Visual Abstract
Keywords: pediatrics, peritoneal dialysis, plasma, solutions, uremia
Abstract
Background and objectives
Residual native kidney function confers health benefits in patients on dialysis. It can facilitate control of extracellular volume and inorganic ion concentrations. Residual kidney function can also limit the accumulation of uremic solutes. This study assessed whether lower plasma concentrations of uremic solutes were associated with residual kidney function in pediatric patients on peritoneal dialysis.
Design, setting, participants, & measurements
Samples were analyzed from 29 pediatric patients on peritoneal dialysis, including 13 without residual kidney function and ten with residual kidney function. Metabolomic analysis by untargeted mass spectrometry compared plasma solute levels in patients with and without residual kidney function. Dialytic and residual clearances of selected solutes were also measured by assays using chemical standards.
Results
Metabolomic analysis showed that plasma levels of 256 uremic solutes in patients with residual kidney function averaged 64% (interquartile range, 51%–81%) of the values in patients without residual kidney function who had similar total Kt/Vurea. The plasma levels were significantly lower for 59 of the 256 solutes in the patients with residual kidney function and significantly higher for none. Assays using chemical standards showed that residual kidney function provides a higher portion of the total clearance for nonurea solutes than it does for urea.
Conclusions
Concentrations of many uremic solutes are lower in patients on peritoneal dialysis with residual kidney function than in those without residual kidney function receiving similar treatment as assessed by Kt/Vurea.
Introduction
The presence of residual kidney function is associated with health benefits in patients on peritoneal dialysis and hemodialysis (1–4). Residual kidney function facilitates control of extracellular fluid volume and electrolyte concentrations. Residual kidney function also removes uremic solutes. This raises the question of how well solutes are removed by residual kidney function as compared with dialysis. Present guidelines assess dialysis adequacy by measuring the clearance of urea relative to body water volume (5–7). In patients on peritoneal dialysis, the residual urea clearance is added directly to the dialytic urea clearance and divided by the total body water volume to yield the parameter Kt/Vurea. Measurement of the single solute urea thus serves to establish the relative value of residual kidney function and peritoneal dialysis. Many solutes are, however, cleared more rapidly than urea by residual kidney function, whereas no solutes are known to be cleared more rapidly than urea by dialysis. Thus, plasma levels for many solutes could be lower in patients with residual kidney function than in patients without residual kidney function who are receiving equal treatment as assessed by Kt/Vurea. This study tested this hypothesis by metabolomic analysis of a large number of small uremic solutes. The metabolomic analysis was supplemented by assays of selected solutes using more precise methods. We also looked for a solute whose plasma level would allow estimation of residual kidney function without the need for urine collection. Metabolomic analysis of plasma from a prior study of pediatric patients on hemodialysis allowed us to make a preliminary comparison of the control of plasma solute levels using the two forms of treatment.
Materials and Methods
Patient Enrollment
Thirty-eight pediatric patients maintained on peritoneal dialysis at four centers were enrolled in the study. Written informed consent was obtained for all patients. The study was approved by the institutional review boards at each center and was conducted in accordance with the Declaration of Helsinki. Samples were collected from 32 patients (Supplemental Figure 1). Six patients reported to produce some urine were excluded because timed urine samples could not be obtained either for this study or for routine clinical evaluation. One patient was excluded from the analysis because chemical analysis revealed a residual urea clearance of >15 ml/min, another was excluded because chemical analysis suggested contamination of the dialysate, and a third was excluded because samples were mistakenly collected during a peritoneal equilibration test rather than a regular daily treatment. The remaining 23 patients were divided into a group of 13 patients considered to have no residual kidney function and a group of ten patients with residual kidney function from whom a timed urine sample was obtained.
Chemical Analysis
Samples of spent dialysate and plasma were collected at routine clinic visits along with a timed urine sample when possible. Plasma ultrafiltrate was prepared using Nanosep 30K Omega separators as previously described (8). Relative concentrations of uremic solutes in the plasma and plasma ultrafiltrate were assessed by metabolomic analysis. This analysis was performed at Metabolon Inc. using an established platform that uses liquid chromatography/mass spectrometry to identify solutes from among a chemical library including >4000 confirmed metabolites (9,10). Because we did not collect blood samples from pediatric subjects without kidney disease, solutes were identified as uremic on the basis of comparison of plasma from adult patients on hemodialysis and adults without known kidney disease and as summarized in Supplemental Table 1. Concentrations of selected uremic solutes were also assayed using chemical standards. Urea was measured by a commercial enzymatic assay, and creatinine was measured by HPLC as previously described (8). The modified nucleoside pseudouridine was measured by stable isotope dilution liquid chromatography mass spectrometry (LC/MS/MS) as previously described (11). The organic acids phenylacetylglutamine, hippurate, indoxyl sulfate, and p-cresol sulfate were also measured using stable isotope dilution LC/MS/MS as previously described (8). Quality control samples containing solutes at concentrations observed in patients with kidney failure are analyzed with each LC/MS/MS run. The ratios of the measured to expected concentrations of the quality control samples averaged 0.97±0.04, 1.01±0.09, 1.04±0.11, 1.06±0.08, and 1.02±0.11 for pseudouridine, phenylacetylglutamine, hippurate, indoxyl sulfate, and p-cresol sulfate, respectively, in LC/MS/MS runs over the period during which these studies were conducted. Metabolon Inc. reported a median relative SD of 7% for metabolites identified in replicates of pooled human plasma analyzed concurrently with samples run for this study.
