Applying Transcriptomics to Fistula Failure – A Novel Approach

An arteriovenous fistula (AVF) is a unique vascular structure leading to distinctive endothelial insults which may contribute to non-maturation or failure of the access. Up to 50% of AVFs undergo aberrant remodeling leading to stenotic segments, increased vascular resistance, and often, lack of maturation. The use of omics technologies is influencing most areas of nephrology and now it’s the turn of vascular access.

For a brief review of how omics technologies are being used in nephrology, check out this previous AJKDBlog post

A study by Martinez et al published in AJKD compares RNA expression in native veins (pre-access) and post-AVF samples to determine signatures associated with non-maturation (failure to achieve an internal diameter ≥ 6 mm). The study included 64 patients undergoing 2-stage procedures from a single surgeon (brachio-basilic AVFs). The initial sample was harvested from a segment of the vein used to create the AV anastomosis and the second sample from the juxta-anastomotic area of both mature and failed AVFs during the transposition stage or salvage procedure.

Twelve patients in each group (similar age, sex, basic demographics in failed & matured cohorts) had RNA-seq performed for gene discovery. Confirmatory analysis was performed in 64 patients using a variety of methods (58 samples using real-time polymerase chain reaction (rtPCR), 45 using immunohistochemistry, and 19 using Western blot). It is not clear to me whether this was planned variability in testing or based on availability. Multiple comparisons were controlled for using false discovery rate assessment.

Sample selection for RNA sequencing (RNA-seq) and confirmatory analyses. (A) Total number of veins and arteriovenous fistulas (AVFs) selected for RNA-seq from each maturation outcome and exclusion criteria. Nine RNA samples did not pass the quality control (QC) requirements for RNA-seq (RNA integrity score [RIN] > 5), and 1 additional vein generated low sequence quality (<80% uniquely mapped reads). After these exclusions, the bottom box indicates the number of samples compared in each experimental group in the differential expression analyses. (B) Total number of veins included in the confirmatory analyses of pre-existing gene expression differences between veins of distinct maturation outcomes. The confirmatory analyses included the veins analyzed in (A) the discovery cohort and were prioritized based on sample availability in the following order: (1) real-time polymerase chain reaction (PCR), (2) immunohistochemistry (IHC), and (3) Western blot. Figure 2 from Martinez et al, AJKD, © National Kidney Foundation.

Higher RNA expression of CSF3R, FPR1, S100A8, S100A9, and VNN2 was found in native veins of non-maturing AVFs which was confirmed using rtPCR in the larger confirmatory cohort. Expression of these genes was not associated with any baseline characteristics of the patients but they were correlated between each other, suggesting a common regulatory pathway or at least expressed by similar cell type. Interestingly, these genes have apparently been implicated in activation of the inflammatory NF-κB and NADPH oxidase pathways.

Protein levels of the S100A8 and S100A9 genes were elevated using Western blotting in the vein segments. There was weak correlation of these proteins with histological intimal hyperplasia but no fibrosis. Regarding post-operative samples, tissues from the two outcome groups were largely transcriptionally similar. Importantly, this suggests that most transcriptional changes that determine maturation outcomes are already present in native veins of AVFs that fail to mature.

Table 2 from Martinez et al, AJKD, © National Kidney Foundation.

This is a novel approach to understanding molecular pathways involved in AVF non-maturation, a big problem in nephrology, and the results unsurprisingly hint at some inflammatory pathways which may be at play. Additional studies with larger patient numbers will be required to fully understand the molecular mechanisms involved in AVF failure, especially as this event is a complex trait with many environmental insults undoubtedly contributing.

– Post prepared by Paul Phelan, AJKDBlog Contributor. Follow him @paulphel.


To view Martinez et al (subscription required), please visit

Title: Transcriptomics of Human Arteriovenous Fistula Failure: Genes Associated With Nonmaturation
Authors: L. Martinez, M. Tabbara, J.C. Duque, G. Selman, N. Santos Falcon, A. Paez, A.J. Griswold, G. Ramos-Echazabal, D.R. Hernandez, O.C. Velazquez, L.H. Salman, and R.I. Vazquez-Padron
DOI: 10.1053/j.ajkd.2018.12.035




Leave a Reply

%d bloggers like this: