Periodontitis and metabolic syndrome

Background + Aims

• Diet strongly influences the oral microbiome and gum inflammation.
• Fasting is known to have anti-inflammatory and metabolic effects (lower blood sugar, reduced IGF-1, improved stress resistance), but its impact on the subgingival microbiome had not been examined.
• A previous study by the same team hinted at reduced periodontal inflammation after fasting in women. This analysis extended that work to include men.
• This study aimed to:
  • Evaluate whether a prolonged multimodal fast alters the oral microbiome and periodontal inflammation in patients with metabolic syndrome.

Materials + Methods

• Exploratory secondary analysis of a prospective cohort:
  • 43 non-smoking adults with metabolic syndrome (84% female; mean age 64; mean HbA1c 6.17%).
  • At baseline 72% had suspected severe periodontal disease (PSI code 4).
• Intervention: 2-week inpatient Buchinger-style fast at Immanuel Clinic Berlin.
  • Energy intake 300–500 kcal/day (vegetable broth, fruit juice, tea, water, mineral supplements) for 5–10 days, followed by a 1000–1300 kcal/day reintroduction diet.
• Assessments at 3 timepoints: baseline (T1), after fasting (T2), and 4–5-month follow-up (T3, n = 21 only).
• Outcomes measured: bleeding on probing (BoP), gingival bleeding index (GBI), plaque index (PI), gingival crevicular fluid (GCF), oral pH, c-reactive protein (CRP), fasting glucose, HbA1c, BMI.
• Supra- and subgingival plaque sampled from the deepest pocket, analysed by 16S rRNA V4 gene sequencing
• Diversity assessed as alpha-diversity (variety within a single sample) and beta-diversity (differences between samples), plus a subgingival microbiome dysbiosis index. Differential abundance was cross-checked with 3 statistical models (ANCOM-BC, LinDA, Wilcoxon).

Results

• Clinical changes (T1 → T2)
  • BoP: 36.4% → 30.4% (p = 0.01)
  • GBI: 36.6% → 29.7% (p = 0.02)
  • PI: 66.9% → 58.8% (p < 0.01)
  • GCF: 83.6 → 67.9 Periotron units (p < 0.01)
  • Oral pH: 6.96 → 6.35 (p < 0.01)
  • Fasting glucose: 115 → 84 mg/dL (p < 0.01)
  • BMI: 35.9 → 34.3 kg/m² (p < 0.01)
  • CRP did not change significantly.
• Microbiome
  • Overall composition (Firmicutes, Bacteroidetes, Fusobacteria, Proteobacteria) remained largely stable.
  • Alpha- and beta-diversity and the dysbiosis index did not change significantly.
  • A small set of genera shifted consistently across all 3 statistical models:
    • Decreased: Mitsuokella, Atopobium.
    • Increased: Peptostreptococcaceae [XI][G-7], Eikenella (species-level: Eikenella corrodens), Lachnospiraceae [G-3] (subgingival only).
  • Some increased genera (e.g., Peptostreptococcaceae, Eikenella) have been linked to periodontitis, so not all shifts point in a “healthier” direction.
• 4–5 month follow-up (n = 21)
  • No significant diversity or genus-level differences from baseline but the power was limited.

Limitations