Bicarbonate versus lactate solutions for acute peritoneal dialysis
Abstract
Background
The high mortality rate among critically ill patients with acute kidney injury (AKI) remains an unsolved problem in intensive care medicine, despite the use of renal replacement therapy (RRT). Increasing evidence from clinical studies in adults and children suggests that the new peritoneal dialysis (PD) fluids may allow for better long‐term preservation of peritoneal morphology and function. Formation of glucose degradation products (GDPs) can be reduced and even avoided with the use of newer "biocompatible" solutions. However, it is still unclear if there are any differences in using conventional (lactate) solutions compared with low GDP (bicarbonate) solutions for acute PD.
Objectives
To look at the benefits and harms of bicarbonate versus lactate solutions in acute PD.
Search methods
We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE (from 1966), EMBASE (from 1980), Latin American and Caribbean Health Sciences Literature Database LILACS (from 1982), and reference lists of articles.
Date of last search: 6 May 2014.
Selection criteria
Randomised controlled trials (RCTs) comparing bicarbonate to lactate solution for acute PD.
Data collection and analysis
Two authors independently assess the methodological quality of studies. One author abstracted data onto a standard form, and a second author checked data extraction. We used the random‐effects model and expressed the results as relative risk (RR) for dichotomous outcomes and mean difference (MD) for continuous outcomes with 95% confidence intervals (CI).
Main results
We included one study (20 patients) in this review. In shock patients, bicarbonate did not differ from lactate with respect to mortality (RR 0.50, 95% CI 0.06 to 3.91); however there were significant differences in blood lactate (MD ‐1.60 mmol/L, 95% CI ‐2.04 to ‐1.16), serum bicarbonate (MD 5.00 mmol/L, 95% CI 3.26 to 6.74) and blood pH (MD 0.12, 95% CI 0.06 to 0.18). In non‐shock patients there was a significance difference in blood lactate (MD ‐0.60 mmol/L, 95% CI ‐0.85 to ‐0.35) but not in serum bicarbonate (MD 1.10 mmol/L, 95% CI ‐0.27 to 2.47) or blood pH (MD ‐0.02, 95% CI ‐0.02 to ‐0.06). Other outcomes could not be analysed because of the limited data available.
Authors' conclusions
There is no strong evidence that any clinical advantage for patients requiring acute PD for AKI when comparing conventional (lactate) with low GDP dialysis solutions (bicarbonate).
Plain language summary
Two dialysis solutions for acute peritoneal dialysis
Acute kidney injury (AKI) can be treated with either bicarbonate or lactate in acute peritoneal dialysis (PD). The aim of this review was to compare the effectiveness of bicarbonate versus lactate solution.
We identified only one small randomised controlled trial (RCT) (20 patients) after an extensive literature search, and we found no difference between bicarbonate and lactate for clinically important outcomes, such as mortality and adverse events. An initial reported benefit of bicarbonate over lactate was not confirmed by subsequent studies. In this systematic review of one RCT of patients with AKI, no significance difference was demonstrable between these two dialysis solutions.
Authors' conclusions
Background
Acute kidney injury (AKI) is a complex disorder that occurs in a variety of settings, with clinical manifestations ranging from a minimal elevation in serum creatinine (SCr) to anuric kidney failure (Mehta 2004; Palevsky 2006). AKI is present in 1% to 5% of patients at hospital admission (Kaufman 1991). The condition affects 15% to 20% of patients in intensive care units (ICUs) and reported mortality rates range from 50% to 70% in these patients (Albright 2001; Singri 2003; Thadhani 1996; Uchino 2006). The high mortality rate among critically ill patients with AKI remains an unsolved problem in intensive care medicine, despite the fact that renal replacement therapy (RRT) has been available for decades.
AKI affects 5% to 7% of hospitalised patients and has an unacceptably high mortality, particularly when RRT is required. Unfortunately, there are currently no specific therapeutic interventions available to accelerate recovery, once AKI has developed, with the mainstay of treatment being supportive with RRT (Lewington 2007). There are more than 130,000 patients on peritoneal dialysis (PD) worldwide, representing approximately 15% of the total world population requiring dialysis. It is anticipated that the number will increase during the next decade, especially in developing countries (de Freitas 2005). PD may be used not only in an immediate crisis, but also in stable disease situations to alleviate patients' suffering. Acute PD aims to remove endogenous and exogenous toxins and maintain fluid, electrolyte and acid‐base equilibrium until kidney function returns (Cadwallader 2002). Patients with AKI need constant assessment while on PD, and adequacy should be judged in terms of clinical status, ultrafiltration achieved, and biochemical parameters, particularly urea, creatinine, and bicarbonate levels (Fischbach 2002). The choice of PD as therapy has always to be individualised, balancing advantages against disadvantages.
