Background
Diabetic foot infections (DFIs) are the most common cause of hospitalization for patients with diabetes. Studies have shown diabetic patients have high readmission rates. It is important to identify variables that contribute to readmission. This study aimed to investigate clinical variables associated with 30-day hospital readmission in patients with DFI.
Methods
We conducted a retrospective study of adults admitted to the hospital for DFI between July 1, 2012, and July 1, 2015. We identified patients by International Classification of Diseases, Ninth Revision codes and randomly selected 35% of medical records for review. Patients were excluded if they did not have a DFI by review, were pregnant, or were incarcerated. The primary outcome was 30-day readmission. Data collected included baseline demographics, medical comorbidities, substance abuse, homelessness, tobacco use, and laboratory and surgical pathology data. Univariate and multivariate logistic regression models were used to identify independent predictors.
Results
Of 140 included patients, 106 (76%) were male. Median age was 55 years and length of stay (LOS) was 7 days. In univariate analysis, 31 patients (22%) were readmitted in the 30 days after the index hospitalization. Factors associated with readmission included treatment failure, elevated C-reactive protein level, and hospital LOS (P < .05). In multivariate analyses, LOS and treatment failure were independent predictors of readmission.
Conclusions
The 30-day readmission rate for patients with DFI is high. Treatment failure, C-reactive protein, and LOS are independently associated with readmission. More work is needed to determine reasons for readmission so that appropriate measures can be taken before discharge.
Clinical predictors of treatment failure for diabetic foot infections: data from a prospective trial