Background: Dyspepsia is a common gastrointestinal complaint globally, affecting approximately 21.8% of the population. Among patients presenting with dyspeptic symptoms, over 80% are diagnosed with functional dyspepsia (FD), while approximately 16% are found to have chronic atrophic gastritis (CAG). CAG represents an important precancerous condition in the gastric cancer cascade, yet the relationship between pathologically confirmed CAG and dyspeptic symptoms remains poorly understood. The significant symptom overlap between CAG and FD creates diagnostic challenges in clinical practice. Study Objectives: The primary objective is to determine whether there are significant differences in the prevalence and severity of dyspeptic symptoms (including epigastric pain, burning sensation, early satiety, and postprandial fullness) between patients with pathologically confirmed CAG and those without CAG (non-CAG group) among individuals who present with endoscopic features suggestive of atrophic gastritis. Secondary objectives include: (1) analyzing the independent effects of various covariates (Helicobacter pylori infection, dietary habits, sleep quality, psychological factors) on dyspeptic symptoms; (2) developing a symptom-based predictive model for pathological CAG; and (3) conducting exploratory serum metabolomics analysis to identify potential biomarkers and metabolic pathways associated with FD symptoms. Study Design: This is a single-center, prospective, observational study conducted at the Third Affiliated Hospital of Zhejiang Chinese Medical University. The study will enroll approximately 258-315 adult patients (aged 18-75 years) who undergo endoscopy showing features suggestive of CAG within the past year. All participants will undergo standardized 5-point gastric mucosal biopsy according to the Updated Sydney System. Based on histopathological results, patients will be classified into pathological CAG group (presence of gastric mucosal atrophy) or non-CAG group (absence of atrophy). The study aims to recruit at least 80 pathologically confirmed non-CAG patients for comparison. Study Procedures: After obtaining informed consent, all enrolled patients will complete a comprehensive assessment at baseline including: demographic information, medical history, endoscopy and pathology results, Gastrointestinal Symptom Scale (GOSS) questionnaire using a 7-point Likert scale, H. pylori infection status (serology), dietary habits assessment, Pittsburgh Sleep Quality Index (PSQI), Self-Rating Anxiety Scale (SAS), Self-Rating Depression Scale (SDS), and perceived stress evaluation. A subset of participants will provide fasting blood samples for non-targeted metabolomics analysis using liquid chromatography-mass spectrometry (LC-MS) to identify metabolites related to amino acids, organic acids, lipids, and neurotransmitter precursors. This is a non-interventional study with all data and sample collection completed at enrollment without long-term follow-up. Primary Outcome: The primary outcome is the difference between pathological CAG and non-CAG groups in the prevalence and severity of dyspeptic symptoms, particularly cardinal FD symptoms (epigastric pain, burning, early satiety, postprandial fullness), assessed using the GOSS scale. A symptom score ≥4 on any cardinal symptom will define the presence of clinically significant FD symptoms. Expected Duration: The study is expected to last 24 months, including preparation, patient recruitment with data collection, and final statistical analysis and reporting phases. Significance: This study will provide evidence-based insights into the relationship between pathologically confirmed CAG and dyspeptic symptoms, potentially improving symptom management strategies and patient counseling. The metabolomics component may reveal novel biomarkers and pathways underlying symptom generation, laying groundwork for future mechanistic studies and personalized therapeutic approaches. Results will inform clinical practice and serve as preliminary data for larger-scale investigations.