Intermittent modes of RRT include haemodialysis (HD) and PD. HD may be the preferred mode of treatment in more stable patients with adequate vascular access treated on renal units where specialist nurses are available (Strazdins 2004). PD has been shown to be an effective therapy in paediatric patients developing AKI after cardiac surgery, and for treating adults with AKI (Lewington 2007). The choice of dialysis therapy for AKI depends upon the clinical circumstances, patient location, and expertise available. PD has generally been considered the preferred therapy if there is isolated failure of the kidneys in children, such as haemolytic uraemic syndrome (HUS). It is regarded as a simpler technique that is universally available (Strazdins 2004). However, haemofiltration (HF) and haemodiafiltration (HDF) are increasing in popularity in paediatric ICUs where the facilities to perform HD may not be available. The main advantage of PD is that it is continuous therapy that requires neither anticoagulation nor vascular access, and the technique can be used in haemodynamically unstable patients (Flynn 2001).
For more than 15 years, PD fluids containing lactate as the buffer have been used. However lactate may compromise local cell functions independently of pH by affecting the cellular REDOX state and reducing cellular energy sources. The exact mechanism responsible for these structural changes is unclear. Conventional PD fluids with glucose as the osmotic agent and, more importantly, the glucose degradation products (GDPs) generated during the heat sterilisation of these solutions seem to be responsible for inducing many of these changes in the peritoneum (Krishnan 2005). Toxic GDPs are formed during heat sterilisation of conventional PD solutions and these may be directly involved in the pathogenesis of peritoneal hypervascularisation and fibrosis, by stimulating local vascular endothelial growth factor (VEGF) and transforming growth factor‐beta (TGF‐β) synthesis and release (Haas 2003). GDPs, in addition to causing structural and functional alterations of the peritoneal cells, are also a leading cause of advanced glycation end‐products (AGE) production. Newer solutions have been recently introduced to minimise the affect on the peritoneal membrane. These include fluids containing bicarbonate alone or a mixture of bicarbonate and lactate as the source of the buffer (MacKenzie 1998). Bicarbonate buffering has recently become possible as a result of the development of a two‐chambered bag system to separate calcium and magnesium from bicarbonate, which have been shown to improve peritoneal cell function (Dratwa 2003). The use of bicarbonate buffer could confer specific advantages among young children undergoing PD (Nash 1977); and improved correction of metabolic acidosis with bicarbonate‐buffered PD fluid has been suggested for adult patients undergoing continuous ambulatory PD (CAPD) (Feriani 1998).
Increasing evidence from clinical studies in adults and children suggests that the new PD fluids may allow for better long‐term preservation of peritoneal morphology and function. These newer solutions have been shown to have several local and systemic advantages over the conventional PD solutions. However, what kind of buffers is more favourable for clinical (e.g. neutral‐pH, low‐GDP fluids) is still unclear (Nau 2004). As a consequence, there is no consensus regarding the preferred acute PD solutions for AKI.
Objectives
This review aimed to look at the benefits and harms of bicarbonate with low GDP versus lactate (conventional) solutions in acute PD for AKI.
Methods
Criteria for considering studies for this review
Types of studies
All randomised controlled trials (RCTs) and quasi‐RCTs (RCTs in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) looking at the efficiency and safety of bicarbonate versus lactate or bicarbonate/lactate solutions for patients with AKI were included. We included the first period of randomised cross‐over studies.
Types of participants
Inclusion criteria
-
Patients with AKI, regardless of sex or age, as estimated by either SCr, creatinine clearance (CrCl) or blood urea nitrogen (BUN).
-
AKI, for the purpose of this review, was defined as an increase in SCr > 0.5 mg/dL (44 mmol/L) over the baseline value, an increase of more than 50% over the baseline value, a reduction in the calculated CrCl of more than 50%, or a decrease in kidney function that results in the need for RRT (Moore 1998; Solomon 1994; Zanardo 1994).
Exclusion criteria
-
Patients with circumscribed peritonitis, infection of the abdominal wall, extensive internal organ adhesion or pregnancy.
Types of interventions
-
Studies comparing bicarbonate versus lactate solutions for acute PD.
-
Studies using any concentration of lactate or bicarbonate.
-
Studies using any concentration of sodium, calcium, magnesium, chloride or osmotic agent (glucose, glucose polymer or amino acid mixture).
