spectroscopy readings to classify individuals according to volume status. Which parameters from bioimpedance spectroscopy were used for this classification (e.g. extracellular water to intracellular water ratio, fluid overload, raw impedance data). Data on who applied the bioimpedance spectroscopy evaluation, whether they were previously trained, and the number of planned bioimpedance spectroscopy evaluations that were lost. Whether a co‐intervention to evaluate hydration status was employed (such as lung or superior vena cava ultrasound evaluation), and with which frequency, timing, and length it was employed. Which decision algorithm was employed (if any was used).
Control group
The frequency and timing of clinical evaluation to estimate target weight, who performs and which components of the clinical assessment (e.g. chest auscultation, blood pressure evaluation, search for oedema, anamnesis questions), how many evaluations were lost, any co‐intervention to evaluate hydration status that was employed (detailing frequency, timing, and length), the decision algorithm that was utilised (if any was used) and the similarity with that employed in the intervention group.
It will be evaluated if the operationalisation structure and the frequency of reassessment for both arms (intervention and control groups) were similar. Any differences in these variables between the arms will be evaluated as a confounding factor.
Outcomes
The outcomes domains proposed in protocols or registries and the methods section of the paper will be compared with the outcomes results described in the reports, and differences will be reported. The time points of interest will be six months (short‐term follow‐up) or 12 months (long‐term follow‐up), with a preference for long‐term follow‐up.
Survival: time from the beginning of the survey to death, and exposed persons during this time. The data will be aggregated as time to death in months and the number of exposed persons per group.
QoL: post‐intervention standardised score of the domains physical role limitations, bodily pain, and vitality, and the standardised scores of the physical and mental component summary of the SF‐36 (score 0 to 100, with higher scores indicating better QoL). Data will be aggregated as the mean and SD of the scores in each group.
All‐cause hospitalisation: the number of hospital admissions and the number of exposed persons at each time point, from which the rate of hospitalisations will be generated, will be extracted. Data will be aggregated as the sum of the number of hospital admissions and the number of exposed individuals at each time period. If the variable is described in a different format (e.g. time to event (yes/no)), we will contact the author to obtain access to the raw data for generating the hospitalisation rate, facilitating a proper meta‐analysis.
Acute volume overload or cardiovascular events: the number of hospitalisations related to cardiovascular (acute myocardial infarction) or cerebrovascular (stroke) events, the number of extra treatments due to volume overload, and the number of hospitalisations due to episodes of volume overload. These events will be extracted separately. Acute volume overload events will be considered acute dyspnoea relieved by ultrafiltration in HD or pulmonary congestion detected by image exam. The number of exposed persons in the short and long‐term periods will also be extracted. Data will be analysed as the sum of the number of events and the number of exposed persons at each period of time, or separately, according to the data available from the included studies. If the variable is described differently (e.g. time to event (yes/no)), we will contact the author to access the raw data to generate the event rate and allow a proper meta‐analysis.
Blood pressure: the systolic blood pressure, measured in mmHg, will be extracted for each group at the last follow‐up. Data will be aggregated as the mean systolic blood pressure at each time period. The time point and setting at which the blood pressure was measured will also be extracted (before, during, or after dialysis, in the centre, or at home), and the type of equipment used in this measurement (aneroid, digital, or ambulatory monitor). Only data measured at similar time points and settings and with similar equipment will be summarised together.
Intra‐dialytic complications: the number of episodes of intra‐dialytic hypotension, dizziness, or cramps and the number of exposed persons per follow‐up period for each group will be extracted separately. The data will be analysed individually or aggregated as the sum of the events and the sum of exposed persons in each group per period, according to the data available in the included studies. If the variable is described differently (e.g. time to event (yes/no)), we will contact the author to access the raw data to generate the event rate and allow a proper meta‐analysis.
Vascular access survival: the time from access creation to loss of access or low‐flow arteriovenous fistula or grafts, and the number of exposed persons per follow‐up time for each group will be extracted. Data will be aggregated as the sum of the events and the sum of exposed persons in each group per period.
Results
Results will be described for each group and for each outcome at each planned time point, the number of participants randomly assigned and included in the analysis, the number of participants who withdrew, were lost to follow‐up, or were excluded, and the reasons for their exclusion. These results will be presented as a flowchart.
Summary data will be presented for each planned outcome and each group, as mean time and standard deviation (SD) for time‐related outcomes (survival and vascular access survival), means and SD for continuous variables (QoL and blood pressure), and incidence rates for all‐cause hospitalisations, acute volume overload or cardiovascular events and intra‐dialytic complications.
Between‐group estimates will be presented as hazard ratios (HR) for survival and vascular access survival, mean differences for quality of life and blood pressure, and incidence rate ratios for all‐cause hospitalisations, acute volume overload or cardiovascular events, and intra‐dialytic complications.
