From helicobacter pylori to glucose metabolism: can DOB values serve as a predictive marker?

Helicobacter pylori infection and abnormal glucose metabolism are prevalent, interconnected contributors to chronic disease. Although metabolic changes have been studied in infected individuals, the independent association between the delta-over-baseline (DOB) value of the ¹³C-urea breath test and fasting blood glucose (FBG) remains unclear. We investigated whether DOB could predict abnormal FBG in adults receiving routine health examinations.

To assess the association between H. pylori infection and metabolic abnormalities, and to evaluate the predictive utility of the DOB value for glycemic abnormalities.

In this retrospective study, 594 patients underwent both the ¹³C-UBT and metabolic parameter assessments. Patients were stratified by DOB values, and metabolic abnormalities were defined by predefined criteria. Logistic regression analyzed the relationship between H. pylori status and metabolic parameters, adjusting for confounders. A restricted cubic spline (RCS) model and receiver operating characteristic (ROC) curve assessed non-linear associations and diagnostic performance of DOB.

Compared with the H. pylori-negative group, the positive group exhibited significantly higher triglyceride (1.667 ± 1.173 vs. 1.447 ± 0.954 mmol/L; p = 0.020) and FBG levels (5.655 ± 1.704 vs. 5.363 ± 1.028 mmol/L; p = 0.024). In multivariable models, H. pylori infection was independently associated with abnormal FBG (OR 2.10; 95% CI 1.30-3.40; p = 0.003) but not with TG abnormalities. The DOB value showed modest overall discriminatory ability for abnormal FBG (AUC = 0.590), with enhanced performance in participants < 40 years (AUC = 0.721).

H. pylori infection is associated with higher fasting glucose and triglyceride levels, and the ¹³C-UBT DOB value showed modest predictive ability for glycemic abnormalities-especially in adults under 40 (AUC = 0.721). DOB may serve as an adjunct risk-stratification marker in younger populations. However, the single-center, cross-sectional design and lack of lifestyle and mechanistic biomarker data limit causal inference. Prospective multicenter cohort studies with serial UBT, clinical (diet, medications, exercise, socioeconomic factors) and biomarker (cytokines, GLP-1) measurements are needed to validate these findings.