Calculations
The free fraction of the bound solutes was calculated as the plasma ultrafiltrate concentration divided by the total plasma concentration. Free fractions could not be calculated for six solutes because peak areas for these solutes were detected in the patients' plasma but not in their plasma ultrafiltrate. For solutes measured with chemical standards, the dialytic removal was calculated as the dialytic concentration multiplied by the daily volume of spent dialysate, and the dialytic clearance was calculated as the removal rate divided by the plasma concentration. The residual removal rate was calculated as the urine concentration multiplied by the urine volume corrected for the duration of the collection (12–24 hours), and the residual clearance was calculated as the residual removal rate divided by the plasma concentration. Body surface area was calculated using the formula of Gehan and George (12). Kt/Vurea was calculated using the body water formula of Morgenstern et al. (13)
Statistical Analyses
Demographic and dialysis prescription data in the patients with and without residual kidney function were compared using the Mann–Whitney U Test. Plasma concentrations and production rates of solutes measured using chemical standards were compared using generalized linear regression adjusted for age, sex, and body surface area. Plasma concentrations and solute production rates were log transformed when necessary to obtain a normal distribution. The relations of solute levels and their residual clearances to residual kidney function as represented by residual urea clearance were assessed by linear regression. Statistical analyses were performed using SPSS v27. Metabolomic peak areas representing plasma and plasma ultrafiltrate levels in the patients with and without residual kidney function were compared by the unpaired t test, and false discovery rates (q values) were calculated from the respective P values using software from qvalue.princeton.edu.
Results
Characteristics of the study subjects and their dialysis treatments are summarized in Table 1. The patients were all maintained on automated nocturnal peritoneal dialysis, and 21 of the 23 also were prescribed a daytime dwell. Patients with residual kidney function were prescribed shorter dialysis treatment times than those without residual kidney function. The number of cycles was significantly lower only in those patients with residual kidney function. Total Kt/Vurea exceeded the minimum of 1.7 recommended by various guidelines (5–7). As assessed by Kt/Vurea, the patients without residual kidney function received treatment equal to the patients with residual kidney function. The dialytic component of Kt/Vurea was, however, lower in the patients with residual kidney function.
Table 1.
Demographics and dialysis prescription parameters
| Variable | No Residual Kidney Function | Residual Kidney Function |
| n | 13 | 10 |
| Age, yr | 6±5 | 12±4a |
| Sex, girls/boys | 6/7 | 7/3 |
| Weight, kg | 20±10 | 43±24a |
| BSA, m2 | 0.8±0.3 | 1.3±0.4a |
| Vintage, yr | 1.9±1.3 | 1.1±1.2 |
| Urine volume, ml/min | — | 0.52±0.44 |
| Treatment time, h/d | 10.7±1.3 | 8.6±1.3a |
| No. of cycles | 9±2 | 7±2a |
| Ultrafiltration, ml/d | 382 (297–1089) | 567 (304–910) |
| Kt/V dialytic per week | 2.55±0.51 | 1.43±0.34a |
| Kt/V residual per week | — | 0.88 + 0.41 |
| Kt/V total per week | 2.55±0.51 | 2.31±0.38 |
Values are mean ± SD or median (interquartile range). The duration of the urine collection averaged 20±5 hours, with a range of 12–24 hours. BSA, body surface area.
P=0.05.
Metabolomic analysis identified a total of 256 named uremic solutes that could be detected in the plasma of all of the patients without residual kidney function and with residual kidney function as summarized in Supplemental Table 2. Comparison of the patients with and without residual kidney function revealed a marked disparity in average plasma levels of uremic solutes as illustrated in Figure 1. The plasma levels of individual uremic solutes in patients with residual kidney function averaged 64% (interquartile range [IQR], 51%–81%) of their levels in patients without residual kidney function. Using a false discovery rate of 5%, plasma levels were significantly lower for 59 of the 256 uremic solutes in the patients with residual kidney function and significantly higher for none of them, as summarized in Table 2. Comparison of solute levels in the plasma ultrafiltrate yielded similar results as summarized in Supplemental Table 2. The plasma ultrafiltrate levels of individual uremic solutes in patients with residual kidney function averaged 64% (IQR, 50%–78%) of their levels in patients without residual kidney function as depicted in Supplemental Figure 2. Again, using a false discovery rate of 5%, plasma ultrafiltrate levels were significantly lower for 86 solutes and significantly higher for none, as further summarized in Supplemental Table 2. All of the 59 uremic solutes identified as having significantly lower levels in the plasma of patients with residual kidney function were also identified as having significantly lower levels in the plasma ultrafiltrate.
Figure 1.