Risk of bias assessment in included studies
All the critical and important outcomes (seven) will be evaluated for risk of bias in the included studies, using the Risk of Bias 2 (RoB 2) assessment tool [49]. The risk of bias will be evaluated for the proposed time points of six and 12 months follow‐up and measures defined in Outcome measures, using the effect of assignment as the effect of interest. The risk of bias in cluster trials will be evaluated with a RoB 2 tool for clusters [50]. In cross‐over trials, considering that only the first stage of the study will be included in the review, RoB 2 will be used to assess the risk of bias. The risk of bias will be independently evaluated by two review authors (GP, RKN), and any disagreements will be discussed between the evaluators, aiming for a consensual decision. If it is impossible to achieve an agreement, a third review author (NF) will be consulted.
The following domains will be assessed by the RoB 2 assessment tool using its specific signalling questions.
Bias arising from the randomisation process
Bias due to deviations from intended interventions
Bias due to missing outcome data
Bias in the measurement of the outcome
Bias in the selection of the reported result.
The response options for these questions are:
Yes;
Probably yes;
Probably no;
No; and
No information.
The response to the signalling questions will be used in the RoB 2 algorithms to reach a proposed risk of bias judgement and assign one of three levels to each domain:
Low risk of bias;
Some concerns; or
High risk of bias.
The overall risk of bias in the trial will be considered low if the risk of bias for all domains is also considered low; some concerns if at least one domain is considered to raise some concerns; and high risk of bias if at least one domain is considered to be at high risk of bias or if several domains are judged as raising some concerns. The decisions about the risk of bias will be justified by short quotations and information in risk of bias tables to justify the judgment made as ‘Support for judgment.’
Measures of treatment effect
Results will be expressed as mean difference (MD) and 95% confidence interval (CI) for continuous outcomes (blood pressure and QoL). For time‐related outcomes (survival and vascular access survival), the HR with 95% CI will be used. The treatment effect on intra‐dialytic complications, acute volume overload or cardiovascular‐related events, and hospital admissions will be expressed as an incidence rate ratio with a 95% CI. According to the data available in most of the included studies, the measures of the treatment effect could be modified for some of the planned outcomes.
Unit of analysis issues
The unit of analysis will be the individual participant for most included trials. For cluster‐RCTs, we will use effect estimates from analyses that appropriately account for the clustering (e.g., multilevel models or generalised estimating equations) when reported. If such analyses are not available, we will adjust the data following the Cochrane Handbook for Systematic Reviews of Interventions guidance, using: 1) the number of clusters randomised to each group and the average cluster size or total number of participants; 2) outcome data that ignore clustering; and 3) an estimate of the intracluster correlation coefficient (ICC). When the ICC is not reported, we will use external estimates from similar studies or other reliable sources. Sensitivity analyses will be conducted to assess the impact of these assumptions on the robustness of the results.
We will include only the relevant arms in the analysis for trials with more than two intervention groups. When two or more arms are eligible for inclusion (e.g. two different BIS interventions compared to a control), we will combine the relevant bioimpedance spectroscopy groups into a single comparator if appropriate. If combining is not clinically or statistically appropriate, we will include each comparison separately and divide the shared control group equally among them, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions [51, 52].
We will include only data from the first period in cross‐over RCTs to avoid potential carryover effects.
During data extraction and analysis, multiple reports of the same study will be identified and treated as a single study. Any assumptions made during these adjustments will be explored in sensitivity analyses.
Dealing with missing data
Median and interquartile ranges will be converted to mean and SD by validated techniques for continuous outcomes[53]. When a 95% CI is available for an absolute effect measure, the standard error (SE) will be calculated as (upper‐limit ‐ lower limit)/3.92. The SE will be calculated from the P value as intervention effect estimation/Z (estimated from a standard normal distribution table). Effect estimates reported in the trial, such as MD or standardised MD (SMD) accompanied by measures of uncertainty, such as SE, 95% CI, or P value, will be used in the meta‐analysis in cases where the number of participants, mean, or SD is not reported, using the generic inverse variance method. For studies without information on variability, the SD will be borrowed from the study with the highest SD (worst‐case scenario), with subsequent sensitivity analysis to assess the impact of the assumption. Whenever we need it, missing information will be requested from the trial authors.
Reporting bias assessment
To assess the risk of bias due to missing a particular result of an identified study, the following approaches will be employed.
Identify the protocol, study plan, and/or analysis plan
Compare projected results with published results
Identification of outcomes described in methods and not presented in results
Contact authors for missing results.
Reasons for missing results will be carefully assessed using Outcome Reporting Bias In Trials (ORBIT) [54]. Studies with missing results because of the P value, magnitude, or direction of the result will be reported and considered to present a high risk of reporting bias.
The review authors will consult multiple bibliographic databases, trial registers, manufacturers, conference abstracts, theses, government reports, regulatory websites, and study authors or sponsors to assess the risk of bias due to an entire study report being selectively unavailable. In addition, for outcomes with ten or more studies included in the analysis, funnel plots will be used to evaluate the possibility of non‐reporting biases in cases where protocols or trial register records for most studies were unavailable. If funnel plots indicate a small study effect and there is evidence of between‐study heterogeneity (I2 > 0%), sensitivity analysis will be conducted by comparing fixed‐effect and random‐effects estimates. Selection models and regression‐based methods would also be considered, considering the shortcomings and limitations of these sensitivity analyses.