Scientific Background and Rationale Chronic atrophic gastritis (CAG) represents a critical precancerous lesion in the Correa cascade (chronic gastritis → atrophic gastritis → intestinal metaplasia → dysplasia → gastric adenocarcinoma). Despite its clinical importance, the relationship between histopathologically confirmed CAG and dyspeptic symptom profiles remains poorly characterized. Current clinical practice faces a diagnostic challenge: endoscopic features suggestive of atrophic gastritis correlate imperfectly with histopathological confirmation, and substantial symptom overlap exists between CAG and functional dyspepsia (FD), which accounts for over 80% of dyspeptic presentations. Previous studies examining symptom profiles in CAG patients have yielded inconsistent results, with limited data using standardized pathological criteria (Updated Sydney System, OLGA staging) combined with validated symptom assessment tools. Additionally, the independent contributions of H. pylori infection, psychological comorbidities, sleep disturbances, and dietary patterns to dyspeptic symptoms in CAG remain incompletely understood. This prospective observational study addresses these knowledge gaps by comparing symptom burden between pathologically confirmed CAG and non-CAG patients, while systematically evaluating multifactorial influences on symptom generation. Study Design and Methodology Patient Identification and Recruitment: Patients will be recruited from the gastroenterology outpatient clinic and endoscopy unit at the Third Affiliated Hospital of Zhejiang Chinese Medical University. Potential participants include those with recent (≤1 year) endoscopic findings suggestive of atrophic gastritis, those undergoing CAG surveillance, and those presenting with new-onset dyspepsia requiring endoscopic evaluation. Based on estimated prevalence (approximately 25-30% of endoscopically suspected CAG cases show no atrophy histologically), screening 258-315 patients is projected to yield at least 80 pathologically confirmed non-CAG controls, providing \>80% power to detect a 15% between-group difference in FD symptom prevalence (α=0.05, two-sided). Histopathological Classification: All participants undergo standardized 5-point gastric mucosal biopsy following the Updated Sydney System: lesser curvature of antrum (2-3 cm from pylorus), greater curvature of antrum, lesser curvature of corpus, greater curvature of corpus, and incisura angularis. Specimens are evaluated by experienced gastrointestinal pathologists blinded to clinical and symptom data. Atrophy is graded using the Operative Link for Gastritis Assessment (OLGA) staging system. Patients are classified as CAG group (confirmed glandular atrophy in any biopsy site) or non-CAG group (no atrophic changes despite endoscopic suspicion). Comprehensive Assessment Protocol: Following informed consent, participants complete a single baseline assessment without longitudinal follow-up: Symptom Quantification: The Gastrointestinal Symptom Scale (GOSS) measures dyspeptic symptoms on a 7-point Likert scale (0=absent, 6=very severe) over the preceding two weeks. Cardinal FD symptoms (epigastric pain, burning, early satiety, postprandial fullness) are analyzed individually and compositely. A score ≥4 on any cardinal symptom defines clinically significant FD symptoms. Multifactorial Covariate Assessment: H. pylori status: Serological testing (anti-H. pylori IgG); documentation of recent eradication therapy if applicable Dietary habits: Structured questionnaire assessing meal regularity, spicy/fried/cold food consumption, alcohol, smoking, tea/coffee intake Sleep quality: Pittsburgh Sleep Quality Index (PSQI; 19 items across 7 components; global score \>5 indicates poor sleep) Psychological status: Self-Rating Anxiety Scale (SAS) and Self-Rating Depression Scale (SDS); 20 items each; index scores ≥50 indicate clinically significant symptoms Perceived stress: Visual analog scale (0-10) Exploratory Metabolomics Substudy: A subset of 60-80 participants (balanced across CAG/non-CAG and symptom presence/absence) provide fasting blood samples for serum metabolomics. Ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UHPLC-Q-TOF-MS) in dual ionization modes enables non-targeted metabolite profiling. Metabolite identification utilizes HMDB, METLIN, and KEGG databases. Multivariate analyses (PCA, OPLS-DA) identify differentially abundant metabolites; pathway enrichment analysis explores biological mechanisms potentially linking metabolic alterations to symptom generation. Target metabolite classes include amino acids and derivatives (tryptophan, tyrosine, glutamate), organic acids, lipid species, bile acids, and neurotransmitter precursors implicated in brain-gut axis dysfunction. Analytical Approach Primary Analysis: Between-group comparisons (CAG vs. non-CAG) for symptom prevalence (chi-square/Fisher's exact test) and severity scores (independent t-test/Mann-Whitney U test based on distribution). Effect sizes reported as odds ratios or mean differences with 95% confidence intervals. Secondary Analyses: Multivariable logistic regression identifying independent predictors of clinically significant FD symptoms, adjusting for age, sex, H. pylori status, BMI, psychological factors, sleep quality, and dietary habits Development of a symptom-based predictive model for pathological CAG using logistic regression with backward stepwise selection; model performance evaluated via AUC-ROC, calibration plots, and bootstrap internal validation Metabolomics analysis with false discovery rate correction for multiple comparisons; metabolites with VIP scores \>1.0 and adjusted p\<0.05 considered significant; correlation analyses between differential metabolites and symptom severity Data Management and Quality Assurance Data entry into secure electronic case report forms with built-in validation checks ensures accuracy. Quality control measures include: standardized staff training, regular monitoring of completeness/consistency, double data entry for critical variables, and periodic audits. Patient identifiers are replaced with unique study codes; only authorized personnel access the database. Clinical and Scientific Significance This investigation will provide evidence-based insights into whether pathologically confirmed CAG associates with distinct symptom profiles compared to endoscopically suspected but histologically non-atrophic gastritis. Results may inform clinical decision-making regarding surveillance strategies, symptom management approaches, and patient counseling. The symptom-based predictive model may assist clinicians in estimating CAG probability using readily available clinical variables. The exploratory metabolomics component may identify novel biomarkers and metabolic pathways underlying symptom generation, generating hypotheses for future mechanistic studies and potentially supporting development of non-invasive diagnostic tools or targeted therapeutic interventions. Study duration: 24 months encompassing preparation, recruitment/data collection, and analysis/reporting phases.