Histogram showing lower average plasma solute concentrations (denoted as [solute]) for 256 uremic solutes in patients with residual kidney function than in patients without residual kidney function. The log(10)-transformed concentration ratios for individual solutes are presented on the horizontal axis, and the numbers of solutes in the various concentration ratio ranges are presented on the vertical axis. The log(10) concentration ratio is negative when the concentration is lower in the patients with residual kidney function than in the patients with no residual kidney and positive when the plasma concentration is higher in the patients with residual kidney function than in the patients with no residual kidney function. Solute concentrations for the 256 uremic solutes have been presumed proportional to their mass spectrometric peak areas on metabolomic analysis.
Table 2.
Solutes found at lower levels in patients with residual kidney function than in patients without residual kidney function as identified by metabolomics
| Solute | With Residual Kidney Function-Without Residual Kidney Function Plasma |
| Picolinoylglycine | 0.18 |
| Methyl-4-hydroxybenzoate sulfate | 0.19 |
| Methylsuccinoylcarnitine | 0.30 |
| 4-Hydroxyhippurate | 0.32 |
| Tigloylglycine | 0.35 |
| Hydantoin-5-propionate | 0.35 |
| 1-Methylguanidine | 0.37 |
| Phenylacetylglycine | 0.38 |
| 3-Methylcrotonylglycine | 0.38 |
| 2-Hydroxyadipate | 0.38 |
| 2-Methylbutyrylglycine | 0.42 |
| Kynurenate | 0.43 |
| 4-Hydroxyphenylacetylglutamine | 0.45 |
| Gluconate | 0.46 |
| Vanillactate | 0.47 |
| Azelate | 0.48 |
| Orotidine | 0.48 |
| N-acetyltaurine | 0.48 |
| Quinolinate | 0.50 |
| Lyxonate | 0.50 |
| Carboxyethyl-gamma-aminobutyric acid | 0.51 |
| Glucuronate | 0.52 |
| N-acetyl-aspartyl-glutamate | 0.52 |
| Succinylcarnitine | 0.53 |
| 1-Methyl-4-imidazoleacetate | 0.54 |
| N6-succinyladenosine | 0.54 |
| 4-acetamidobutanoate | 0.54 |
| 2-O-methylascorbic acid | 0.54 |
| 2-methylmalonylcarnitine | 0.54 |
| N6-carbamoylthreonyladenosine | 0.55 |
| Arabonate/xylonate | 0.55 |
| 1-Methylurate | 0.56 |
| Pseudouridine | 0.56 |
| 5-(Galactosylhydroxy)-L-lysine | 0.57 |
| (S)-a-amino-omega-caprolactam | 0.57 |
| N-acetylneuraminate | 0.57 |
| N1-methylinosine | 0.57 |
| N-acetylhistidine | 0.58 |
| Vanillylmandelate | 0.60 |
| 3-(3-Amino-3-carboxypropyl)uridinea | 0.61 |
| N-acetyl-β-alanine | 0.62 |
| 3-Hydroxy-3-methylglutarate | 0.62 |
| 5,6-Dihydrouridine | 0.62 |
| O-sulfo-L-tyrosine | 0.64 |
| Ribonate | 0.64 |
| 3-Methoxytyrosine | 0.64 |
| C-glycosyltryptophan | 0.65 |
| N-acetylisoleucine | 0.68 |
| N2,N2-dimethylguanosine | 0.70 |
| 2,3-Dihydroxy-5-methylthio-4-pentenoatea | 0.71 |
| Prolylglycine | 0.71 |
| Mannonatea | 0.74 |
| N-acetylserine | 0.74 |
| Hydroxyasparagineb | 0.75 |
| N-acetylvaline | 0.75 |
| N-acetylthreonine | 0.75 |
| S-adenosylhomocysteine | 0.77 |
| N-formylmethionine | 0.78 |
| Erythronatea | 0.81 |
With residual kidney function-without residual kidney function refers to the ratio of the average peak areas for individual solutes in the plasma of patients with residual kidney function compared with those without residual kidney function. The 59 solutes listed had plasma peak areas that were significantly lower in the patients with residual kidney function than in the patients without residual kidney function , with q value <0.05.
Indicates compound was not confirmed based on an authentic chemical standard but confidence in its identity is high.
Indicates that a compound may be one of several structural isomers that cannot be distinguished by mass spectrometry.
More precise analysis of selected solute levels with chemical standards provided results similar to those obtained by the metabolomic analysis. Concentration ratios obtained by analysis with standards were similar to those obtained by metabolomic analysis as summarized in Supplemental Table 3, suggesting that metabolomic analysis provided an accurate view of solute levels. Plasma levels for individual solutes measured with chemical standards are summarized in Table 3. The solute whose plasma level was proportionally lowest in the patients with residual kidney function was pseudouridine. The plasma levels of pseudouridine in individual patients were, however, not closely correlated with the amount of residual kidney function, as illustrated in Figure 2. Plasma levels of other solutes measured with chemical standards also failed to reflect the extent of residual kidney function as illustrated in Supplemental Figure 3. A search among uremic solutes included in the untargeted metabolomic analysis also failed to identify a solute whose plasma level reflected the extent of residual kidney function, as further summarized in Supplemental Table 2.
Table 3.
