Abstract
Background and Aims
Peritoneal dialysis (PD) are the main renal replacement treatment for children and adolescents with end stage kidney disease (ESRD). Peritoneal fibrosis is a major complication in long-term PD patients. Aim of the present study is to record the metabolic "fingerprint" of children on PD and to investigate its correlation with PD history and dialysis adequacy as well as the emergence of potential biomarkers that could detect early or predict peritoneal dysfunction.
Method
Samples of serum, urine and peritoneal effluent collected from 15 children with ESRD on PD. At the same time and 6 months later was performed PET-test as well as assessment of KT/V and creatinine clearance. Samples were subjected to targeted metabolomic analysis of amino acids and derivatives. Regarding metabolic technologies used, all samples (peritoneal fluid, urine, blood) were analyzed by a hydrophilic interaction liquid chromatography coupled to mass spectrometry (HILIC-MS / MS) method previously developed and validated in our laboratory for the simultaneous determination of amino acids and their derivatives in biological fluids. Αalso, high flow analysis was carried out (LC-q-tof analysis – HPLC /MS ).
Results
Using (HILIC-MS / MS) method, we found out that peritoneal dialysis duration, presence or absence of diuresis and PD creatinine clearance values are associated with significant differences in the levels of several metabolites, including glycine, creatinine, proline and 4-hydroxyproline, leucine, valine, glutamine and glutamic acid. Using (LC-q-tof analysis – HPLC /MS) approximately 200 metabolites were analyzed in the aforementioned samples. Figure 1 illustrates the number of metabolites detected in each matrix, as well as the common metabolites between the three matrices. These metabolites were associated with peritoneal dialysis duration, creatine clearance and presence or absence of diuresis. Several metabolites showed statistical difference between the examined groups. In detail, regarding serum analysis, five metabolites, including hydroxy phenyl acetic acid, glutathione ox, glucosamine-6-P, indole acetic acid and riboflavin showed statistical difference between the examined groups based on PD vintage. Based on urine excretion four metabolites named histidine, shikimic acid, thiamine and methionine were statistically different. Concerning urine analysis, two metabolites namely uridine and itaconic acid showed statistical difference when patients sub grouped based on PD vintage. Peritoneal fluid analysis highlighted one metabolite, uridine, that levels are significantly lower in patients on PD therapy for more than 4 years while based on creatinine clearance levels subgroups, two metabolites, lactate and pantothenic acid present statistically significant difference.
Conclusion
Metabolomics may be a tool in the evaluation of patients with ESRD on PD as it appears to reflect the clinical phenotype of the patient and the functional phenotype of the peritoneal membrane. Our results are the preliminary results of an ongoing prospective study. Limitation of the study is the small sample of patients, which does not allow safe clinical interpretation.