Incremental versus conventional haemodialysis in end-stage kidney disease: a systematic review and meta-analysis

ABSTRACT Background Appropriate dialysis prescription in the transitional setting from chronic kidney disease to end-stage kidney disease is still challenging. Conventional thrice-weekly haemodialysis (HD) might be associated with rapid loss of residual kidney function (RKF) and high mortality. The benefits and risks of incremental HD compared with conventional HD were explored in this systematic review and meta-analysis. Methods We searched MEDLINE, Scopus and Cochrane Central Register of Controlled Trials up to April 2023 for studies that compared the impacts of incremental (once- or twice-weekly HD) and conventional thrice-weekly HD on cardiovascular events, RKF, vascular access complications, quality of life, hospitalization and mortality. Results A total of 36 articles (138 939 participants) were included in this meta-analysis. The mortality rate and cardiovascular events were similar between incremental and conventional HD {odds ratio [OR] 0.87 [95% confidence interval (CI)] 0.72–1.04 and OR 0.67 [95% CI 0.43–1.05], respectively}. However, hospitalization and loss of RKF were significantly lower in patients treated with incremental HD [OR 0.44 (95% CI 0.27–0.72) and OR 0.31 (95% CI 0.25–0.39), respectively]. In a sensitivity analysis that included studies restricted to those with RKF or urine output criteria, incremental HD had significantly lower cardiovascular events [OR 0.22 (95% CI 0.08–0.63)] and mortality [OR 0.54 (95% CI 0.37–0.79)]. Vascular access complications, hyperkalaemia and volume overload were not statistically different between groups. Conclusions Incremental HD has been shown to be safe and may provide superior benefits in clinical outcomes, particularly in appropriately selected patients. Large-scale randomized controlled trials are required to confirm these potential advantages.


INTRODUCTION
In the USA, > 130 000 patients each year are newly diagnosed with end-stage kidney disease ( ESKD) requiring haemodialysis ( HD) [1 ].Patients with ESKD are routinely initiated on a conventional thrice-weekly HD regimen as a standard HD prescription worldwide.Despite being a life-sustaining treatment, HD is a life-altering process with a high cost and is associated with marked declines in health-related quality of life.In addition, an abrupt transition from pre-dialysis to dialysis with thriceweekly HD is associated with a significant increase in loss of residual kidney function ( RKF) and mortality during the first 3 months [2 , 3 ].
Incremental HD, with once-or twice-weekly HD as a means of allowing a gradual transition period, has recently been advocated.The dialysis dose of this approach is personalized according to RKF.Incremental HD provides a gentler start with less frequent HD and gradually increases the amount of dialysis as the RKF of the patient is progressively lost [4 ].It has been suggested that incremental HD may have several benefits, including preserving RKF, lowering costs and improving quality of life compared with conventional treatment.Despite the several potential benefits of incremental HD, not all maintenance HD patients are suitable for this approach.The safety-related issues of incremental HD, including insufficient clearance of uraemic solutes, volume overload and hyperkalaemia, are still a major concern.
There is growing interest and an increasing number of reports of the benefits of incremental HD on RKF; however, most of these studies include a small number of patients and observational designs.Therefore, the actual effects of incremental HD on various outcomes remain inconclusive.To address these important issues, we aim to synthesize the available evidence on the safety and efficacy of incremental HD.This systematic review and meta-analysis was conducted to evaluate the impacts of incremental HD on cardiovascular events, RKF, vascular access complications, quality of life, hospitalization and mortality compared with conventional HD in ESKD patients.

