Early changes of fecal short-chain fatty acid levels in patients with mild cognitive impairments

CNS Neurosci Ther. 2023 Nov;29(11):3657-3666. doi: 10.1111/cns.14252. Epub 2023 May 5.

Abstract

Aims: To compare the fecal levels of short-chain fatty acids (SCFAs) in patients with mild cognitive impairment (MCI) and normal controls (NCs) and to examine whether fecal SCFAs could be used as the biomarker for the identification of patients with MCI. To examine the relationship between fecal SCFAs and amyloid-β (Aβ) deposition in the brain.

Methods: A cohort of 32 MCI patients, 23 Parkinson's disease (PD) patients, and 27 NC were recruited in our study. Fecal levels of SCFAs were measured using chromatography and mass spectrometry. Disease duration, ApoE genotype, body mass index, constipation, and diabetes were evaluated. To assess cognitive impairment, we used the Mini-Mental Status Examination (MMSE). To assess brain atrophy, the degree of medial temporal atrophy (MTA score, Grade 0-4) was measured by structural MRI. Aβ positron emission tomography with 18 F-florbetapir (FBP) was performed in seven MCI patients at the time of stool sampling and in 28 MCI patients at an average of 12.3 ± 0.4 months from the time of stool sampling to detect and quantify Aβ deposition in the brain.

Results: Compared with NC, MCI patients had significantly lower fecal levels of acetic acid, butyric acid, and caproic acid. Among fecal SCFAs, acetic acid performed the best in discriminating MCI from NC, achieved an AUC of 0.752 (p = 0.001, 95% CI: 0.628-0.876), specificity of 66.7%, and sensitivity of 75%. By combining fecal levels of acetic acid, butyric acid, and caproic acid, the diagnostic specificity was significantly improved, reaching 88.9%. To better verify the diagnostic performance of SCFAs, we randomly assigned 60% of participants into training dataset and 40% into testing dataset. Only acetic acid showed significantly difference between these two groups in the training dataset. Based on the fecal levels of acetic acid, we achieved the ROC curve. Next, the ROC curve was evaluated in the independent test data and 61.5% (8 in 13) of patients with MCI, and 72.7% (8 in 11) of NC could be identified correctly. Subgroup analysis showed that reduced fecal SCFAs in MCI group were negatively associated with Aβ deposition in cognition-related brain regions.

Conclusion: Reductions in fecal SCFAs were observed in patients with MCI compared with NC. Reduced fecal SCFAs were negatively associated with Aβ deposition in cognition-related brain regions in MCI group. Our findings suggest that gut metabolite SCFAs have the potential to serve as early diagnostic biomarkers for distinguishing patients with MCI from NC and could serve as potential targets for preventing AD.

Keywords: biomarker; early diagnosis; mild cognitive impairment; fecal short-chain fatty acids.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Acetates
  • Alzheimer Disease* / complications
  • Alzheimer Disease* / diagnostic imaging
  • Amyloid beta-Peptides / metabolism
  • Atrophy / complications
  • Butyric Acid
  • Caproates
  • Cognitive Dysfunction* / complications
  • Cognitive Dysfunction* / diagnostic imaging
  • Fatty Acids, Volatile
  • Humans
  • Positron-Emission Tomography / methods

Substances

  • hexanoic acid
  • Caproates
  • Butyric Acid
  • Amyloid beta-Peptides
  • Fatty Acids, Volatile
  • Acetates