Predictive Model for Hypoglycaemia Risk in Type 2 Diabetes Mellitus Patients During the Peri-Colonoscopy Period: A Retrospective Cohort Study

Abstract

Aims

To identify factors influencing hypoglycaemia in patients with type 2 diabetes mellitus (T2DM) undergoing colonoscopy and to construct a predictive model for assessing hypoglycaemia risk.

Design

A retrospective cohort study.

Methods

We retrospectively collected data on 598 T2DM patients who underwent colonoscopy and randomised them into a developmental cohort and a validation cohort in a 7:3 ratio. We used multivariate logistic regression to develop a predictive model for hypoglycaemia during colonoscopy and identify independent predictors in pre- and post-colonoscopy hypoglycaemia groups.

Results

We identified 112 of 598 (18.7%) T2DM patients who experienced hypoglycaemia during the peri-colonoscopy period: 43 pre-colonoscopy, 61 post-colonoscopy and 8 at both junctures. Ultimately, five predictors—insulin, SGLT2 inhibitors, fasting after colonoscopy, fasting C-peptide and estimated glomerular filtration rate (eGFR)—were integrated into the predictive model. The AUC for predicting hypoglycaemia was 0.78 (95% CI, 0.71–0.84) and 0.82 (95% CI, 0.74–0.90) in the development and validation cohort, respectively. Variables associated with pre-colonoscopy hypoglycaemia included SGLT2 inhibitors, fasting C-peptide and eGFR, whereas the post-colonoscopy hypoglycaemia group was associated with metformin, duration of diabetes, fasting C-peptide and fasting after the examination.

Conclusion

This study successfully developed and validated a predictive model for assessing hypoglycaemia risk in T2DM patients during peri-colonoscopy.

Implications for the Profession and/or Patient Care

Early identification of patients at high risk for peri-colonoscopy hypoglycaemia allows nurses to implement personalised preventive strategies. The predictive model enables clinical nurses to deliver tailored interventions based on individual risk factors, potentially reducing hypoglycaemia-related complications and improving patient safety outcomes.

Impact

This study provides nurses with a validated risk prediction tool for identifying high-risk type 2 diabetes patients during colonoscopy, enabling targeted blood glucose monitoring protocols and preventive interventions in clinical practice.

Reporting Method

This study follows the STROBE guidelines for reporting cohort studies.

Patient or Public Contribution

Diabetes patients contributed electronic health record datasets.