Study Type
OBSERVATIONAL
Enrollment
315
The Third Affiliated Hospital of Zhejiang Chinese Medicinal University
Hangzhou, Zhejiang, China
RECRUITINGDifference in Incidence and Severity of Dyspepsia Symptoms
Comparison of the incidence and severity of typical functional dyspepsia symptoms (epigastric pain, burning sensation, early satiety, postprandial fullness) between pathologically confirmed CAG and non-CAG groups
Time frame: At enrollment (baseline)
Independent Effect of Helicobacter pylori Infection on Dyspepsia Symptom Severity
Adjusted odds ratio (OR) with 95% confidence interval for the association between H. pylori infection status (positive vs. negative based on serology) and clinically significant FD symptoms (GOSS score ≥4 on any cardinal symptom), derived from multivariable logistic regression analysis adjusting for age, sex, BMI, psychological factors, sleep quality, and dietary habits.
Time frame: At enrollment (baseline)
Independent Effect of Sleep Quality on Dyspepsia Symptom Severity
Adjusted odds ratio (OR) with 95% confidence interval for the association between poor sleep quality (Pittsburgh Sleep Quality Index global score \>5) and clinically significant FD symptoms (GOSS score ≥4 on any cardinal symptom), derived from multivariable logistic regression analysis adjusting for age, sex, BMI, H. pylori status, psychological factors, and dietary habits.
Time frame: At enrollment (baseline)
Independent Effect of Anxiety on Dyspepsia Symptom Severity
Adjusted odds ratio (OR) with 95% confidence interval for the association between clinically significant anxiety (Self-Rating Anxiety Scale index score ≥50) and clinically significant FD symptoms (GOSS score ≥4 on any cardinal symptom), derived from multivariable logistic regression analysis adjusting for age, sex, BMI, H. pylori status, depression, sleep quality, and dietary habits.
Time frame: At enrollment (baseline)
Independent Effect of Depression on Dyspepsia Symptom Severity
Adjusted odds ratio (OR) with 95% confidence interval for the association between clinically significant depression (Self-Rating Depression Scale index score ≥50) and clinically significant FD symptoms (GOSS score ≥4 on any cardinal symptom), derived from multivariable logistic regression analysis adjusting for age, sex, BMI, H. pylori status, anxiety, sleep quality, and dietary habits.
Time frame: At enrollment (baseline)
Independent Effect of Perceived Stress on Dyspepsia Symptom Severity
Adjusted odds ratio (OR) with 95% confidence interval for the association between high perceived stress (visual analog scale score \>7 out of 10) and clinically significant FD symptoms (GOSS score ≥4 on any cardinal symptom), derived from multivariable logistic regression analysis adjusting for age, sex, BMI, H. pylori status, anxiety, depression, sleep quality, and dietary habits.
Time frame: At enrollment (baseline)
Independent Effect of Dietary Habits on Dyspepsia Symptom Severity
Adjusted odds ratio (OR) with 95% confidence interval for the association between unhealthy dietary patterns (composite score based on irregular meal timing, frequent consumption of spicy/fried/cold foods, and alcohol intake) and clinically significant FD symptoms (GOSS score ≥4 on any cardinal symptom), derived from multivariable logistic regression analysis adjusting for age, sex, BMI, H. pylori status, psychological factors, and sleep quality.
Time frame: At enrollment (baseline)
Discriminative Ability of CAG Prediction Model (AUC-ROC)
Area under the receiver operating characteristic curve (AUC-ROC) with 95% confidence interval for a multivariable logistic regression prediction model for pathological CAG. The model incorporates cardinal FD symptom severity scores (epigastric pain, burning, early satiety, postprandial fullness from GOSS scale), H. pylori infection status, sleep quality (PSQI score), psychological factors (SAS and SDS scores), perceived stress, and dietary habits. Model performance is evaluated on both the derivation cohort and via bootstrap internal validation (1000 iterations).