Data source and searches
This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses ( PRISMA) statement.To identify eligible studies, a search was conducted in MEDLINE, Scopus and the Cochrane Central Register of Controlled Trials from 1990 to April 2023.Reference lists of the obtained articles were also searched for relevant publications.Also, unpublished data were sought from ClinicalTrials.gov and conference abstracts.The study protocol was registered with PROSPERO ( CRD42022378215) .For the search, the following terms were used:

Study selection
We included studies if they were either randomized controlled trials ( RCTs) , non-randomized trials or observational cohorts.In order to be eligible for inclusion, the studies had to meet the following criteria: an original article or intervention studies in humans with ESKD that compared the effects and/or safety of incremental and conventional HD in ESKD patients.The intervention of interest was incremental HD ( once-or twiceweekly HD) compared with conventional HD ( thrice-weekly HD) .The outcomes included changes in urine output and quality of life before and after the intervention and clinical outcomes, comprising vascular access complications, volume overload, hyperkalaemia, cardiovascular events, hospitalization and mortality.Narrative reviews and case reports/series were excluded.
Two authors ( K.T. and K.J.) separately screened the titles and abstracts of all electronic citations and obtained the full-text articles for the comprehensive review.Both authors then independently re-evaluated these articles.Disagreements were resolved through consensus and arbitration by the third author ( P.S.) .

Data extraction
The study characteristics and outcomes of interest were independently reviewed by the authors.The data extracted included study authors, country, year of publication, number of patients, mean age, population characteristics and dialysis frequency ( once-, twice-or thrice-weekly HD) .The outcomes of interest that were extracted included urine output ( ml/24 hours) , quality of life and depression scores at baseline and at the end of the study.The incidence of death, hospitalization, cardiovascular events, loss of renal function, vascular access complications, volume overload and hyperkalaemia were also assessed.

Quality assessment
Two reviewers ( K.T. and K.J.) independently assessed the quality of the studies using the Newcastle-Ottawa Scale ( NOS) for observational studies and the Cochrane Risk of Bias Tool ( RoB 2) for RCTs.
The NOS consists of three quality domains: selection, comparability of the groups and outcome assessment.Each study received a score between 0 and 9, with a score ≥7 indicating high quality [5 ].The RoB 2 tool includes five domains: randomization process, deviations from the intended interventions, missing outcome data, measurement of the outcome and selection of the reported result.The summary graphic was created using the RobVis22 application to show the independent domains for risk of bias.The studies were classified as 'high risk,' 'some concerns' or 'low risk' based on their scores [6 ].

Data analysis
We used random effects model meta-analysis to compute the odds ratio ( OR) in binary variables, including loss of RKF, vascular access and HD complications, hospitalization, cardiovascular events and mortality.To analyse continuous variables, we used the mean difference ( MD) from baseline to the end of the study between incremental and conventional HD.All pooled estimates are presented with a 95% confidence interval ( CI) .Heterogeneity was assessed using the I 2 index and the Q-test P -value, with an I 2 index > 75% indicating medium-high heterogeneity [7 ].
To identify possible effect modifiers on the pooled analyses, sources of heterogeneity were explored by subgroup analyses according to study design and patient characteristics.We conducted a sensitivity analysis to assess the consistency of the results.Statistical significance was determined when the P -value was < .05.Publication bias was formally assessed using funnel plots and the Egger test [8 ].All analyses were performed using Comprehensive Meta-Analysis software version 2.0 ( www.metaanalysis.com;Biostat, Englewood, NJ, USA) .The forest plots were created by the 'forestplot' package in R software version 4.2.2 ( R Foundation for Statistical Computing, Vienna, Austria) .

Search results and characteristics of the studies
A total of 928 potentially relevant records were identified.The full-text screening was performed for 70 articles, and 36 studies [9 -44 ] ( 35 officially published articles and 1 abstract) fulfilled the eligibility criteria.A flow diagram of the study selection is presented in Fig. 1 .
The main characteristics of the studies are summarized in Table 1 .The included studies were published from 1999 to June 2022.There were 138 939 patients with a mean age ranging from 43.8 to 70 years.Thirty-one studies were observational studies, four were RCTs and one was a non-randomized controlled study.Fifteen studies were performed in Asia, ten in Europe, seven in North America, two in Africa, one in South Africa and one in Australia.
Incremental HD prescription in the enrolled studies referred to two HD sessions per week in 27 studies, one or two HD sessions per week in 8 studies and one HD session per week in combination with a very low protein diet in 1 study.There were RKF ( urea clearance ranging from ≥2-≥3 ml/min) or urine output criteria ( ≥500 ml/day) for incremental HD enrolment in 10 studies.

