Taste perception and food preferences in patients with diabetic foot ulcers before and after hyperbaric oxygen therapy

The study investigates whether hyperbaric oxygen therapy (HBOT) alters gustatory function and food preferences in patients with type 2 diabetes who have diabetic foot ulcers. Taste perception substantially influences eating behavior and nutrient intake, and diabetes is associated with oral pathologies and neuropathy that can impair taste. Prior work links chemosensory dysfunction with diabetic complications and suggests HBOT benefits wound healing in diabetic foot ulcers. The research question asks if HBOT can improve sensitivity to the five basic tastes (salty, sweet, umami, sour, bitter) and shift food preferences in this population. This is important because diet is central to diabetes management, and modifying taste perception could support healthier eating behaviors.

Prior studies document increased incidence of oral mucosal disorders and altered taste in type 2 diabetes, with associations to glycemic control, vascular complications, peripheral neuropathy, and microalbuminuria. Interventions for taste disorders include zinc supplementation, intranasal theophylline, and trigeminal nerve magnetic stimulation, but treating the underlying condition yields the best outcomes. HBOT improves healing in diabetic foot ulcers and may influence chemosensory function; earlier work in heterogeneous non-healing wounds suggested HBOT-related taste benefits. Evidence in diabetics shows elevated recognition thresholds for salty and sour tastes and altered sweet taste function, with mixed findings for bitter. Mechanistically, animal models implicate TRPM5 pathways affecting sweet, umami, and bitter taste and insulin secretion. HBOT has been reported to improve metabolic and inflammatory parameters in diabetic patients and animal models, and to enhance oral mucosal vascularity and saliva in irradiated patients, suggesting plausible pathways for chemosensory improvement.

Design: Comparative study with a healthy control group and a within-subject pre/post HBOT assessment in diabetic patients with foot ulcers. Groups: 75 healthy controls (Group C) tested once; 23 patients with type 2 diabetes and diabetic foot ulcers tested before HBOT (Group Db) and after 25–30 HBOT sessions (Group Da). Group C was age- and BMI-matched to Group D. Inclusion/exclusion criteria were applied by a hyperbaric physician; severe conditions impeding testing were excluded. HBOT protocol: Multiplace chamber, 2.5 ATA sessions using compressed air, patients breathing 100% oxygen via mask; session duration 87 minutes; 25–30 sessions on consecutive weekdays. Diabetic patients had capillary glucose measured before each chamber entry. Taste testing: Conducted per ISO 3972 with modifications. Five tastants: sucrose (sweet), NaCl (salty), monosodium glutamate (umami), citric acid (sour), quinine HCl (bitter). Sip-and-spit method with blinded, randomized 15 mL solutions in series of 10 ascending concentrations to determine recognition threshold (lowest correctly recognized concentration). After a 15-minute rest, three suprathreshold concentrations were presented for each taste to rate intensity on a 10-cm line scale (0–10) and hedonic valence on a −5 to +5 line scale. Suprathreshold concentrations: NaCl 0.18%, 0.36%, 0.90%; sucrose 1%, 10%, 30%; MSG 0.1%, 0.3%, 1.0%; citric acid 0.02%, 0.04%, 0.10%; quinine HCl 0.001%, 0.002%, 0.005%. Testing for controls occurred over two mornings fasting; diabetics tested in mornings, allowed breakfast 2 hours prior to avoid hypoglycemia. Food preference testing: Participants rated the pleasantness (0–10 scale) of 20 food categories from photographs (e.g., vegetables/salads, fruits, desserts, chocolate, milk dishes, cheese, dumplings, pasta, bread, beef/pork, poultry, broth, eggs, fish, seafood, fast food, salty products, crisps, sour products, spicy dishes). Ulcer assessment: Ulcer stage classified by the University of Texas Diabetic Foot Ulcer Classification System; planimetric area measured pre/post HBOT when feasible; 87% showed area decrease post-HBOT. Statistics: Mann–Whitney U test compared Group C vs Group Db. Wilcoxon signed-rank test compared Db vs Da. Significance at p < 0.05. Data handled in Excel 2010 and Statistica 12.0.