Synthesis methods
We will include studies of bioimpedance spectroscopy interventions delivered in maintenance HD with the following comparison.
Target weight guided by bioimpedance spectroscopy compared to target weight estimation based on clinical examination alone.
Meta‐analysis will only be considered if the studies are sufficiently homogeneous in terms of:
Types of participants;
Types of interventions;
Outcome measures.
The primary analysis will include all eligible studies, followed by a sensitivity analysis to assess the effects of restricting the analysis to RCTs with an overall ‘low’ or ‘low/some concerns’ risk of bias. The certainty of evidence will be summarised according to GRADE guidelines.
We will sort the studies by the size of the study effect, year of publication, and risk of bias. Risk‐of‐bias assessments will be presented in the review as both a full table and as forest plots, which will display risk‐of‐bias judgments alongside the results of each study included in the meta‐analysis. Meta‐analysis will only be carried out if we deem it appropriate for the studies included. If it is not possible to conduct a meta‐analysis, we will justify and describe the reasons, and the synthesis will be conducted and presented according to the Synthesis without meta‐analysis (SWiM) in systematic reviews: reporting guideline [55].
If meta‐analysis is done, it will be carried out using RevMan [56]. Due to the predicted diversity of intervention applications (frequency of bioimpedance spectroscopy assessment, bioimpedance spectroscopy alone or combined with other tools, structured or pragmatic decision‐making), we will adopt a random effect model using the DerSimonian and Laird inverse variance method of meta‐analysis. We will use Restricted Maximum Likelihood (REML) to calculate heterogeneity variance.
Investigation of heterogeneity and subgroup analysis
We will employ visual inspection of the forest plot and the I² statistic to evaluate heterogeneity. The I2 statistic evaluates the percentage of total variation across studies, which is likely due to statistical heterogeneity rather than sampling error [57]. A guide to the interpretation of I² values will be as follows:
0% to 40%: might not be important;
30% to 60%: may represent moderate heterogeneity;
50% to 90%: may represent substantial heterogeneity;
75% to 100%: considerable heterogeneity.
The judgment of the observed value of I2 will take into account (1) the magnitude and direction of effects and (2) the strength of evidence for heterogeneity (P value from the Chi2 test or a CI for I2).
To investigate the causes of substantial or considerable heterogeneity, we will thoroughly review the extracted data and its entry into RevMan. Whenever a sufficient number of studies are available, subgroup analysis will be conducted using study‐level variables.
We plan to stratify subgroups based on the method, using bioimpedance spectroscopy data for target weight estimation. This includes structured decision‐making, where using bioimpedance spectroscopy data informs an algorithm, and pragmatic decision‐making, where bioimpedance spectroscopy data loosely complements physical examination. Furthermore, we will consider the frequency of bioimpedance spectroscopy evaluations, categorised as monthly or more frequent versus less than monthly. Subgroup analysis will also be employed to evaluate interventions that associate bioimpedance spectroscopy with co‐interventions. These subgroup analyses were chosen as the method, frequency of the active intervention, and co‐interventions could potentially act as effect modifiers. To compare the subgroup effects of the intervention, we will conduct a test for heterogeneity across the subgroup results.
Equity‐related assessment
The authors do not feel an equity‐related assessment is appropriate in the context of our present review question. We initially aim to analyse the effectiveness of bioimpedance spectroscopy as an adjunct to clinical examination to estimate target weight in the general dialysis population through the proposed protocol. Whether this analysis demonstrates benefits and low risks in the general population analysed, we plan to proceed to analyse PROGRESS factors [58] through a study focusing on individual participants of the trials. This future study will present an appropriate review question focused on this issue. This next study involves contacting trial authors to gather information on the socio‐economic status, residence, race or ethnicity, occupation, gender, religion, education, and social capital of each individual included in the trials. Based on this data, we will compare the potential benefits and harms of bioimpedance spectroscopy as an adjunct to clinical examination to estimate the target weight of HD recipients according to the PROGRESS factors [58].
Sensitivity analysis
Sensitivity analyses will explore the influence of high risk of bias, small sample size, industry‐funded studies, and missing data on the meta‐analysis results. If the need is identified during the review process, other sensitivity analyses may be used [59].
The following criteria will be used for each sensitivity analysis.
High risk of bias according to RoB 2 assessment tool [49]
Intervention through bioimpedance spectroscopy alone or combined with another tool
Cluster RCT
Source of funding (industry‐funded studies)
Findings of small study effect by funnel plot plus heterogeneity (I2 > 0%) (missing data)
Studies with SD imputation.
Certainty of the evidence assessment
We will summarise the treatment effects, preferably evaluated at 12 months, and the certainty of the evidence using the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) guidelines [60] for each of the following outcomes in the summary of findings table.
