Quantification of short chain fatty acids for use in the prognosis of chronic kidney disease

It has recently been suggested that circulating SCFAs play a protective role through the regulation of inflammation caused after the onset of renal disease. Initial evidence has been provided through in vitro, animal model and small-scale human studies, but for translation of SCFAs into a clinical space, further investigations, with larger cohorts, need to be established. This research describes the development and validation of a gas chromatography-mass spectrometry (GC-MS) based assay, for the quantification of seven circulating SCFAs, and its use in the investigation of a healthy and a large chronic kidney disease (CKD) cohort, with the overall aim to investigate the prognostic capability of SCFA concentrations in circulation. </p><p dir="ltr">An assay for the analysis of seven underivatized straight and branched-chain SCFAs (C2-C5) was developed and optimised using entry level GC-MS equipment, employing the use of deuterated internal standards and a straight forward liquid extraction protocol.

The lower limit of detection of the assay was observed to be ≤ 28.6 ng/mL for all SCFAs. Assay precision from multiple injections of, low, medium and high, quality control samples resulted in relative standard deviations of ≤9.2%, and calibrations experiments demonstrated full linearity with correlation coefficients (r 2 ) of ≥0.99. Experiments understanding the influence of matrix effects, analyte recovery, and extraction reproducibility demonstrated acceptable outcomes for validation purposes. </p><p dir="ltr">The method was validated across six different collection modalities and variation between collection methods was assessed to ensure consistency.

Across the multiple biofluid 2 collections, it was determined that samples collected into lithium heparin and serum clotting activator tubes were most appropriate for future analyses. Circulating SCFA concentrations were analysed in 51 healthy participants and associations between SCFA concentrations and clinically relevant measurables were assessed. A significant negative association between serum acetic acid, propanoic acid and butyric acid concentrations and diastolic blood pressure (p<0.05) and a significant negative association between all serum SCFA concentrations, except isovaleric acid and valeric acid, and white blood cell count (p<0.05) was observed.

Additionally, circulating SCFAs were also assessed in a small cross-sectional non-dialysis CKD and end stage kidney disease (ESKD) cohort prior to hemodialysis treatment. Within the ESKD cohort, patients showed increased levels of acetic acid and 2-methlybutyric acid, but lower levels of butyric acid and isovaleric acid compared with non-dialysis CKD groups. </p><p dir="ltr">Finally, an assessment into the association of SCFAs and adverse outcomes of all-cause mortality or a composite outcome of all-cause mortality or any cardiovascular event (CVE) (death or CVE) in a large 1472 patient CKD stage 3 cohort was performed.

Assessments were performed through time independent binary logistic regression and time dependant Cox proportional hazard ratios regression. Results demonstrated that circulating concentrations of butyric acid inversely associated with both adverse outcomes univariately and in CKD-relevant variable adjusted models at basic, intermediate and advanced levels when tested both as time independent and time dependent outcomes.

Butyric acid was investigated further and a dichotomised butyric acid variable was established using the optimal Youden Index estimated via the receiver operator characteristic curve. The prognostic ability of circulating butyric acid for adverse outcomes within CKD patients within five years was investigated using both dichotomised and continuous circulating butyric acid variables. Predictive models including and excluding butyric acid variables, were compared using both net reclassification index (NRI) analysis and integrated discrimination improvement (IDI) analysis.

Results found that models including the butyric acid dichotomised variable showed a significant improvement in predictions for adverse outcome death or CVE in both IDI [0.004 (95% CI, 0.001 to 0.008), p=0.01] and NRI [0.113 (95% CI, 0.004 to 0.2226), p=0.04] analysis. This provided initial evidence for the prognostic benefit of circulating butyric acid levels for prediction of adverse outcomes in CKD. 3 To date, one journal article has been published for the work detailed from Chapter 3 and 4 in this thesis: Bain, J. T., Taal, M. W., Selby, N. M., Reynolds, J. C. & Heaney, L. M. Simple, high throughput measurement of gut-derived short-chain fatty acids in clinically relevant biofluids using gas chromatography-mass spectrometry.

J. Mass Spectrom. Adv. Clin.

Lab 25, 36–43 (2022).</p>