Comparative metabolomic study of high-flux hemodialysis and high volume online hemodiafiltration in the removal of uremic toxins using 1H NMR spectroscopy

Uremic toxins (UTs) accumulate in the circulation of patients with chronic kidney disease (CKD). High volume hemodiafiltration (HDF) improves clearance of low and medium molecular weight UTs compared to HD. The present study is a post-hoc analysis comparing the metabolomic profile in serum from patients under high flux HD (hf-HD) and HDF in HDFIT, a multicentric randomized controlled trial (RCTs). Per protocol, serum samples were collected pre- and post- dialysis treatments at randomization (baseline) and at the end of the follow up (6 months) and stored in a biorepository. Random (pre- and post-dialysis) samples from nine patients in study arm were selected at baseline and at the end of the follow up. To compare the samples, 26 possibly matching metabolites were identified by a t-test among the four groups using 1H nuclear magnetic resonance (NMR). To evaluate the comparison between the modalities is a single treatment session, the clearance rates (CRs) of each metabolite were calculated based on pre-dialysis and post-dialysis samples. In addition, to evaluate to effect of UT removal during the trial follow up period, the pre-dialysis metabolite concentrations at the baseline and at 6 months were compared among the two arms of the study. There was no significant difference between in the single session CRs of metabolites when hf-HD and HDF were compared. On the other hand, the comparison between baseline and 6-month (long-term evolution) led to the identification of 16 metabolites that differentiated the hf-HD and the HDF evolutions. Most of these 16 metabolites are involved in several important metabolic pathways, such as metabolism of phenylalanine and biosynthesis of phenylalanine, tyrosine, and tryptophan, which are related to UTs and cardiovascular disease development. Although no difference was observed between hf-HD and HDF samples before and after a single session, concentrations of CKD-relevant metabolites and associated pathologies were stable in the HDF samples, but not in the hf-HD samples, over the six-month period, suggesting that HDF enhances long-term stability. [Display omitted] •1H NMR spectroscopy and multivariate analysis provided a comprehensive profile of serum metabolites from CKD patients.•Compared to a single dialysis session, hf-HD and HDF showed no metabolites clearance rate difference.•CKD metabolites with associated pathologies showed stability in HDF, suggesting that HDF enhances long-term stability.