Survival (time until death)
QoL (using SF‐36)
Hospitalisations (number of hospital admissions due to any cause)
Intra‐dialytic complications (number of episodes of intra‐dialytic hypotension, dizziness, or cramps per month)
Acute fluid overload and cardiovascular events (number of hospitalisations related to cardiovascular or cerebrovascular events or episodes of acute dyspnoea relieved by HD ultrafiltration or associated with pulmonary congestion detected by image exam)
Systolic blood pressure (mmHg)
Vascular access survival (time from confection until loss of vascular access).
Intra‐dialytic complications and vascular access survival are potential adverse effects. Two review authors (EK, NF) will independently assess the certainty of the evidence. Disagreements will be discussed between the evaluators, aiming for a consensual view of any downgrading decision. If consensus is not attained, a third review author will be consulted (RKN). The certainty of the evidence will be described as high, moderate, low, or very low. All assessments of the certainty of evidence (downgrading using GRADE) will be documented and justified.
Certainty of the evidence assessment will be measured from five domains.
Risk of bias or study limitations using the overall RoB 2 judgment
Heterogeneity or inconsistency of results
Indirectness of evidence
Imprecision of results
High probability of publication bias.
The body of evidence from RCTs begins with a high certainty rating. RCTs maintain the highest certainty rating when no concerns exist in any of the GRADE domains. If a reason is found for downgrading the evidence, it should be classified as 'serious' (downgrading the certainty by one level) or 'very serious' (downgrading the certainty by two levels). The rating of the overall RoB 2 judgment will feed directly into this GRADE domain. In particular, a ‘low’ risk of bias would indicate ‘no limitation’; ‘some concerns’ would indicate either ‘no limitation’ or ‘serious limitation’; and a ‘high’ risk of bias would indicate either ‘serious limitation’ or ‘very serious limitation.’ When an article exhibits concerns in three or more topics, the certainty of evidence will be deemed to be very low.
Consumer involvement
Consumers and other stakeholders were not involved in the development of this protocol. However, they will be actively engaged during the systematic review process to ensure the findings are relevant, accessible, and aligned with patient and public health priorities.
Level of involvement and approach
We will adopt a collaborative approach, where stakeholders – particularly patients and nephrology professionals – will contribute as advisors in key stages of the review. Their role will be to provide feedback on the interpretation of findings, enhance the accessibility of results, and support dissemination efforts.
Roles and stages of involvement
Stakeholders will be involved in the following stages of the review process.
Interpretation of findings: they will contribute insights into the clinical and real‐world implications of the results.
Review of the plain language summary: consumers will provide feedback to improve clarity and readability.
Dissemination and advocacy: stakeholders will help translate findings into key messages for policymakers, patient organisations, and the general public. Authors and stakeholders will disseminate the results on websites as press releases, on institutional websites, on social media, and at congresses as abstract presentations.
Authorship and engagement planning: one stakeholder will be invited to join the author team, and two additional stakeholders will collaborate on developing an engagement plan based on patient priorities.
Research methods for consumer involvement
Stakeholders will be engaged through structured discussions and feedback sessions to ensure their contributions are systematically integrated into the review. They will be recruited through a formal invitation posted on the Cochrane Engage platform. This approach will allow for transparent and inclusive recruitment, enabling diverse contributions to shape the review question, interpret findings, and enhance dissemination strategies. Their involvement will be transparently reported using the GRIPP2‐LF (Guidance for Reporting Involvement of Patients and Public) framework [61].
Supporting Information
Supplementary materials are available with the online version of this article: 10.1002/14651858.CD016044.
Supplementary materials are published alongside the article and contain additional data and information that support or enhance the article. Supplementary materials may not be subject to the same editorial scrutiny as the content of the article and Cochrane has not copyedited, typeset or proofread these materials. The material in these sections has been supplied by the author(s) for publication under a Licence for Publication and the author(s) are solely responsible for the material. Cochrane accordingly gives no representations or warranties of any kind in relation to, and accepts no liability for any reliance on or use of, such material.
Supplementary material 1 Search strategies
New
Additional information
Acknowledgements
We are grateful to Cochrane Kidney and Transplant who supported the authors in developing this review.
Editorial and peer‐reviewer contributions
The following people conducted the editorial process for this article:
Sign‐off Editor (final editorial decision): Catherine M Clase, Mc Master University, Canada;
Managing Editor (selected peer reviewers, provided editorial guidance to authors, edited the article): Justin Mann, Cochrane Central Editorial Service;
Editorial Assistant (conducted editorial policy checks, collated peer‐reviewers comments and supported editorial team): Addie‐Ann Smyth, Cochrane Central Editorial Service;
Copy Editor (copy editing and production): Narelle Willis, Cochrane Central Production Service;
Peer‐reviewers (provided comments and recommended an editorial decision): Jo‐Ana Chase, Cochrane Evidence Production and Methods Directorate (methods review), Jo Platt, Central Editorial Information Specialist (search review), Daniel Schneditz, Medical University of Graz, Division of Physiology & Pathophysiology, Otto Loewi Research Center for Vascular Biology, Immunology and Inflammation (Clinical Reviewer 1), Dr. Elangovan Krishnan, MBBS, M. Tech., MS, Ph.D, Executive Medical Director, STAR Education (Clinical Reviewer 2), Steven Brantlov, Department of Procurement & Clinical Engineering, Aarhus, Denmark (Clinical Reviewer 3).