Risk-of-bias assessment
Table 1 summarizes the results of the risk of bias assessment of the included studies.The NOS scores of observational and nonrandomized studies enrolled in this meta-analysis ranged from 5 to 8 ( moderate to high quality) .Four RCTs had a low risk of bias according to the Cochrane RoB tool.

Volume overload
Two studies examined the incidence of volume overload.Murea et al. [12 ] and Vilar et al. [15 ] reported no significant difference in volume overload between patients who received incremental or conventional HD.

Quality of life
Four studies examined the impacts of incremental and conventional HD on quality of life.The data could not be combined into a meta-analysis due to the diversity of the quality-of-life ≥ 2 mL/min or UO ≥ 500 mL/ RKF ≥ 2 mL/min or UO ≥ 500 mL/d assessments.Overall, there were no significant differences in the quality of life between incremental and conventional HD. Park et al. [27 ] revealed no significant difference in the Kidney Disease Quality of Life short form and Beck Depression Inventory ( BDI) scores.Davenport et al. [20 ] reported no significant difference in BDI-II and nine-item Patient Health Questionnaire ( PHQ-9) scores between the two groups.In two RCTs, Vilar et al. [15 ] did not find a significant difference in the quality of life and depression as measured by the EQ-5D-5L ( EuroQol 5-dimension, 5-level) value and PHQ-9 score between the incremental and conventional HD groups.Murea et al. [11 ] observed no significant differences in PHQ-9 and Dialysis Symptom Index changes while revealing a significant reduction in anxiety using the seven-item Generalized Anxiety Disorder questionnaire in comparing conventional HD at 6 weeks.

Publication bias
As Egger's test results were mainly non-significant ( P > .05), together with a generally symmetrical funnel plot for the outcomes of the studies included in this meta-analysis, publication bias was less likely to occur ( Supplementary Figs.S1-S4) .

