Glycemic variability, ulcer severity, and 30-day post-discharge emergency department revisits in geriatric patients with diabetic foot ulcers: A single-center retrospective cohort study

Aim: The prognostic value of continuous glucose monitoring (CGM)-derived glycemic variability (GV) in inpatients with diabetic foot ulcers (DFUs) is uncertain.

Materials and Methods: This retrospective cohort included 114 patients aged ≥ 65 years with type 2 diabetes and a DFU. CGM metrics were mean glucose, standard deviation (SD), glucose coefficient of variation (GluCV), and high and low blood glucose indices (HBGI/LBGI). Wagner grade was categorized as moderate (1–2) or high (3–5). Outcomes were 30-day emergency department (ED) revisit, sepsis, intensive care unit (ICU) admission, length of stay, and mortality. Adjusted models included age, sex, diabetes duration, glycated hemoglobin (HbA1c), GluCV, Wagner grade, peripheral arterial disease (PAD), and hypertension.

Results: Mean age was 76.9 years and 44.7% were male. ED revisit occurred in 43.9%, sepsis in 57.0%, ICU admission in 56.1%, and 30-day mortality in 46.5%. Mean glucose was higher with sepsis (198 vs 185 mg/dL; p = 0.037). High Wagner grade showed non-significant associations with mortality (adjusted odds ratio (aOR) 1.91; p = 0.11) and ICU admission (aOR 1.93; p = 0.11). Male sex was associated with lower ED revisit odds (aOR 0.41; p = 0.029), but this was the only nominally significant adjusted association and was not corrected for multiple testing. HbA1c and GluCV were not independently associated with any outcome; SD, HBGI, and LBGI showed no significant unadjusted associations.

Conclusion: Short inpatient CGM-derived GV metrics did not demonstrate incremental prognostic value. Wagner grade provided the most consistent signal, whereas the male-sex association was exploratory and requires external confirmation.