Contributions of authors
KS: write‐up
ERK: sift and study selection, data extraction, synthesis, and GRADE assessment
NF: sift and study selection, data extraction, synthesis, risk of bias and GRADE assessment
RKN: sift and study selection, data extraction, synthesis, risk of bias, and GRADE assessment
GP: sift and study selection, data extraction, synthesis, risk of bias
MB: write‐up, and reviewing manuscript
Declarations of interest
The authors KS and MB have potential articles to be included in the systematic review.
The study is not industry‐controlled or industry‐supported.
ERK, NF, RKN, and GP declare no conflicts of interest.
No funding was received for this study. Authors with potential conflicts of interest were excluded from the sift and study selection, data extraction, synthesis, risk of bias assessment, and GRADE assessment. All restrictions imposed by the Cochrane Collaboration on authors of potentially included studies will be respected.
Sources of support
Internal sources
No sources of support provided
External sources
No sources of support provided
Registration and protocol
Cochrane approved the proposal for this review in September 2023.
Data, code and other materials
Data sharing is not applicable to this article as it is a protocol, so no datasets were generated or analysed.
References
- 1.Jha V, Garcia-Garcia G, Iseki K, Li Z, Naicker S, Plattner B, et al. Chronic kidney disease: global dimension and perspectives [Erratum in: Lancet. 2013 Jul 20;382(9888):208]. Lancet 2013;382(9888):260-72. [DOI: 10.1016/S0140-6736(13)60687-X] [DOI] [PubMed] [Google Scholar]
- 2.Kovesdy CP. Epidemiology of chronic kidney disease: an update 2022. Kidney International Supplements 2022;12(1):7-11. [DOI: 10.1016/j.kisu.2021.11.003] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.International Society of Nephrology. ISN-Global Kidney Health Atlas. A comprehensive resource on kidney care worldwide. htpps://www.theisn.org/initiatives/global-kidney-health-atlas/ (accessed 4 September 2025).
- 4.Foreman KJ, Marquez N, Dolgert A, Fukutaki K, Fullman N, McGaughey M, et al. Forecasting life expectancy, years of life lost, and all-cause and cause-specific mortality for 250 causes of death: reference and alternative scenarios for 2016-40 for 195 countries and territories. Lancet 2018;392(10159):2052-90. [DOI: 10.1016/S0140-6736(18)31694-5] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Batte A, Shahrin L, Claure-Del Granado R, Luyckx VA, Conroy AL. Infections and acute kidney injury: a global perspective. Seminars in Nephrology 2023;43(5):151466. [DOI: 10.1016/j.semnephrol.2023.151466] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stevens P, Ahmed SB, Carrero JJ, Foster B, Francis A, Hall RK, et al. Clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney International 2024;105(4 Suppl):S117-314. [DOI: 10.1016/j.kint.2023.10.018] [DOI] [PubMed] [Google Scholar]
- 7.Bello AK, Levin A, Lunney M, Osman MA, Ye F, Ashuntantang GE, et al. Status of care for end stage kidney disease in countries and regions worldwide: international cross sectional survey. BMJ 2019;367:l5873. [DOI: 10.1136/bmj.l5873] [DOI] [PubMed] [Google Scholar]
- 8.Sinha AD, Agarwal R. Can chronic volume overload be recognized and prevented in hemodialysis patients? The pitfalls of the clinical examination in assessing volume status. Seminar in Dialysis 2009;22(5):480-2. [DOI: 10.1111/j.1525-139X.2009.00641.x] [DOI] [PubMed] [Google Scholar]
- 9.Kim KE, Neff M, Cohen B, Somerstein M, Chinitz J, Onesti G, et al. Blood volume changes and hypotension during hemodialysis. Transactions - American Society for Artificial Internal Organs 1970;16:508-14. [PMID: ] [PubMed] [Google Scholar]
- 10.Agarwal R, Andersen MJ, Pratt JH. On the importance of pedal edema in hemodialysis patients. Clinical Journal of the American Society of Nephrology: CJASN 2008;3(1):153-8. [DOI: 10.2215/CJN.03650807] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ok E, Asci G, Chazot C, Ozkahya M, Mees EJ. Controversies and problems of volume control and hypertension in haemodialysis. Lancet 2016;388(10041):285-93. [DOI: 10.1016/S0140-6736(16)30389-0] [DOI] [PubMed] [Google Scholar]