DISCUSSION
The present meta-analysis explored the efficacy and safety of incremental versus conventional HD.Thirty-six articles were identified.We assessed the impacts on cardiovascular events, urine output, vascular access and HD complications, quality of life, hospitalization and mortality.We reported significantly lower hospitalization, loss of RKF and reduction of urine output in incremental versus conventional HD.Mortality and cardiovascular events were lower in incremental HD when the participants had RKF ≥2 ml/min or urine output ≥500 ml/day.Regarding safety, there were no significant differences in HD complications, including hyperkalaemia and volume overload.
RKF provides effective and continuous solute clearance and is associated with lower mortality in HD patients.Initiation with abruptly intense HD, particularly where significant ultrafiltration or intradialytic hypotension occurs, may lead to a rapid decrease in RKF.The concept of a gradual approach as incremental HD, derived from peritoneal dialysis, is expected to preserve RKF in HD patients.A previous meta-analysis revealed that urine volume was higher in patients on incremental HD than in those on conventional HD [45 ].This finding is strengthened in our study, which included a larger number of studies and analysed the changes in urine output at baseline and the end of the study.We found that patients treated with incremental HD had less 24-hour urine output reduction and a lower RKF loss, defined by urine volume < 200 ml/day, compared with conventional HD.It is important to note that the wide variability of RKF measurements among the studies, including urea and creatinine clearances, with or without normalization by body surface area, or reported in terms of the rate of RKF decline, leads to limitations in combining all available data for meta-analysis.
Other potential benefits of less frequent HD include reductions in vascular access problems and dialysis-associated complications such as IDH, myocardial ischaemia, arrhythmia and catheter-related infections.We found that patients undergoing incremental HD had a significantly lower incidence of IDH and a tendency towards lower rates of vascular access complications, although this difference was not statistically significant.Additionally, our results showed that incremental HD is associated with fewer hospitalizations.In support of the present study, Caton et al. [46 ] also reported reduced hospitalization rates in patients receiving incremental HD compared with conventional HD.
In the present meta-analysis, overall mortality did not differ significantly between incremental and conventional HD, as discovered in the previous meta-analysis.However, participants in some studies received less frequent HD due to financial problems or lack of adequate healthcare services and therefore were not the optimal candidates for incremental HD.According to the Kidney Disease Outcomes Quality Initiative Clinical Practice Guidelines for Hemodialysis Adequacy [47 ], patients with urea clearance < 2 ml/min should undergo HD three times per week.We found that in studies in which participants had RKF of at least 2 ml/min or urine output ≥500 ml/day, incremental HD had significantly fewer cardiovascular events and mortality than conventional HD.Furthermore, complications associated with inadequate dialysis, including hyperkalaemia and volume overload, were not different between incremental and thrice-weekly These findings emphasize the importance of patient selection and RKF measurement in the incremental HD approach.However, due to the variations in RKF between patients undergoing incremental and conventional HD in these studies, it is important to note that better outcomes in patients with incremen-tal HD who have RKF might not result solely from a different HD prescription, but could also be influenced by the presence of RKF, which is known as an important factor associated with survival [48 ].In addition, when we excluded the studies that enrolled patients in once-weekly HD, we discovered that twice-weekly HD was associated with lower mortality than thrice-weekly HD.It is possible that the lack of strict requirements for RKF and regular monitoring in several studies that administered once-weekly HD may have contributed to suboptimal HD and poor outcomes.Implementing appropriate once-weekly HD remains challenging and insufficient data are available.Few reports described the cohorts or experience in a single centre with the successful application of once-weekly HD [49 , 50 ].Further work should be conducted through well-designed RCTs with optimal selection criteria to investigate the feasibility, safety and efficacy of once-weekly HD.
We observed that standardized Kt / V was lower in patients undergoing the incremental approach compared with those receiving thrice-weekly HD.However, there were no significant differences in serum haemoglobin, phosphate and hyperkalaemia levels between the two groups.In addition, patients treated with incremental HD had significantly lower levels of serum β2M than those on conventional HD.This could be due to better RKF preservation in incremental HD and may contribute to fewer cardiovascular events and lower mortality in this group [51 ].Our results also showed that while patients undergoing incremental HD had lower protein intake as measured by a lower nPCR compared with those on conventional HD, there was no significant difference in their serum albumin levels.It is important to consider the appropriate level of protein intake for patients on incremental HD, as excess protein can lead to impaired renal function while insufficient protein intake can cause malnutrition.Further research should be conducted to determine the optimal protein intake in this population.
An improvement in quality of life, both physical and mental, is another possible benefit of incremental HD.Common complaints that can negatively impact the patient's quality of life, including prolonged transit time to and from the dialysis centre, intradialytic symptoms and the recovery period post-dialysis, may be alleviated by applying a less frequent dialysis regimen [52 ].However, assessments of the quality of life, anxiety and depression of patients were based on a variety of scores and were difficult to combine or compare across the studies.Most of the studies reported insignificant impacts of incremental HD on quality of life compared with conventional HD.Of note, one RCT demonstrated the superiority of incremental HD over conventional HD in reducing anxiety.The impacts of incremental HD on quality of life should be explored in future studies.
There are some limitations in our study.First, the decision to prescribe incremental or conventional HD varied among the included studies and may contribute to heterogeneity in the results.Second, variations in outcome definition and the duration of follow-up may confound the analysis.Lastly, it is important to note that the majority of the included studies are observational, which may introduce limitations in terms of bias and the interpretation of causality.Currently, several RCTs have been registered for the protocol and are being conducted to determine the efficacy and impact of incremental HD [53 -56 ].

CONCLUSIONS
Current evidence demonstrates that incremental HD is safe and potentially provides reductions in hospitalization and loss of RKF compared with conventional HD.In appropriately selected patients, incremental HD may improve cardiovascular outcomes and reduce mortality.Further well-designed RCTs are needed to fully assess the efficacy and impact of incremental HD compared with conventional HD and to determine whether more frequent than twice-weekly HD should remain the standard of care.

Table 1 :
Characteristics of the studies included in the meta-analysis.Scale for assessing the quality of observational studies.b ROB 2 for assessing the quality of RCTs.eGFR: estimated glomerular filtration rate; KRU: renal urea clearance.

Figure 4 :
Figure 4: Subgroup analyses for cardiovascular events and mortality.