Pyridoxamine does not reduce arterial stiffness in an 8-week randomized double-blind placebo-controlled intervention trial with abdominally obese individuals

Arterial stiffening is associated with cardiovascular disease (CVD). Several studies suggest an association of arterial stiffness with advanced glycation end products (AGEs).^{1, 2} The accumulation of AGEs in long-lived proteins such as collagen may lead to cross-linking and possibly to increased arterial stiffness. The dicarbonyl compound methylglyoxal (MGO), a reactive metabolite mainly formed as a by-product of glycolysis, is the major precursor in the formation of AGEs. MGO levels are elevated in obesity and MGO has been associated with the development of many age-related complications³ and incident CVD in diabetes.^{4, 5}

Pyridoxamine (PM), a B6 vitamer and scavenger of MGO, was able to reduce diabetes-induced artery calcification⁶ and was able to prevent aortic stiffening in rat and old mouse models.^7-9 We recently showed that PM prevents vascular dysfunction in mice.¹⁰ These findings suggest the role of PM in reducing arterial stiffening by targeting the formation of MGO stress and AGE-induced collagen cross-linking.

In an RCT with abdominally obese individuals, we recently demonstrated a reduction in MGO and endothelial dysfunction markers by 8 weeks of PM supplementation.¹¹ In this post-hoc analysis of the RCT study, we aim to investigate the effects of this 8 week PM supplementation on arterial stiffness.

Apparently, healthy individuals with abdominal obesity (54% female; mean age 50 years; mean body mass index 32 kg/m²) were randomized to an 8-week intervention with either placebo (n = 36), a low dose of 25 mg PM (n = 36) or a high dose of 200 mg PM (n = 36). An overview of the population characteristics and enrolment log is provided in Table S1 and Figure S1. For a complete overview of the study design, population characteristics and in- and exclusion criteria, see.¹¹

All measurements were performed at baseline (BL) (test day before participants started taking PM capsules) and at follow-up (FU) (after the 8-week intervention period). Regional stiffness measurements (carotid-femoral pulse wave velocity (cfPWV) and augmentation index (AIx)), as well as local carotid artery stiffness measurements (compliance coefficient (CC), distensibility coefficient (cDC), carotid intima-media thickness (cIMT), pulse wave velocity (cPWV) and Young's elastic modulus (cYEM)), were performed and corrected for office blood pressure. An overview of these vascular measurements and corresponding calculations is described in the supplementary methods. Treatment effects were evaluated by one-way analysis of covariance with adjustment for baseline values.

The study was registered at the ClinicalTrials.gov database as NCT02954588 and was approved by the Ethics Committee of Maastricht University Medical Centre.

The total study population comprised 112 participants, of which 108 finished all primary outcome measurements. Table S1 gives an overview of the general characteristics at baseline, Figure S1 describes the enrolment log. The mean compliance during the intervention was >80% based on the number of returned capsules. A dose-dependent trend of pyridoxamine plasma metabolites over the three groups was apparent (p < 0.01) as previously published.¹¹

In this RCT with abdominally obese individuals, we found no treatment effect of an 8-week intervention with PM on arterial stiffness.

Experimental studies have previously described significant improvements in vasculature after PM treatment.^{6-8, 10, 12, 13} We previously found a reduction of MGO, MG-H1 and the endothelial dysfunction markers sICAM-1 and sVCAM-1 by PM.¹¹ Although these studies are of great value to assess the effect of PM in cardiovascular disease, limited clinical studies are available to determine its effect on the macrovasculature in humans.

MGO may affect the arterial wall by both structural and functional alterations. MGO can react irreversibly with lysine and arginine residues to form cross-links, affecting the properties of the arterial wall. In addition to structural alterations of the arterial wall, MGO can affect vascular cell function via intracellular glycation of proteins. Furthermore, the MGO-derived MG-H1 is known to bind to the receptor for AGEs (RAGE) which may lead to low-grade inflammation, all indirectly linked to vascular stiffness.

Although PM supplementation reduces plasma MGO and MG-H1 and improves endothelial function, as estimated by specific biomarkers, this clinical trial also had several limitations. For safety reasons, we used a moderate dosage of PM, and our study population consisted of apparently healthy individuals. As a result, the PM concentration may have been too low, and the participants were too healthy to observe significant improvements in the arterial wall structure. An intervention involving diabetic individuals, who typically exhibit more pronounced structural changes in the vasculature, might yield greater effects on these outcomes. Furthermore, the 8-week intervention may have been too short, although several studies have reported changes in vascular stiffness within an 8-week timeframe.¹⁴ Additionally, the lack of statistically significant findings could be attributed to limited statistical power. To fully elucidate the clinical effect of PM on arterial stiffening, it remains of great interest and relevance to study the effects of PM supplementation over a longer intervention period and/or in patients whose vascular health tends to be more heavily compromised.

In this RCT with abdominally obese individuals, we showed that PM supplementation reduces plasma MGO and MG-H1 levels but does not influence arterial stiffness over an 8-week period.

The present study is funded by the Centre for Translational Molecular Medicine (CTMM) and research grant CH001 from the Top Institute of Food and Nutrition (TIFN). BS was supported by the Netherlands Organisation for Scientific Research (grant no. Rubicon 452172006) and the European Union's Horizon 2020 research and innovation programme (grant no. 793805).

The authors declare no conflict of interest.

The current study was approved by the Ethics Committee of Maastricht University Medical Centre (METC163003).