Plasma solute levels
| Solute | No Residual Kidney Function | Residual Kidney Function |
| Urea N | 61±25 | 52±16 |
| Creatinine | 8.9±3.9 | 7.4±3.0a |
| Pseudouridine | 2.6±1.3 | 1.1±0.8a |
| Phenylacetylglutamine | 5.2±4.0 | 2.7±3.8 |
| Hippurate | 4.9±3.1 | 2.2±3.3 |
| Indoxyl sulfate | 3.1±1.6 | 2.1±1.6 |
| p-Cresol sulfate | 2.4±2.0 | 2.9±1.8 |
Values are in milligrams per deciliter (mean ± SD).
P=0.05 versus no residual kidney function adjusted for age, sex, and body surface area.
Figure 2.
The relation of plasma levels of pseudouridine (PU) to residual urea clearance (Kru) in pediatric patients on peritoneal dialysis. PU levels were on average lower in patients with residual kidney function but could not be used to estimate the level of residual kidney function in individual patients (r2=0.40).
Assays of dialysate and urine using chemical standards allowed us to examine the solute clearances and generation rates that determine solute levels in the plasma. As summarized in Table 4, the dialytic clearances of the nonurea solutes were less than the dialytic clearance of urea. The residual clearances of the nonurea solutes varied in relation to the residual clearance of urea. The ratios of residual clearance to dialytic clearances for the nonurea solutes were compared with the ratios of residual clearance to dialytic clearance for urea to test how well residual kidney function cleared the nonurea solutes compared with urea. These clearance ratios showed that residual kidney function afforded high clearances relative to urea for the nonurea solutes. The values obtained for each solute varied widely, however, among individual patients. Solute generation rates were also highly variable among individual patients, as summarized in Table 5.
Table 4.
Clearance ratios
| Solute | No Residual Kidney Function | Residual Kidney Function | ||
| KdSolute-Kdurea | KdSolute-Kdurea | KrSolute-Krurea | KrSolute-Krurea/ KdSolute-Kdurea | |
| Creatinine | 0.63±0.16 | 0.57±0.10 | 1.9±0.5 | 3.4±1.0 |
| Pseudouridine | 0.31±0.14 | 0.22±0.09 | 1.3±0.4 | 7.2±3.6 |
| Phenylacetylglutamine | 0.44±0.14 | 0.46±0.26 | 5.6±7.0 | 11.0±9.1 |
| Hippurate | 0.29±0.07 | 0.31±0.15 | 4.6±3.4 | 15.4±6.8 |
| Indoxyl sulfate | 0.07±0.03 | 0.05±0.01 | 0.7±0.5 | 14.1±11.2 |
| p-Cresol sulfate | 0.06±0.03 | 0.04±0.01 | 0.3±0.2 | 7.0±5.1 |
Values are mean ± SD. KdSolute-Kdurea is the ratio of the dialytic clearance of each solute to the dialytic clearance of urea. KrSolute-Krurea is the ratio of the residual kidney clearance of each solute to the residual kidney clearance of urea. KrSolute-Krurea divided by KdSolute-Kdurea provides a measure of the degree to which a solute has a higher clearance relative to urea by residual kidney function than peritoneal dialysis.
Table 5.
Solute generation rates
| Solute | No Residual Kidney Function | Residual Kidney Function |
| Urea N | 5200±1913 | 4767±1723 |
| Creatinine | 473±213 | 700±315 |
| Pseudouridine | 58±11 | 53±17 |
| Phenylacetylglutamine | 171±118 | 191±92 |
| Hippurate | 122±81 | 185±117 |
| Indoxyl sulfate | 20±14 | 39±24 |
| p-Cresol sulfate | 12±11 | 34±23 |
Values are milligrams per 1.73 m2 per day (mean ± SD). Rates of solute generation were assumed equal to the removal in the dialysate combined with the removal in the urine when present.
The availability of samples saved from a previous study allowed us to make a preliminary comparison of plasma solute levels in patients without residual kidney function treated by hemodialysis and by peritoneal dialysis (11). Clinical characteristics of the patients included in this comparison are summarized in Supplemental Table 4. For the solutes examined, the plasma levels in the patients on peritoneal dialysis averaged 0.84 (IQR, 0.71–1.03) times the midweek pretreatment plasma levels in the patients on hemodialysis as depicted in Supplemental Figure 4 and summarized in Supplemental Table 5. By far, the largest difference was observed for o-cresol sulfate, for which average plasma levels were 30 times higher in the patients on peritoneal dialysis than in the patients on hemodialysis. Levels for other plasma solutes in the patients on peritoneal dialysis ranged from 0.27 to 2.7 times the levels in the patients on hemodialysis.
Discussion
Residual kidney function contributes importantly to the welfare of patients on dialysis (1–4). Its value has been demonstrated most extensively in patients on peritoneal dialysis. A landmark reanalysis of the Canada-USA study showed that the better survival achieved with higher combined peritoneal and residual kidney clearances was attributable entirely to higher residual kidney clearances (14). Subsequent studies showed that residual kidney function was associated with improved physical function and other benefits (15,16). Residual kidney function tends to be preserved better in patients on peritoneal dialysis than those on hemodialysis. It may, therefore, be particularly important in children, for whom peritoneal dialysis is the predominant form of treatment (17).