-
Studies using acidic or neutral pH fluids.
Types of outcome measures
Primary outcomes
-
Mortality
-
Resolution of AKI (time taken)
-
Length of stay: ICU, hospitalisation
-
Quality of life
Secondary outcomes
-
Blood pH
-
Serum bicarbonate (mmol/L)
-
Blood lactate (mmol/L)
-
CrCl (mL/min)
-
Adverse events
-
episodes of relapsing peritonitis
-
episodes of acute fluid overload
-
aggravated hypertension
-
number of patients with dialysis‐associated amyloidosis
-
symptomatic hypotension or hypotension requiring intervention
-
headache
-
nausea and vomiting
-
pruritus
-
Search methods for identification of studies
Electronic searches
We searched the Cochrane Renal Group's Specialised Register through contact with the Trials' Search Co‐ordinator using search terms relevant to this review. The Cochrane Renal Group’s Specialised Register contains studies identified from the following sources.
-
Monthly searches of the Cochrane Central Register of Controlled Trials CENTRAL
-
Weekly searches of MEDLINE OVID SP
-
Handsearching of renal‐related journals & the proceedings of major renal conferences
-
Searching of the current year of EMBASE OVID SP
-
Weekly current awareness alerts for selected renal‐journals
-
Searches of the International Clinical Trials Register (ICTRP) Search Portal & ClinicalTrials.gov
Studies contained in the Specialised Register are identified through search strategies for CENTRAL, MEDLINE, and EMBASE based on the scope of the Cochrane Renal Group. Details of these strategies as well as a list of handsearched journals, conference proceedings and current awareness alerts are available in the 'Specialised Register' section of information about the Cochrane Renal Group.
See Appendix 1 for search terms.
Searching other resources
-
Reference lists of clinical practice guidelines, review articles and relevant studies.
-
Letters seeking information about unpublished or incomplete studies to investigators known to be involved in previous studies.
Data collection and analysis
Selection of studies
We used the search strategy described to obtain titles and abstracts of studies that might be relevant to the review. Two authors screened the titles and abstracts independently, and discarded any studies that were not applicable, after checking them for any relevant data or information on additional studies. The same authors independently retrieved the full text of these studies to determine which studies satisfied the inclusion criteria.
Data extraction and management
The same authors carried out data extraction independently, using standard data extraction forms. We planned to translate any studies reported in non‐English language journals before assessment. Where more than one publication of one study existed, we planned to group reports together and use the most recent or most complete dataset. We planned to highlight any discrepancy between published versions
Assessment of risk of bias in included studies
The following items were independently assessed by two authors using the risk of bias assessment tool (Higgins 2011) (see Appendix 2).
-
Was there adequate sequence generation (selection bias)?
-
Was allocation adequately concealed (selection bias)?
-
Was knowledge of the allocated interventions adequately prevented during the study (detection bias)?
-
Participants and personnel
-
Outcome assessors
-
-
Were incomplete outcome data adequately addressed (attrition bias)?
-
Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?
-
Was the study apparently free of other problems that could put it at a risk of bias?
Measures of treatment effect
For dichotomous outcomes (mortality, adverse events) we have expressed the results as relative risk (RR) with 95% confidence intervals (CI). Where continuous scales of measurement were used to assess the effects of treatment (blood pressure (BP), SCr, bicarbonate levels, blood pH, serum lactate level), we have used the mean difference (MD), or the standardised mean difference (SMD) if different scales had been used.
Dealing with missing data
We planned to request further information as required from the original authors. We planned to include any relevant information obtained in this manner in the review. We resolved any disagreements in consultation with a third author.
Assessment of heterogeneity
We planned to analyse heterogeneity using a Chi² test on N‐1 degrees of freedom, with an alpha of 0.05 used for statistical significance and with the I² test (Higgins 2003). I² values of 25%, 50% and 75% correspond to low, medium and high levels of heterogeneity.
Data synthesis
We planned to pool data using the random‐effects model but also planned to carry out the analyses using the fixed‐effect model to ensure robustness of the model chosen and susceptibility to outliers.
Subgroup analysis and investigation of heterogeneity
We planned to use subgroup analysis to explore possible sources of heterogeneity (e.g. participants, interventions and study quality). Heterogeneity among participants could be related to age and renal pathology. Heterogeneity in treatments could be related to prior agent(s) used and the agent, dose and duration of therapy. We planned to tabulated and assess adverse effects with descriptive techniques, as they were likely to be different for the various agents used. Where possible, we had planned to calculate the risk difference (RD) with 95% CI for each adverse effect, either compared to no treatment or to another age.
