Aims
Colorectal cancer (CRC) remains a major global health burden, and its rising incidence in younger adults underscores the limitations of age-based screening alone. Increasing evidence suggests that metabolic exposures, particularly diabetes and chronic hyperglycemia, may contribute to earlier and more aggressive colorectal neoplasia. This reinforces the need for metabolic risk–based stratification to identify high-risk individuals.This study aimed to assess the association between glycemic control and the presence of colorectal adenomas or cancers, and to develop a simple combined risk score.
Methods
We conducted a single-center retrospective study at Hôtel-Dieu de France in Beirut, including 462 patients who underwent colonoscopy between 2020 and 2025 and had available HbA1c measurements. Demographic, cardiovascular, and metabolic data (age, sex, BMI, hypertension, diabetes, dyslipidemia) were extracted from medical records. The closest HbA1c to colonoscopy and the highest recorded HbA1c were collected. All lesions and cancers were confirmed histologically. Advanced neoplasia was defined as lesions ≥10 mm, villous histology, or high-grade dysplasia or invasive cancer.Statistical analysis included χ² testing, logistic regression (OR, 95% CI), ROC curve analysis for HbA1c discrimination, and Youden-based optimal threshold selection. A multivariable risk score was built using the strongest predictors (age, sex, BMI, diabetes, HbA1cMax).
Results
Among the 462 patients, colorectal lesions were present in 80.7%, and advanced neoplasia in 7.4%. ROC analysis identified an optimal HbA1c cut-off of 5.9% (AUC 0.62, p<0.05).Advanced colorectal lesions were more frequent in patients with HbA1c ≥5.9% for both the closest HbA1c (12.2% vs 5.7%; OR 2.31, 95% CI 1.13–4.71) and HbA1cMax (11.1% vs 5.3%; OR 2.22, 95% CI 1.10–4.48).The final score was:0.35×(Age) + 0.071×(Male sex) + 0.042×(BMI ≥30) + 0.054×(Diabetes) + 0.488×(HbA1cMax)Model performance was satisfactory (AUC 0.708, 95% CI 0.629–0.786). Sensitivity was 88.2% and specificity was 47.2%.A score ≥25 identified high-risk patients (OR 6.70, 95% CI 2.32–19.36, p<0.001).
Conclusions
In this cohort, HbA1c ≥5.9% was associated with approximately double the risk of advanced neoplasia or colorectal cancer. Simple metabolic markers, particularly HbA1c, may improve pre-colonoscopy risk stratification and guide personalized screening intervals. Prospective multicentric studies are needed to validate this metabolic-based risk score.