Residual kidney function may benefit patients by several means. It facilitates control of plasma electrolyte levels and reduces the requirement for dialytic volume removal. This study showed that it also lowers the plasma levels of a large number of uremic solutes. The levels of the small uremic solutes included in our metabolomic analysis averaged about 30% lower in patients with residual kidney function than in patients without residual kidney function whose treatment provided a similar total Kt/V. Additionally, the difference in plasma levels attained statistical significance for a large number of individual solutes, even though the number of subjects in our pediatric study was limited.
Lower plasma solute levels in patients with residual kidney function can be explained by differences between solute clearances by residual kidney function and dialysis. Peritoneal solute clearance decreases rapidly with increasing solute size (18–20). The lower peritoneal clearances of larger solutes may be approximately proportional to their reduced diffusivity in water (21). On this basis, a solute with the mass of 240 D would be predicted to have a peritoneal clearance about half that of urea. The effect of solute size on clearance by glomerular filtration is much less marked. Increasing size reduces the filtration rate by <10% for solutes with mass up to 10 kD (22). Another difference is that with residual kidney function, tubular secretion increases the clearance of some solutes above the GFR, whereas urea is partially reabsorbed. Secretion may be impaired as the kidney fails, but at least some solutes are still secreted by the residual kidney (23–26). The combined result of these effects is that residual kidney function provides a higher portion of the total clearance for nonurea solutes than for urea. Of note, a greater extent of secretory clearance has recently been associated with a lower symptom burden in adult patients on peritoneal dialysis (27).
Studies by Bammens et al. (28,29) initially described the disproportionate contribution of residual kidney function to the clearance of β2-microglobulin and selected small solutes in patients on peritoneal dialysis. Subsequent studies have focused on the contribution of residual kidney function to the clearance of low molecular weight proteins (30–32). The average plasma levels of low molecular weight proteins decline with increasing residual kidney function because residual kidney function clears them more rapidly than urea, whereas dialysis clears them more slowly (33–36). Measurement of their plasma levels has, however, not provided estimates of residual kidney function, which are accurate enough to adjust the dialysis prescription in individual patients.
This study tested whether plasma levels of a small uremic solute might provide a measure of residual kidney function. Such a solute must have two properties in addition to a higher clearance relative to urea by residual kidney function than peritoneal dialysis. It must be produced at a predictable rate and not cleared significantly by mechanisms other than residual kidney function and dialysis. In people without kidney disease, the modified nucleoside pseudouridine has been shown to have these properties (37–39). We recently found that pseudouridine production remains predictably related to body size in pediatric patients on hemodialysis (11). Struijk et al. (40) had earlier shown that pseudouridine is poorly cleared by peritoneal dialysis, suggesting that its removal in patients on peritoneal dialysis should depend heavily on residual kidney function. We, therefore, tested whether plasma pseudouridine levels would provide a useful index of the extent of residual kidney function. The results were disappointing. Plasma pseudouridine levels were significantly lower in patients with residual kidney function but were not closely correlated with the amount of residual kidney function in individual patients. The lack of correlation of plasma levels with residual kidney function was attributable largely to variability in the clearances of pseudouridine relative to urea by both peritoneal dialysis and the residual kidney. Pseudouridine production rates were also less closely related to body size in patients on peritoneal dialysis than we had previously found them to be in patients on hemodialysis.
We also assessed the clearances relative to urea of several solutes that are cleared by secretion in both the normal and residual kidneys (8,23). The solutes we examined were organic anions of gut microbial origin whose production is known to be highly variable among individuals. As expected, their clearances relative to urea were much greater by residual kidney function than by peritoneal dialysis. However, their clearances relative to urea by both peritoneal dialysis and residual kidney function were quite variable. Review of the untargeted analysis likewise failed to reveal a uremic solute whose plasma levels could reliably be used to assess residual kidney function. As with the solutes measured using chemical standards, we presume that variability in solute production causes variation in plasma levels independent of residual kidney function. Production of many solutes is likely affected by diet. Urea removal was similar in patients with and without residual kidney function in our study, suggesting that protein intake was also similar. We do not, however, have information on the type of protein consumed or on other dietary constituents. Solute production may be further varied by medication use, diseases responsible for kidney failure, and for colon-derived solutes, differences in the colon microbiome. It is possible that the variability in the production of individual solutes might be overcome by measurement of a panel of solutes, but identification of such a panel would require a larger study.
We took advantage of the availability of stored samples from a prior study to make a preliminary comparison of solute levels in pediatric patients on peritoneal dialysis and on hemodialysis. The most striking finding was the 30-fold higher average peak area for o-cresol sulfate in the plasma of pediatric patients on peritoneal dialysis. We hypothesize that this may reflect conversion of aromatic amino acids to o-cresol rather than p-cresol in patients on peritoneal dialysis. We previously found that although p-cresol sulfate is tightly plasma protein bound and poorly cleared by peritoneal dialysis, its plasma level is not higher in patients on peritoneal dialysis than in patients on hemodialysis (26). This study reproduced this finding, which may reflect reduced production of p-cresol sulfate in patients on peritoneal dialysis without residual kidney function.