- 12.Arneson TJ, Liu J, Qiu Y, Gilbertson DT, Foley RN, Collins AJ. Hospital treatment for fluid overload in the medicare hemodialysis population. Clinical Journal of the American Society of Nephrology: CJASN 2010;5(6):1054-63. [DOI: 10.2215/CJN.00340110] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Enia G, Torino C, Panuccio V, Tripepi R, Postorino M, Aliotta R, et al. Asymptomatic pulmonary congestion and physical functioning in hemodialysis patients. Clinical Journal of the American Society of Nephrology: CJASN 2013;8(8):1343-8. [DOI: 10.2215/CJN.11111012] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kalantar-Zadeh K, Regidor DL, Kovesdy CP, Van Wyck D, Bunnapradist S, Horwich TB, et al. Fluid retention is associated with cardiovascular mortality in patients undergoing long-term hemodialysis. Circulation 2009;119(5):671-9. [DOI: 10.1161/CIRCULATIONAHA.108.807362] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wizemann V, Wabel P, Chamney P, Zaluska R, Moissl U, Rode C, et al. The mortality risk of overhydration in haemodialysis patients. Nephrology Dialysis Transplantation 2009;25(5):1574-83. [DOI: 10.1093/ndt/gfn707] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Dekker MJ, Kooman J. Fluid status assessment in hemodialysis patients and the association with outcome: review of recent literature. Current Opinion in Nephrology & Hypertension 2018;27(3):188-93. [DOI: 10.1097/MNH.0000000000000409] [DOI] [PubMed] [Google Scholar]
- 17.Agarwal R, Alborzi P, Satyan S, Light RP. Dry-weight reduction in hypertensive hemodialysis patients (DRIP): a randomized, controlled trial. Hypertension 2009;53(3):500-7. [DOI: 10.1161/HYPERTENSIONAHA.108.125674] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jansen MA, Hart AA, Korevaar JC, Dekker FW, Boeschoten EW, Krediet RT, et al. Predictors of the rate of decline of residual renal function in incident dialysis patients. Kidney International 2002;62(3):1046-53. [DOI: 10.1046/j.1523-1755.2002.00505.x] [DOI] [PubMed] [Google Scholar]
- 19.Costa AS, Tiffin-Richards FE, Holschbach B, Frank RD, Vassiliadou A, Krüger T, et al. Clinical predictors of individual cognitive fluctuations in patients undergoing hemodialysis. American Journal of Kidney Diseases 2014;64(3):434-42. [DOI: 10.1053/j.ajkd.2014.02.012] [DOI] [PubMed] [Google Scholar]
- 20.Chang TI, Paik J, Greene T, Desai M, Bech F, Cheung AK, et al. Intradialytic hypotension and vascular access thrombosis. Journal of the American Society of Nephrology 2011;22(8):1526-33. [DOI: 10.1681/ASN.2010101119] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Stéfansson BV, Brunelli SM, Cabrera C, Rosenbaum D, Anum E, Ramakrishnan K, et al. Intradialytic hypotension and risk of cardiovascular disease. Clinical Journal of the American Society of Nephrology: CJASN 2014;9(12):2124-32. [DOI: 10.2215/CJN.02680314] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Leunissen KM, Kouw P, Kooman JP, Cheriex EC, DeVries PM, Donker AJ, et al. New techniques to determine fluid status in hemodialyzed patients. Kidney International - Supplement 1993;41:S50-6. [PMID: ] [PubMed] [Google Scholar]
- 23.Ohashi Y, Sakai K, Hase H, Joki N. Dry weight targeting: the art and science of conventional hemodialysis. Seminar in Dialysis 2018;31(6):551-6. [DOI: 10.1111/sdi.12721] [DOI] [PubMed] [Google Scholar]
- 24.Hur E, Usta M, Toz H, Asci G, Wabel P, Kahvecioglu S, et al. Effect of fluid management guided by bioimpedance spectroscopy on cardiovascular parameters in hemodialysis patients: a randomized controlled trial. American Journal of Kidney Diseases 2013;61(6):957-65. [DOI: 10.1053/j.ajkd.2012.12.017] [DOI] [PubMed] [Google Scholar]
- 25.Machek P, Jirka T, Moissl U, Chamney P, Wabel P. Guided optimization of fluid status in haemodialysis patients. Nephrology Dialysis Transplantation 2010;25(2):538-44. [DOI: 10.1093/ndt/gfp487] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Liu L, Sun Y, Chen Y, Xu J, Yuan P, Shen Y, et al. The effect of BCM guided dry weight assessment on short-term survival in Chinese hemodialysis patients: Primary results of a randomized trial - BOdy COmposition MOnitor (BOCOMO) study. BMC Nephrology 2020;21(1):135. [DOI: 10.1186/s12882-020-01793-x] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Covic A, Onofriescu M. Time to improve fluid management in hemodialysis: should we abandon clinical assessment and routinely use bioimpedance? Clinical Journal of the American Society of Nephrology: CJASN 2013;8(9):1474-5. [DOI: 10.2215/CJN.06930613] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Matthie JR, Withers PO, Van Loan MD, Mayclin PL. Development of commercial complex bio-impedance spectroscopic system for determining intracellular and extracellular water volumes. Proceedings of the 8th International Conference on Electrical Bio-Impedance 1992;1:203-5. [Google Scholar]