Results
Description of studies
Results of the search
The search strategy identified 167 abstracts of potential studies (Figure 1). After screening, we assessed the full text articles of six studies. One study met our inclusion criteria (Thongboomkerd 2001), and five studies were excluded (Cancarini 1998; Fusshoeller 2004; Schmitt 2002; Leblanc 1995; Nau 2004).
Study flow diagram.
Included studies
-
Thongboomkerd 2001 reported on kidney failure patients, who were indicated for dialysis and needed acute PD, were classified as shock and non‐shock group. These two groups were then randomised to receive either bicarbonate or lactate solution. Twenty patients were enrolled in this study (five in each subgroup).
Excluded studies
-
Cancarini 1998 was excluded because no data were available for extracting.
-
Fusshoeller 2004 was excluded because the patients had CKD. This was a study of 14 chronic PD patients assigned to receive automated PD with either conventional or bicarbonate/lactate based fluids.
-
Leblanc 1995 was excluded because the study was about HD, not PD.
-
Nau 2004 was excluded as the intervention was automated PD, not acute PD.
-
Schmitt 2002 was excluded because the patients had chronic kidney disease (CKD). This was a randomised crossover study among children undergoing automated PD. After a four‐week run‐in period the children underwent two consecutive 12‐week study periods, in which automated PD was performed with a neutral‐pH PD fluid containing 34 mM bicarbonate or a conventional PD fluid with 35 mM lactate The two treatment phases were separated by a four‐week washout period.
Risk of bias in included studies
It was unclear whether allocation was concealed; whether analysis was carried out on an intention‐to‐treat basis; and how complete the follow‐up of study participants was. In terms of blinding, it was unclear whether investigators, outcome assessors or data analysts were blinded to treatment allocation. Given the nature of the interventions, it is unlikely that participants could have been blinded to treatment allocation. Some outcome data have not been reported, such as resolution of AKI (time taken); length of stay: ICU, hospitalisations; quality of life. It is unclear whether selective reporting is present, which can lead to some bias.
Effects of interventions
Primary outcome
Mortality
Thongboomkerd 2001 reported the results for mortality. The RR for mortality with administration of bicarbonate versus lactate appears in Analysis 1.1 (RR 0.50, 95% CI 0.06 to 3.91).
Secondary outcomes
Blood pH
Thongboomkerd 2001 reported the results for blood pH (Analysis 1.2).
-
Shock group (MD 0.12, 95% CI 0.06 to 0.15).
-
Non‐shock group (MD 0.02, 95% CI ‐0.02 to 0.06).
Blood lactate (mmol/L)
Thongboomkerd 2001 reported the results for blood lactate (Analysis 1.4).
-
Shock group (MD ‐1.60 mmol/L, 95% CI ‐2.04 to ‐1.16).
-
Non‐shock group (MD ‐0.06 mmol/L, 95% CI ‐0.85 to ‐0.35).
Serum bicarbonate (mmol/L)
Thongboomkerd 2001 reported the results for serum bicarbonate (Analysis 1.3)
-
Shock group (MD 5.00 mmol/L, 95% CI 3.26 to 6.74).
-
Non‐shock group (MD 1.10 mmol/L, 95% CI ‐0.27 to 2.47).
We were unable to analyse data as planned for the following primary outcomes.
-
Resolution of AKI (time taken)
-
Length of stay: ICU, hospitalisation
-
Quality of life
-
Adverse events.
We were unable to analyse data as planned for the following secondary outcomes and adverse events.
-
CrCl (mL/min)
-
Number of patients with dialysis‐associated amyloidosis
-
Symptomatic hypotension or hypotension requiring intervention
-
Headache
-
Nausea and vomiting
-
Pruritus.
Discussion
Summary of main results
There is no strong evidence to support a clinical advantages for either bicarbonate or lactate solution in patients with AKI who require acute PD. Our systematic review has demonstrated the following.
-
In shock patients there were significant differences between bicarbonate and lactate in blood pH, blood lactate and serum bicarbonate.
-
In non‐shock patients there were significant differences between bicarbonate and lactate in blood lactate.
-
There were no significant different between bicarbonate and lactate for mortality for either shock or non‐shock patients.
-
With the exception of mortality, we were not able to assess the main primary outcomes of this review.
Overall completeness and applicability of evidence
There was one small RCT identified comparing bicarbonate and lactate solutions for acute PD (Thongboomkerd 2001) identified after checking approximately 164 abstracts identified in the literature search.