In summary, this study showed that average levels of a large number of small uremic solutes are lower in patients on peritoneal dialysis with residual kidney function than in those without residual kidney function receiving equally adequate treatment as assessed by Kt/Vurea. The study has important limitations. The number of subjects was small, and the study groups were not evenly matched for age, sex, or size. The higher dialytic clearance in patients without residual kidney function may have affected solute removal in ways we cannot account for. We studied uremic solutes identified in adults using a single metabolomic platform. This metabolomic platform allowed us to examine a large number of uremic solutes but does not include all uremic solutes identified using other methods or reported in the literature (41–43). Overall, however, our results provide further evidence that total urea clearance is an imperfect measure of dialysis adequacy. This is consistent with the recent recommendation of the International Society of Peritoneal Dialysis that the dialysis prescription does not need to be adjusted to achieve a target Kt/Vurea (44). Better knowledge of which solutes are toxic could provide a superior index of treatment adequacy.
Disclosures
P. Brakeman received honoraria from the Hill Physicians Group for a lecture series in 2019; serves as a scientific advisor or member of Horizon Therapeutics USA, Inc.; and is a member of the Kidney Research Network Executive Committee. T.W. Meyer reports consultancy agreements with Baxter and Daiichi Sankyo; receiving research funding from Outset Medical; receiving honoraria from Baxter; and serving on the JASN and Kidney International editorial boards. T.W. Meyer has applied for a patent to improve the dialytic removal of uremic solutes that bind to plasma proteins. R. Sheth reports receiving research funding from Alexion Pharmaceuticals and Rockwell Pharmaceuticals and reports American Society of Pediatric Nephrology (no financial relationship) committee membership. T.L. Sirich has served as a consultant for Baxter. S.M. Sutherland does ad hoc consulting for the Gerson Lehrman Group and serves as a key opinion leader for CHF Solutions. S.M. Sutherland receives an honorarium for serving as Continuous Renal Replacement Therapy University faculty from the Acute Kidney Injury Critical Care Research Foundation, which receives an educational grant from Baxter Healthcare for running CRRT University. All remaining authors have nothing to disclose.
Funding
This work was supported by National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases awards R01 DK101674 (to T.W. Meyer) and R01 DK118426 (to T.L. Sirich). L.L. Ganesan was supported by a fellowship from the Stanford Child Health Research Institute and the American Society of Nephrology Ben Lipps Fellowship.
Supplementary Material
Acknowledgments
The authors thank the dialysis staff members of the four participating pediatric centers for help with recruiting patients and collecting samples.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
Supplemental Material
This article contains the following supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.01430121/-/DCSupplemental.
Supplemental Table 1. List of solutes previously identified as uremic.
Supplemental Table 2. Metabolomic comparison of plasma solute levels in patients on peritoneal dialysis without residual kidney function and with residual kidney function.
Supplemental Table 3. Comparison of the plasma solute concentration ratios obtained by metabolomic analysis and by quantitative analysis using chemical standards.
Supplemental Table 4. Demographics and dialysis prescription parameters of patients included in the preliminary analysis of plasma solute levels in children without residual kidney function maintained on peritoneal dialysis and on hemodialysis.
Supplemental Table 5. Metabolomic comparison of plasma solute levels in children without residual kidney function maintained on peritoneal dialysis and on hemodialysis.
Supplemental Figure 1. CONSORT flow diagram showing the enrollment of patients.
Supplemental Figure 2. Histogram showing lower average plasma ultrafiltrate levels of uremic solutes in patients with residual kidney function than in patients without residual kidney function.
Supplemental Figure 3. Association of residual kidney clearances and plasma levels of solutes for which absolute concentrations were measured with residual kidney clearances of urea.
Supplemental Figure 4. Histogram comparing average plasma levels of uremic solutes in children without residual kidney function maintained on peritoneal dialysis and on hemodialysis.
References
- 1.Perl J, Bargman JM: The importance of residual kidney function for patients on dialysis: A critical review. Am J Kidney Dis 53: 1068–1081, 2009 [DOI] [PubMed] [Google Scholar]
- 2.Cadnapaphornchai MA, Teitelbaum I: Strategies for the preservation of residual renal function in pediatric dialysis patients. Pediatr Nephrol 29: 825–836, quiz 832, 2014 [DOI] [PubMed] [Google Scholar]