- 29.Cole KS. Membranes, Ions and Impulses: A Chapter of Classical Biophysics. 1st edition. Berkeley, CA: University of California Press, 1968. [AVAILABLE FROM:: https://www.ucpress.edu/books/membranes-ions-and-impulses/epub-pdf] [ISBN:: 9780520326514] [Google Scholar]
- 30.Hanai T. Electrical properties of emulsions. In: Sherman D, editor(s). Emulsions Science. 2nd edition. London: Academic Press, 1968:354-477. [Google Scholar]
- 31.Chertow GM. With bioimpedance spectroscopy, the errors get fat when the patients get slim. Jpen: Journal of Parenteral & Enteral Nutrition 2002;26(2):128-9. [DOI: 10.1177/0148607102026002128] [DOI] [PubMed] [Google Scholar]
- 32.Eyre S, Stenberg J, Wallengren O, Keane D, Avesani CM, Bosaeus I, et al. Bioimpedance analysis in patients with chronic kidney disease. Journal of Renal Care 2023;49(3):147-57. [DOI: 10.1111/jorc.12474] [DOI] [PubMed] [Google Scholar]
- 33.Huan-Sheng C, Yeong-Chang C, Ming-Hsing H, Fan-Lieh T, Chu-Cheng L, Tsai-Kun W, et al. Application of bioimpedance spectroscopy in Asian dialysis patients (ABISAD-III): a randomized controlled trial for clinical outcomes. International Urology & Nephrology 2016;48(11):1897-909. [DOI: 10.1007/s11255-016-1415-8] [DOI] [PubMed] [Google Scholar]
- 34.Onofriescu M, Hogas S, Voroneanu L, Apetrii M, Nistor I, Kanbay M, et al. Bioimpedance-guided fluid management in maintenance hemodialysis: a pilot randomized controlled trial. American Journal of Kidney Diseases 2014;64(1):111-8. [DOI: 10.1053/j.ajkd.2014.01.420] [DOI] [PubMed] [Google Scholar]
- 35.Siriopol I, Siriopol D, Voroneanu L, Covic A. Predictive abilities of baseline measurements of fluid overload, assessed by bioimpedance spectroscopy and serum N-terminal pro-B-type natriuretic peptide, for mortality in hemodialysis patients. Archives of Medical Science 2017;13(5):1121-9. [DOI: 10.5114/aoms.2017.68993] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Stigger K, Ribeiro LR, Cordeiro FM, Böhlke M. Incidence of hospital admissions in bioimpedance-guided fluid management among maintenance hemodialysis patients-Results of a randomized controlled trial. Hemodialysis International 2023;27(3):318-25. [DOI: 10.1111/hdi.13076] [DOI] [PubMed] [Google Scholar]
- 37.Davies SJ, Coyle D, Lindley EJ, Keane D, Belcher J, Caskey FJ, et al. Bio-impedance spectroscopy added to a fluid management protocol does not improve preservation of residual kidney function in incident hemodialysis patients in a randomized controlled trial. Kidney International 2023;104(3):587-98. [DOI: 10.1016/j.kint.2023.05.016] [DOI] [PubMed] [Google Scholar]
- 38.Patel HV, Annigeri RA, Kowdle PC, Rao BS, Seshadri R, Balasubramanian S, et al. Bioimpedance spectroscopy-guided ultrafiltration normalizes hydration and reduces intradialytic adverse events in hemodialysis patients. Indian Journal of Nephrology 2019;29(1):1-7. [DOI: 10.4103/ijn.IJN_150_18] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Ponce PA, Pham J, Gligoric-Fuerer O, Kreuzberg U. Fluid management in haemodialysis: Conventional versus Body Composition Monitoring (BCM) supported management of overhydrated patients. Portuguese Journal of Nephrology and Hypertension 2014;28(3):239-48. [Google Scholar]
- 40.Sommerer C, Felten P, Toernig J, Zeier M, Dikow R. Bioimpedance analysis is not superior to clinical assessment in determining hydration status: a prospective randomized-controlled trial in a Western dialysis population. Hemodialysis International 2021;25(3):380-90. [DOI: 10.1111/hdi.12919] [DOI] [PubMed] [Google Scholar]
- 41.Wang S, Li C, Hu Y, Liang M, Xu S. The protective effect of bioimpedance spectroscopy guided ultrafiltration on residual renal function in hemodialysis patients. Chinese Journal of Disease Control & Prevention 2017;12:939-42. [DOI: 10.1177/030006052110310] [DOI] [Google Scholar]
- 42.Onofriescu M, Mardare NG, Segall L, Voroneanu L, Cusai C, Hogas S, et al. Randomized trial of bioelectrical impedance analysis versus clinical criteria for guiding ultrafiltration in hemodialysis patients: effects on blood pressure, hydration status, and arterial stiffness. International Urology & Nephrology 2012;44(2):583-91. [DOI: 10.1007/s11255-011-0022-y] [DOI] [PubMed] [Google Scholar]
- 43.Covic A, Ciumanghel AI, Siriopol D, Kanbay M, Dumea R, Gavrilovici C, et al. Value of bioimpedance analysis estimated “dry weight” in maintenance dialysis patients: a systematic review and meta-analysis. International Urology & Nephrology 2017;49(12):2231-45. [DOI: 10.1007/s11255-017-1698-4] [DOI] [PubMed] [Google Scholar]