Bicarbonate and lactate solutions are still widely used in many areas. Their application is mostly for AKI patients with hypercatabolic states and intractable metabolic and fluid derangement who cannot be treated with HD. They are also used for continuous renal replacement therapy patients with the same indications. PD with bicarbonate solution is strongly associated with better outcomes in peritoneal macrophage, monocyte, mesothelial, and fibroblast cell functions (Fischer 1995; Jörres 1998). In cases of tumour necrotic factor, a generation of peritoneal macrophage was significantly increased by bicarbonate or bicarbonate combined with lactate solution, compared to lactate solution alone (MacKenzie 1998). Biocompatibility was better with bicarbonate solution (due to its more frequent exchanges and longer contact times with fresh dialysate). Better blood pH and serum bicarbonate levels might be the correct way to explain the improvement in non‐specific systemic host defence and haemodynamics in the bicarbonate subgroup. However, several factors such as antibiotics, inotropic agents and difference in dialysate sodium should be considered. The pH of the dialysis fluid might be particularly relevant for automated PD, in which frequent short cycles continuously expose the peritoneal membrane to a cytotoxic acidic milieu. Blood lactate levels in the lactate subgroup were significantly higher for both shock and non‐shock patients. This implied that, despite the normal conversion of lactate to bicarbonate in the tissues, higher lactate loading caused a greater accumulation of lactate in the blood circulation. It is important to note that, whilst there were no statistically significant differences between bicarbonate and lactate in blood bicarbonate levels, the 95% CIs were wide enough to suggest that clinically important differences may indeed exist.
Quality of the evidence
The quality of the evidence was not high, because methodology, allocation concealment, and blinding were not reported in detail. We were not able to obtain further data from the study authors.
Potential biases in the review process
There were several limitations to our systematic review that are worthy of mention. First, the results are based on one small study, which reduces the evaluation power and increases the possibility of publication bias. Second, it is possible that we have not identified all the relevant studies from computerised searching. We acknowledge that our results may have been affected by the quality of study, blinding process, and other procedures. Finally, this review may suffer from reporting bias, as we found no measurable data on most of the main outcomes and adverse events. We also cannot accurately determine publication bias, as it is inappropriate to undertake a funnel plot with only one included study. Should we identify any additional studies through searches of the grey literature, we will include these in future updates.
Agreements and disagreements with other studies or reviews
The role of bicarbonate solution has been studied extensively in CAPD for CKD (Feriani 1998; Passlick 1996), and these studies supplied some references relevant to acute PD. PD with bicarbonate solution is strongly associated with better outcomes in peritoneal macrophage, monocot, mesothelial, and fibroblast cell functions (Jörres 1998; Pedersen 1994). The in vivo study by Feriani 1998 showed a significant improvement of metabolic acidosis and plasma bicarbonate levels during treatment by bicarbonate solution, while no significant change was observed in acrostic patients treated with the conventional lactate‐buffered solution.
Comparison 1 Bicarbonate versus lactate solutions for acute peritoneal dialysis, Outcome 1 Mortality.
Comparison 1 Bicarbonate versus lactate solutions for acute peritoneal dialysis, Outcome 2 Blood pH.
Comparison 1 Bicarbonate versus lactate solutions for acute peritoneal dialysis, Outcome 3 Serum bicarbonate (mmol/L).
Comparison 1 Bicarbonate versus lactate solutions for acute peritoneal dialysis, Outcome 4 Blood lactate (mmol/L).
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Mortality Show forest plot | 1 | Risk Ratio (M‐H, Random, 95% CI) | Totals not selected | |
| 1.1 Shock group | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 1.2 Non‐shock group | 1 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 2 Blood pH Show forest plot | 1 | Mean Difference (IV, Random, 95% CI) | Totals not selected | |
| 2.1 Shock group | 1 | Mean Difference (IV, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 2.2 Non‐shock group | 1 | Mean Difference (IV, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 3 Serum bicarbonate (mmol/L) Show forest plot | 1 | Mean Difference (IV, Random, 95% CI) | Totals not selected | |
| 3.1 Shock group | 1 | Mean Difference (IV, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 3.2 Non‐shock group | 1 | Mean Difference (IV, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 4 Blood lactate (mmol/L) Show forest plot | 1 | Mean Difference (IV, Random, 95% CI) | Totals not selected | |
| 4.1 Shock group | 1 | Mean Difference (IV, Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 4.2 Non‐shock group | 1 | Mean Difference (IV, Random, 95% CI) | 0.0 [0.0, 0.0] | |
