Related Experiment Video
Updated: Jun 25, 2026
Predicting Amputation using Local Circulating Mononuclear Progenitor Cells in Angioplasty-treated Patients with Critical Limb Ischemia
Published on: September 22, 2020
Short-term prediction of major lower-limb amputation based on clinical indicators on admission: a single
A H Seuc1, M López, L Rodríguez
1National Institute of Angiology and Vascular Surgery, Cuba. metodoli@infomed.sld.cu
Aim:
The aim of this study was to analyze the possibility of predicting short-term major lower-limb amputation (SMLA) in patients with vascular diagnoses, based only on clinical variables measured on admission.
Methods:
A longitudinal, retrospective cohort study of patients with peripheral vascular diagnoses admitted at our Institute was carried out. A stratified sample of 463 patients admitted during 1997, 2000 and 2003, was studied. Logistic regression was used to identify significant predictors of amputation; twelve clinical variables measured on admission were considered as potential predictors.
Results:
Of the 463 patients, 93 (20%) were amputated. Significant predictors of amputation identified by the logistic regression analysis were ''type of lesion'' (none; neuro-infectious; ischemic; mixture), ''initial diagnosis'' (phlebolymphopathies, acute arterial insufficiency, chronic arterial insufficiency, diabetic foot, others), ''plantar region lesion'' (no; yes), ''diabetes'' (no; yes), ''number of toes affected'' (none; 1-2; 3 or more), and ''area of leg affected'' (none; lower third; + lower third). More than 80% of patients were correctly classified with the final model: sensitivity was 42% and specificity 96%.
Conclusions:
It seems that SMLA in patients with vascular diagnoses can be predicted reasonably well using as predictors only clinical variables measured on admission. This is a potentially useful result for Angiology Services located in developing/poor communities. The amputation probability for each patient obtained from the logistic regression model can be used in several ways: 1) the medical care of patients can be customized so that the amputation rate of the whole Service can be reduced, and 2) the amputation probability of the statistical model can be used as an estimation of the severity of the disease in each patient, which in turn can be used to standardize the amputation rates computed on different years; this would allow a better assessment of the Institutional performance over time.