- 3.Mathew AT, Fishbane S, Obi Y, Kalantar-Zadeh K: Preservation of residual kidney function in hemodialysis patients: Reviving an old concept. Kidney Int 90: 262–271, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Vilar E, Farrington K: Emerging importance of residual renal function in end-stage renal failure. Semin Dial 24: 487–494, 2011 [DOI] [PubMed] [Google Scholar]
- 5.Peritoneal Dialysis Adequacy Work Group : Clinical practice guidelines for peritoneal dialysis adequacy. Am J Kidney Dis 48: S98–S129, 2006 [DOI] [PubMed] [Google Scholar]
- 6.Blake PG, Bargman JM, Brimble KS, Davison SN, Hirsch D, McCormick BB, Suri RS, Taylor P, Zalunardo N, Tonelli M; Canadian Society of Nephrology Work Group on Adequacy of Peritoneal Dialysis : Clinical practice guidelines and recommendations on peritoneal dialysis adequacy 2011. Perit Dial Int 31: 218–239, 2011 [DOI] [PubMed] [Google Scholar]
- 7.Woodrow G, Fan SL, Reid C, Denning J, Pyrah AN: Renal Association Clinical Practice Guideline on peritoneal dialysis in adults and children. BMC Nephrol 18: 333, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sirich TL, Funk BA, Plummer NS, Hostetter TH, Meyer TW: Prominent accumulation in hemodialysis patients of solutes normally cleared by tubular secretion. J Am Soc Nephrol 25: 615–622, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kennedy AD, Pappan KL, Donti TR, Evans AM, Wulff JE, Miller LAD, Reid Sutton V, Sun Q, Miller MJ, Elsea SH: Elucidation of the complex metabolic profile of cerebrospinal fluid using an untargeted biochemical profiling assay. Mol Genet Metab 121: 83–90, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhang Q, Ford LA, Evans AM, Toal DR: Structure elucidation of metabolite x17299 by interpretation of mass spectrometric data. Metabolomics 13: 92, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.O’Brien FJ, Sirich TL, Taussig A, Fung E, Ganesan LL, Plummer NS, Brakeman P, Sutherland SM, Meyer TW: Plasma pseudouridine levels reflect body size in children on hemodialysis. Pediatr Nephrol 35: 305–312, 2020 [DOI] [PubMed] [Google Scholar]
- 12.Gehan EA, George SL: Estimation of human body surface area from height and weight. Cancer Chemother Rep 54: 225–235, 1970 [PubMed] [Google Scholar]
- 13.Morgenstern BZ, Wühl E, Nair KS, Warady BA, Schaefer F: Anthropometric prediction of total body water in children who are on pediatric peritoneal dialysis. J Am Soc Nephrol 17: 285–293, 2006 [DOI] [PubMed] [Google Scholar]
- 14.Bargman JM, Thorpe KE, Churchill DN: Relative contribution of residual renal function and peritoneal clearance to adequacy of dialysis: A reanalysis of the CANUSA study. J Am Soc Nephrol 12: 2158–2162, 2001 [DOI] [PubMed] [Google Scholar]
- 15.Wang AY, Sea MM, Ip R, Law MC, Chow KM, Lui SF, Li PK, Woo J: Independent effects of residual renal function and dialysis adequacy on actual dietary protein, calorie, and other nutrient intake in patients on continuous ambulatory peritoneal dialysis. J Am Soc Nephrol 12: 2450–2457, 2001 [DOI] [PubMed] [Google Scholar]
- 16.Termorshuizen F, Korevaar JC, Dekker FW, van Manen JG, Boeschoten EW, Krediet RT; NECOSAD Study Group : The relative importance of residual renal function compared with peritoneal clearance for patient survival and quality of life: An analysis of the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD)-2. Am J Kidney Dis 41: 1293–1302, 2003 [DOI] [PubMed] [Google Scholar]
- 17.Warady BA, Schaefer F, Bagga A, Cano F, McCulloch M, Yap HK, Shroff R: Prescribing peritoneal dialysis for high-quality care in children. Perit Dial Int 40: 333–340, 2020 [DOI] [PubMed] [Google Scholar]
- 18.Babb AL, Johansen PJ, Strand MJ, Tenckhoff H, Scribner BH: Bi-directional permeability of the human peritoneum to middle molecules. Proc Eur Dial Transplant Assoc 10: 247–262, 1973 [PubMed] [Google Scholar]
- 19.Krediet RT, Boeschoten EW, Struijk DG, Arisz L: Differences in the peritoneal transport of water, solutes and proteins between dialysis with two- and with three-litre exchanges. Nephrol Dial Transplant 3: 198–204, 1988 [PubMed] [Google Scholar]
- 20.Montini G, Amici G, Milan S, Mussap M, Naturale M, Rätsch IM, Ammenti A, Sorino P, Verrina E, Andreetta B, Zacchello G; Italian Registry of Pediatric Peritoneal Dialysis : Middle molecule and small protein removal in children on peritoneal dialysis. Kidney Int 61: 1153–1159, 2002 [DOI] [PubMed] [Google Scholar]
- 21.Leypoldt JK, Parker HR, Frigon RP, Henderson LW: Molecular size dependence of peritoneal transport. J Lab Clin Med 110: 207–216, 1987 [PubMed] [Google Scholar]
- 22.Norden AGW, Burling KA, Zeni L, Unwin RJ: A new estimate of the glomerular sieving coefficient for retinol-binding protein 4 suggests it is not freely filtered. Kidney Int Rep 4: 1017–1018, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Leong SC, Sao JN, Taussig A, Plummer NS, Meyer TW, Sirich TL: Residual function effectively controls plasma concentrations of secreted solutes in patients on twice weekly hemodialysis. J Am Soc Nephrol 29: 1992–1999, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hsueh CH, Yoshida K, Zhao P, Meyer TW, Zhang L, Huang SM, Giacomini KM: Identification and quantitative assessment of uremic solutes as inhibitors of renal organic anion transporters, OAT1 and OAT3. Mol Pharm 13: 3130–3140, 2016 [DOI] [PubMed] [Google Scholar]
