Plasma metabolic profile of Chinese coeliac disease patients: an exploratory observational study
Chinese patients with coeliac disease show distinct plasma metabolic alterations compared to healthy controls and IBD patients, independent of dietary intake, in an extremely small sample.
| Outcome | Grade | Direction | Effect | Studies |
|---|---|---|---|---|
| Pentose phosphate pathway metabolites | D | ▼ Unfavorable | reduzidos em CeD vs controles; sem IC ou tamanho de efeito reportado | 1 |
| TCA cycle intermediates | D | ▲ Favorable | aumentados em CeD vs controles; sem IC ou tamanho de efeito reportado | 1 |
| Plasma L-proline and D-proline | D | ▲ Favorable | aumentadas em CeD vs controles; sem IC ou tamanho de efeito reportado | 1 |
| Glycine-conjugated bile acids | D | ▲ Favorable | aumentados em CeD vs controles; sem IC ou tamanho de efeito reportado | 1 |
| Long-chain acylcarnitines | D | ▼ Unfavorable | reduzidas em CeD apesar de maior ingestão lipídica; sem IC ou tamanho de efeito reportado | 1 |
| Microbiota-related aromatic amino acid metabolites | D | ▲ Favorable | aumentados em CeD vs controles; sem IC ou tamanho de efeito reportado | 1 |
| Niacin micronutrient insufficiency | D | ▼ Unfavorable | insuficiência relativa de niacina em CeD vs controles e DII; sem IC ou tamanho de efeito reportado | 1 |
Context
Coeliac disease is well-characterised in Caucasian populations but remains poorly described in East Asian populations. Identifying a specific metabolomic profile may advance understanding of disease pathophysiology in distinct genetic and dietary contexts. Data on the metabolic exposome of coeliac disease in Chinese individuals are virtually absent.
What the study showed
CeD patients showed reduced metabolites in the pentose phosphate pathway and increased TCA cycle intermediates relative to healthy controls. L-proline, D-proline, microbiota-related aromatic amino acid metabolites, glycine-conjugated bile acids, and panuosterone were elevated. Long-chain acylcarnitines were reduced despite higher reported fat intake. Absolute numbers of differentially expressed metabolites and confidence intervals are not reported by the study.
How it was done
Cross-sectional observational study with 60 participants (15 CeD, 15 healthy controls, 30 IBD) in Northwest China. Metabolomic analysis used UHPLC-QTRAP-MS with a pseudo-targeted approach; dietary intake was assessed by food frequency questionnaire. Metabolomic data were adjusted for energy and 17 specific nutrients.
Effect magnitude
The study does not report standardised effect sizes (OR, RR, SMD, MD) or confidence intervals for identified metabolites; the magnitude of between-group differences is not rigorously quantified.
Limitations
Extremely small sample (n=15 in the CeD group), severely limiting statistical power and generalisability; no formal risk-of-bias tool was applied (ROBINS-I, applicable to observational studies, was not used). Cross-sectional design precludes causal inference. Dietary assessment by food frequency questionnaire is subject to recall bias. Absence of 95% CIs and formal effect sizes undermines clinical interpretation.
In clinical practice
This study does not provide sufficient basis for changes in clinical practice. Findings are exploratory and generate hypotheses about metabolic biomarkers in coeliac disease in Chinese populations. Clinicians should await larger, confirmatory studies before any diagnostic or therapeutic application.
What is still missing
Studies with larger, representative samples, longitudinal design, and external validation of identified metabolites are needed. Causality between metabolic alterations and coeliac disease pathogenesis in Asian populations remains to be established.