• Baseline (Db) vs Controls (C): Recognition thresholds were higher (worse sensitivity) in diabetics for all five tastes: salty (p<0.05), sweet (p<0.01), umami (p<0.01), sour (p<0.001), bitter (p<0.01). Median recognition thresholds (C vs Db): salty 6 vs 8 (0.68 g/L vs 2.00 g/L NaCl); sweet 7 vs 8 (7.20 g/L vs 12.00 g/L sucrose); umami 8 vs 11 (1.00 g/L MSG vs no sensation); sour 5 vs 9 (0.0216 g/L vs 0.1300 g/L citric acid); bitter 6 vs 9 (0.3156 µg/L vs 1.5505 µg/L quinine HCl).
• Intensity (C vs Db): Controls perceived higher intensity for 0.1% and 1.0% MSG (medians 1.9 vs 1.0, and 7.2 vs 5.9; both p<0.05), 0.02% citric acid (median 4.3 vs 2.4; p<0.05), and 0.002% quinine HCl (median 7.0 vs 5.1; p<0.05).
• Hedonic response (C vs Db): Controls showed more negative hedonic ratings for 0.18% NaCl (p<0.05), 0.3% MSG (p<0.05), and 0.10% citric acid; e.g., NaCl 0.18% Min and Q1 in C −4.5 and −0.1 vs Db 0.0 and 0.0.
• Food preferences (C vs Db): Diabetics reported higher pleasure for sour products (median 7.3 vs 5.9; p<0.05) and salty products (median 5.0 vs 2.8; p<0.05).
• Effect of HBOT (Db vs Da): Recognition thresholds decreased (improved) for umami (p<0.05; median Db 11 [no sensation] to Da 10 [2.04 g/L MSG]) and sour (p<0.05; median Db 9 [0.1300 g/L] to Da 8 [0.0830 g/L]). Intensity of umami increased at all suprathresholds: 0.1%, 0.3%, 1.0% MSG (all p<0.05); medians increased from 1.0→2.0; 3.4→4.0; 5.9→7.0. No significant HBOT effect on hedonic ratings of the suprathreshold taste solutions. Food preferences after HBOT: decreased pleasure for chocolate (median 6.7→5.1; Q3 8.1→6.5; p<0.05) and a trend toward decreased pleasure for crisps (median 1.3→1.1; Q3 5.3→2.8; p=0.050).

The study demonstrates that patients with type 2 diabetes and diabetic foot ulcers exhibit broad chemosensory impairments across all five basic tastes and distinct hedonic and food preference patterns compared with healthy controls. HBOT partially addressed these deficits, significantly improving sensitivity to umami and sour tastes and reducing pleasure for certain unhealthy, palatable foods (chocolate, crisps). These changes support the hypothesis that HBOT has beneficial effects on chemosensory function and may influence food choices. Potential mechanisms include improved oral mucosal vascularity and function, altered inflammatory and metabolic profiles, and neurohormonal modulation associated with HBOT. Enhanced umami sensitivity may reduce liking for umami-enhanced snack foods (e.g., crisps), aligning taste perception with healthier preference patterns. While effects on salty, sweet, and bitter sensitivity were limited, the observed improvements suggest HBOT could be an adjunct to support dietary behavior in advanced diabetic complications.

People with advanced type II diabetes and diabetic foot ulcers have impaired perception of salty, sweet, umami, sour, and bitter tastes and altered food preferences versus healthy individuals. HBOT produced beneficial changes, increasing sensitivity to umami and sour and decreasing the pleasure derived from eating chocolate and crisps. Further research should establish the clinical relevance, determine how disease severity modulates HBOT’s effects, and assess the durability and impact on actual dietary intake.

• Generalizability: Participants had advanced type 2 diabetes with diabetic foot ulcers; severe cases unable to perform testing were excluded. Findings may apply to a relatively narrow subgroup of diabetics.
• Sample size: Small diabetic cohort (N=23) completing pre/post HBOT.
• No HbA1c data: Glycemic control was not characterized by HbA1c, limiting metabolic interpretation.
• Diet monitoring: No comprehensive dietary assessment before and after HBOT to link taste changes to actual intake.
• Methodological variability: Although ISO 3972-based procedures were used, gustatory testing lacks a gold standard; subjective assessments could introduce bias.
• Some ulcers were assessed visually without planimetry due to location, limiting consistency in wound outcome measures.