ESPN 50th Annual Meeting

ESPN 2017

An exploration of the association between uraemic toxin concentrations and quality of life in children with chronic kidney disease
Snauwaert Evelien 1 De Bruyne Elke 2 Vanherzeele Charlotte 2 Holvoet Els 3 Raes Ann 1 Van Biesen Wim 3 Glorieux Griet 3 Van Hoeck Koen 4 Van Dyck Maria 5 Godefroid Nathalie 6 Roels Sanne 7 Vande Walle Johan 1 Eloot Sunny 3

1- Paediatric Nephrology - Ghent University Hospital
2- Paediatric Psychology - Ghent University Hospital
3- Nephrology - Ghent University Hospital
4- Paediatric Nephrology - Antwerp University Hospital
5- Paediatric Nephrology - Leuven University Hospital
6- Paediatric Nephrology - UCL
7- Data-Analysis - Faculty of Psychology and Pedagogy

As kidney function deteriorates, uraemic toxins accumulate and contribute to the clinical picture of children with chronic kidney disease (CKD). These children are reporting poorer overall quality of life (QoL) and poorer physical, school, emotional, and social functioning. The aim of this study was to explore the association between levels of uraemic toxins and QoL.

Material and methods:

In 23 children (11.0[6.9;14.6]years, 61%boys) with non-dialysis CKD stage 1-5, plasma concentrations of small solutes (asymmetric dimethyl arginine, symmetric dimethyl arginine, creatinine), middle molecules (β2microglobuline, complement factor D), and protein-bound solutes [p-cresylglucuronide, hippuric acid (HA), indole acetic acid (IAA), indoxyl sulfate, p-cresylsulfate, and 3-carboxy-4-methyl-5-propyl-furanpropionic acid (CMPF)] were measured. Their parents were asked to fill in the general (PedsQLTM 4.0 Generic Core: total score, physical & education subscale) and disease-specific QoL questionnaire (PedsQLTM End Stage Renal Disease (ESRD): disease & fatigue subscale). Lasso regression was used as an explorative method to select a set of predictive uraemic toxins (when β≠0) in models for the PedsQL questionnaires.


The mean estimated GFR was 50.4 [31.2;74.5]mL/min/1.73m2. CMPF was found to predict the total PedsQL (β=-0.34) and physical PedsQL subscales score (β=-1.19). Besides CMPF, IAA was predominant in the prediction of the total PedsQL and physical PedsQL subscale score(respectively β=-0.84 and β=-1.26); and HA in the education PesdsQL subscale (β=-0.31). The disease subscale of the PedsQL ESRD questionnaire was predominantly predicted by HA (β=-1.18) and IAA (β=-2.30). Using this model, creatinine was for none of the questionnaires selected as a possible predictor.


This model selected CMPF, IAA and HA as promising predictors for the hard endpoint QoL in children with CKD. Moreover, creatinine was not selected as a possible predictor for any of the QoL measures. A more extensive longitudinal study is necessary to strengthen our findings about the impact of uraemic toxins on the QoL in children with CKD.