C-reactive protein-to-albumin ratio improves prediction of 6-month major adverse limb events in diabetic foot ulcers: a retrospective cohort study

Introduction Diabetic foot ulcers (DFUs) are associated with substantial risks of infection, amputation, and death; however, conventional anatomical classification systems do not fully capture the systemic inflammatory and nutritional status that may influence prognosis. We aimed to evaluate the prognostic value of the C-reactive protein-to-albumin ratio (CAR) and to develop a nomogram for predicting 6-month major adverse limb events (MALE) in hospitalized patients with DFUs. Methods We conducted a retrospective cohort study of patients treated at a regional referral center in Northwest China between January 2020 and January 2025. Among 321 screened patients, 139 with complete data were included in the final analysis. MALE was defined as major amputation, unplanned limb revascularization, or death related to foot deterioration or progression of sepsis. Multivariable logistic regression was used to identify independent predictors. Model performance was assessed by discrimination, calibration, bootstrap internal validation, the Brier score, integrated discrimination improvement, net reclassification improvement, and decision curve analysis. Results During follow-up, 43 patients (31.0%) experienced MALE. CAR (adjusted odds ratio 1.22, 95% confidence interval 1.03–1.48; P = 0.033) and log-transformed ulcer area (adjusted odds ratio 2.43, 95% confidence interval 1.52–4.10; P 0.001) were independently associated with MALE. The model including CAR achieved an area under the receiver operating characteristic curve of 0.852, compared with 0.827 for the base model, although the difference was not statistically significant by DeLong testing. Adding CAR improved the Brier score and yielded a positive integrated discrimination improvement. Bootstrap validation showed an optimism-corrected C-index of 0.831, and calibration remained acceptable. Discussion CAR was independently associated with 6-month MALE in patients with DFUs. A nomogram incorporating CAR and conventional clinical variables demonstrated good discrimination, calibration, and internal validity, with modest incremental improvement in risk stratification. External validation is warranted before broader clinical application.