- 25.Chapron A, Shen DD, Kestenbaum BR, Robinson-Cohen C, Himmelfarb J, Yeung CK: Does secretory clearance follow glomerular filtration rate in chronic kidney diseases? Reconsidering the intact nephron hypothesis. Clin Transl Sci 10: 395–403, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Pham NM, Recht NS, Hostetter TH, Meyer TW: Removal of the protein-bound solutes indican and p-cresol sulfate by peritoneal dialysis. Clin J Am Soc Nephrol 3: 85–90, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wang K, Nguyen M, Chen Y, Hoofnagle AN, Becker JO, Zelnick LR, Kundzins J, Goodling A, Himmelfarb J, Kestenbaum B: Association of tubular solute clearance with symptom burden in incident peritoneal dialysis. Clin J Am Soc Nephrol 15: 530–538, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bammens B, Evenepoel P, Verbeke K, Vanrenterghem Y: Removal of middle molecules and protein-bound solutes by peritoneal dialysis and relation with uremic symptoms. Kidney Int 64: 2238–2243, 2003 [DOI] [PubMed] [Google Scholar]
- 29.Bammens B, Evenepoel P, Verbeke K, Vanrenterghem Y: Time profiles of peritoneal and renal clearances of different uremic solutes in incident peritoneal dialysis patients. Am J Kidney Dis 46: 512–519, 2005 [DOI] [PubMed] [Google Scholar]
- 30.Davenport A: Measuring residual renal function in dialysis patients: Can we dispense with 24-hour urine collections? Kidney Int 89: 978–980, 2016 [DOI] [PubMed] [Google Scholar]
- 31.Shafi T, Levey AS: Measurement and estimation of residual kidney function in patients on dialysis. Adv Chronic Kidney Dis 25: 93–104, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wong J, Kaja Kamal RM, Vilar E, Farrington K: Measuring residual renal function in hemodialysis patients without urine collection. Semin Dial 30: 39–49, 2017 [DOI] [PubMed] [Google Scholar]
- 33.Shafi T, Michels WM, Levey AS, Inker LA, Dekker FW, Krediet RT, Hoekstra T, Schwartz GJ, Eckfeldt JH, Coresh J: Estimating residual kidney function in dialysis patients without urine collection. Kidney Int 89: 1099–1110, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Wong J, Sridharan S, Berdeprado J, Vilar E, Viljoen A, Wellsted D, Farrington K: Predicting residual kidney function in hemodialysis patients using serum β-trace protein and β2-microglobulin. Kidney Int 89: 1090–1098, 2016 [DOI] [PubMed] [Google Scholar]
- 35.Filler G, Huang SH, Lindsay RM: Residual renal function assessment with cystatin C. Pediatr Nephrol 26: 333–335, 2011 [DOI] [PubMed] [Google Scholar]
- 36.Kjaergaard KD, Jensen JD, Rehling M, Jespersen B: Endogenous markers for estimation of renal function in peritoneal dialysis patients. Perit Dial Int 33: 195–204, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Topp H, Armbrust S, Lengger C, Schöch G, Davies J, Stichler W, Manz F, Fusch C: Renal excretion of 8-oxo-7,8-dihydro-2(')-deoxyguanosine: Degradation rates of RNA and metabolic rate in humans. Arch Biochem Biophys 402: 31–37, 2002 [DOI] [PubMed] [Google Scholar]
- 38.Dzúrik R, Lajdová I, Spustová V, Opatrný K Jr.: Pseudouridine excretion in healthy subjects and its accumulation in renal failure. Nephron 61: 64–67, 1992 [DOI] [PubMed] [Google Scholar]
- 39.Dlugajczyk A, Eiler JJ: Lack of catabolism of 5-ribosyluracil in man. Nature 212: 611–612, 1966 [DOI] [PubMed] [Google Scholar]
- 40.Struijk DG, Schoots AC, Koole LH, van der Reijden HJ, Koomen GC, Krediet RT, Arisz L: Transport kinetics of pseudouridine during hemodialysis and continuous ambulatory peritoneal dialysis. J Lab Clin Med 118: 74–80, 1991 [PubMed] [Google Scholar]
- 41.Duranton F, Cohen G, De Smet R, Rodriguez M, Jankowski J, Vanholder R, Argiles A; European Uremic Toxin Work Group : Normal and pathologic concentrations of uremic toxins. J Am Soc Nephrol 23: 1258–1270, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Tanaka H, Sirich TL, Plummer NS, Weaver DS, Meyer TW: An enlarged profile of uremic solutes. PLoS One 10: e0135657, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Rhee EP, Waikar SS, Rebholz CM, Zheng Z, Perichon R, Clish CB, Evans AM, Avila J, Denburg MR, Anderson AH, Vasan RS, Feldman HI, Kimmel PL, Coresh J; CKD Biomarkers Consortium : Variability of two metabolomic platforms in CKD. Clin J Am Soc Nephrol 14: 40–48, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Brown EA, Blake PG, Boudville N, Davies S, de Arteaga J, Dong J, Finkelstein F, Foo M, Hurst H, Johnson DW, Johnson M, Liew A, Moraes T, Perl J, Shroff R, Teitelbaum I, Wang AY, Warady B: International Society for Peritoneal Dialysis practice recommendations: Prescribing high-quality goal-directed peritoneal dialysis. Perit Dial Int 40: 244–253, 2020 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.