Time frame: At enrollment (baseline)
Sensitivity of CAG Prediction Model
Sensitivity (true positive rate) with 95% confidence interval at the optimal predicted probability threshold for the multivariable logistic regression prediction model for pathological CAG. The threshold is determined using Youden's index to maximize the sum of sensitivity and specificity. The model is derived from cardinal FD symptom severity scores, H. pylori infection status, sleep quality, psychological factors, perceived stress, and dietary habits, and validated via bootstrap resampling.
Time frame: At enrollment (baseline)
Specificity of CAG Prediction Model
Specificity (true negative rate) with 95% confidence interval at the optimal predicted probability threshold for the multivariable logistic regression prediction model for pathological CAG. The threshold is determined using Youden's index to maximize the sum of sensitivity and specificity. The model is derived from cardinal FD symptom severity scores, H. pylori infection status, sleep quality, psychological factors, perceived stress, and dietary habits, and validated via bootstrap resampling.
Time frame: At enrollment (baseline)
Calibration of CAG Prediction Model
Calibration plots displaying observed versus predicted probabilities of pathological CAG across deciles of predicted probability, accompanied by the Hosmer-Lemeshow goodness-of-fit test p-value and calibration slope with 95% confidence interval. Calibration assesses whether the prediction model's estimated probabilities match the actual occurrence of CAG. The model is derived from cardinal FD symptom severity scores, H. pylori infection status, sleep quality, psychological factors, perceived stress, and dietary habits, with internal validation via bootstrap resampling (1000 iterations).
Time frame: At enrollment (baseline)
Number and Identity of Significantly Differential Serum Metabolites Between CAG and Non-CAG Groups
Non-targeted metabolomics analysis using ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UHPLC-Q-TOF-MS) in serum samples from 60-80 enrolled participants (balanced between CAG and non-CAG groups). Metabolites are identified using HMDB, METLIN, and KEGG databases and classified into five categories: amino acids and derivatives (tryptophan, tyrosine, glutamate), organic acids, lipid species (phospholipids, sphingolipids), bile acids, and neurotransmitter precursors. Significantly differential metabolites between CAG and non-CAG groups are identified using orthogonal partial least squares discriminant analysis (OPLS-DA) with variable importance in projection (VIP) score \>1.0 and false discovery rate (FDR)-adjusted p-value \<0.05. Reported outputs: number of differential metabolites, metabolite identity, metabolite class, VIP score, and adjusted p-value.
Time frame: At enrollment (baseline)
Fold Change in Serum Concentration of Significantly Differential Metabolites Between CAG and Non-CAG Groups
For each significantly differential metabolite identified (VIP \>1.0, FDR-adjusted p\<0.05), fold change in serum concentration between groups is calculated and reported as the ratio of mean metabolite concentration in CAG group to mean concentration in non-CAG group. This metric quantifies the magnitude of metabolite level differences between comparison groups, indicating whether metabolite levels are elevated or decreased in CAG patients relative to non-CAG controls.
Time frame: At enrollment (baseline)
Correlation Coefficient Between Serum Metabolite Concentration and Cardinal FD Symptom Severity
Exploratory correlation analysis examining relationships between concentrations of significantly differential metabolites and cardinal FD symptom severity scores (epigastric pain, burning, early satiety, postprandial fullness assessed via GOSS scale) in metabolomics substudy participants (n=60-80). Pearson or Spearman correlation coefficients with 95% confidence intervals are calculated for each metabolite-symptom pair. False discovery rate correction is applied for multiple comparisons. Reported outputs: Pearson/Spearman correlation coefficient (r or ρ) and FDR-adjusted p-value for each metabolite-symptom correlation, identifying which metabolites associate with symptom severity.
Time frame: At enrollment (baseline)
Metabolic Pathways Enriched Among Significantly Differential Serum Metabolites
Pathway enrichment analysis of significantly differential metabolites (VIP \>1.0, FDR-adjusted p\<0.05) using metabolic pathway databases (KEGG, MetaboAnalyst). Analysis identifies canonical metabolic pathways over-represented among differential metabolites. Special attention to pathways implicated in brain-gut axis dysfunction, intestinal barrier integrity, immune regulation, and neurotransmitter synthesis. Reported metrics: pathway names identified, number of differential metabolites mapped to each pathway, pathway impact score, and statistical significance of pathway enrichment (p-value), revealing biological mechanisms potentially underlying symptom generation in CAG and FD.
Time frame: At enrollment (baseline)
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.