Association of Vitamin B6 and Homocysteine-ACTH Coupling With Depressive and Anxiety Symptoms in Burning Mouth Syndrome: An Exploratory Case-Control Study.

backgroundBurning mouth syndrome (BMS) is a chronic, idiopathic and debilitating orofacial pain disorder, increasingly discussed within oral dysaesthesia and nociplastic pain frameworks, in which psychological distress and stress-related mechanisms are frequently implicated. Vitamin B6, homocysteine and adrenocorticotropic hormone (ACTH) reflect biologically interconnected micronutrient, one-carbon-metabolic and neuroendocrine pathways, but their relevance to BMS and particularly their interrelationships within patients remains unclear.

objectivesTo compare plasma vitamin B6, homocysteine and ACTH concentrations between patients with BMS and controls, and to characterise the relationships among these biomarkers and with anxiety and depression within each group.

methodsIn this exploratory case-control ancillary analysis of the OPIODYN study, plasma pyridoxal 5'-phosphate (PLP; vitamin B6), homocysteine and ACTH were measured in 21 patients with BMS and 17 controls, and anxiety and depression were assessed with the Hospital Anxiety and Depression Scale (HADS). Between-group comparisons used 2-sided Mann-Whitney U tests with Hodges-Lehmann median differences and 95% confidence intervals (CIs). Within-group Spearman correlations and between-group differences in correlation strength (Δρ, Monte Carlo permutation) were computed, with Benjamini-Hochberg control of the false discovery rate within pre-declared families.

resultsNo statistically significant between-group differences were observed for PLP, homocysteine or ACTH concentrations. Median differences were +5.35 nmol/L (95% CI, -14.88 to 26.78; p = 0.567) for PLP, -0.13 μmol/L (95% CI, -1.81 to 1.49; p = 0.907) for homocysteine and -0.66 pmol/L (95% CI, -1.61 to 0.35; p = 0.186) for ACTH. Most participants in both groups had adequate PLP and within-range homocysteine and basal ACTH. Homocysteine and ACTH were strongly correlated in patients with BMS (ρ = 0.806; 95% CI 0.562-0.891) but not in controls (ρ = 0.204; 95% CI, -0.344 to 0.668), with a significant between-group difference in correlation strength (Δρ = 0.602; p = 0.018). Among patients with BMS, both HADS subscales showed positive correlations with homocysteine and ACTH (ρ = 0.61-0.64; all q = 0.010), whereas no corresponding correlation was significant in controls, and PLP was not correlated with HADS or with the other biomarkers in either group. No single biomarker discriminated BMS from control status (AUC, 0.51-0.63).

conclusionThese exploratory findings do not support vitamin B6, homocysteine, or ACTH as stand-alone biomarkers of BMS. Rather, they suggest that psychological distress maps onto a homocysteine-ACTH correlation axis in patients with BMS, from which PLP appears dissociated, and support a shift in BMS biomarker research from isolated concentrations towards the relationships among metabolic, neuroendocrine and affective dimensions. Larger studies are needed to test the reproducibility of this pattern and its potential value for identifying clinically meaningful BMS subgroups.