- 44.Scotland G, Cruickshank M, Jacobsen E, Cooper D, Fraser C, Shimonovich M, et al. Multiple-frequency bioimpedance devices for fluid management in people with chronic kidney disease receiving dialysis: a systematic review and economic evaluation. Health Technology Assessment (Winchester, England) 2018;22(1):131-7. [DOI: 10.3310/hta22010] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Beaubien-Souligny W, Kontar L, Blum D, Bouchard J, Denault AY, Wald R. Meta-analysis of randomized controlled trials using tool-assisted target weight adjustments in chronic dialysis patients. KI Reports 2019;4(10):1426-34. [DOI: 10.1016/j.ekir.2019.07.003] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Tong A, Crowe S, Gill JS, Harris T, Hemmelgarn B, Manns B, et al. Clinicians’ and researchers’ perspectives on establishing and implementing core outcomes in haemodialysis: semistructured interview study. BMJ Open 2018;8(4):e021198. [DOI: 10.1136/bmjopen-2017-021198] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Verberne WR, Das-Gupta Z, Allegretti AS, Bart HA, Biesen W, García-García G, et al. Development of an international standard set of value-based outcome measures for patients with chronic kidney disease: a report of the international consortium for health outcomes measurement (ICHOM) CKD working group. American Journal of Kidney Diseases 2018;73(3):372-84. [DOI: 10.1053/j.ajkd.2018.10.007] [DOI] [PubMed] [Google Scholar]
- 48.Lefebvre C, Manheimer E, Glanville J. Chapter 6: Searching for studies. In: Higgins JP, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011 (archived). Available from https://training.cochrane.org/handbook/archive/v5.1/.
- 49.Sterne JA, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898. [DOI: 10.1136/bmj.l4898] [DOI] [PubMed] [Google Scholar]
- 50.Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ 2011;343:d5928. [DOI: 10.1136/bmj.d5928] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Higgins JP, Eldridge S, Li T. Chapter 23: Including variants on randomized trials (chapter last updated October 2019). In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.5 (updated August 2024). Cochrane, 2024. Available from https://www.training.cochrane.org/handbook.
- 52.Higgins JP, Li T, Deeks JJ, editor(s). Chapter 6: Choosing effect measures and computing estimates of effect (chapter last updated August 2023). In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.5 (updated August 2024). Cochrane, 2024. Available from https://www.training.cochrane.org/handbook.
- 53.McGrath S, Katzenschlager S, Zimmer AJ, Seitel A, Steele R, Benedetti A. Standard error estimation in meta-analysis of studies reporting medians. Statistical Methods in Medical Research 2022;32(2):373-88. [DOI: 10.1177/09622802221139233] [DOI] [PubMed] [Google Scholar]
- 54.Kirkham JJ, Altman DG, Chan A, Gamble C, Dwan KM, Williamson P. Outcome reporting bias in trials: a methodological approach for assessment and adjustment in systematic reviews. BMJ 2018;362:k3802. [DOI: 10.1136/bmj.k3802] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Campbell M, McKenzie JE, Sowden A, Katikireddi SV, Brennan SE, Ellis S, et al. Synthesis without meta-analysis (SWiM) in systematic reviews: reporting guideline. BMJ 2019;368:l6890. [DOI: 10.1136/bmj.l6890] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Review Manager (RevMan). Version 9.10.0. The Cochrane Colloraboration, 2025. Available at https://revman.cochrane.org.
- 57.Higgins JP, Thompson GS, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327(7414):557-60. [DOI: 10.1136/bmj.327.7414.557] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Cochrane Methods Equity. PROGRESS-Plus. https://methods.cochrane.org/equity/projects/evidence-equity/progress-plus (accessed 4 September 2025).
- 59.Deeks JJ, Higgins JPT, Altman DG, McKenzie JE, Veroniki AA (editors). Chapter 10: Analysing data and undertaking meta-analyses [last updated November 2024]. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 6.5. Cochrane, 2024. Available from https://www.training.cochrane.org/handbook.
- 60.Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008;336(7650):924-6. [DOI: 10.1136/bmj.39489.470347.AD] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Staniszewska S, Brett J, Simera I, Seers K, Mockford C, Goodlad S, et al. GRIPP2 reporting checklists: tools to improve reporting of patient and public involvement in research. BMJ 2017;358:j3453. [DOI: 10.1136/bmj.j3453] [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary material 1 Search strategies
Data Availability Statement
Data sharing is not applicable to this article as it is a protocol, so no datasets were generated or analysed.