Lower Extremity Amputation Shown to Predict Premature Mortality

Researchers found lower extremity amputation was a robust independent predictor of mortality in a study of patients with diabetes in a managed care population with access to high quality medical care.

The study of 6,992 patients with diabetes across 10 years showed the association of lower extremity amputation (LEA) to early mortality was strongest in men and differed by race, with Hispanic patients at a high risk. The results are part of Translating Research into Action for Diabetes (TRIAD), a prospective observational study of care for diabetic patients within managed care organizations. The researchers determined foot complications through administrative claims data, and searched for deaths across 10 years of follow-up via the National Death Index.

Foot conditions and mortality

“The purpose of our study was to assess [whether] diabetes-related foot complications remain a significant predictor of death after adjusting for other known risk factors for death, including cardiovascular disease,” Laura N. McEwen, PhD, MPH, senior epidemiologist in the Department of Internal Medicine, Division of Metabolism, Endocrinology and Diabetes at the University of Michigan, told O&P News. McEwen and colleagues wanted to examine the association of three specific foot conditions — Charcot’s neuro-osteoarthropathy, diabetic foot ulceration and lower extremity amputation — with early mortality.

“These are the most common diabetes-related foot complications that have been described to be associated with premature mortality,” according to McEwen, but previous studies had not determined the independent contribution of each complication to all-cause mortality in patients with diabetes.

Laura N. McEwen, PhD, MPH
Laura N. McEwen

The purpose of TRIAD was to find out whether health plan organization and structure can impact the results of diabetes care.

“The working hypothesis was that access to high quality care would result in improved cardiovascular risk factors and cardiovascular disease and potentially reduce any associations between foot complications and death,” McEwen said.

The researchers further hypothesized lower extremity amputation would remain a risk factor for mortality, but that race and ethnic differences in lower extremity amputation risk would be reduced due to access to high quality medical care available through managed care.

Predictors and variables

Using 1999 to 2003 health plan administrative data, McEwen and colleagues studied the 3-year period of prevalence for three potential mortality predictors: Charcot’s neuro-osteoarthropathy; diabetic foot ulceration; and lower extremity amputation. Minor lower extremity amputation was defined as toe or transmetatarsal amputation. Major lower amputation was described as any below-knee or above-knee amputation.

The researchers also assessed demographic variables as possible effect modifiers and confounders, including age, sex, race/ethnicity; education and income; diabetes-specific variables, including type of diabetes, duration of diabetes, treatment for diabetes and glycated hemoglobin level; and other clinical variables, including BMI, systolic blood pressure, low-density lipoprotein cholesterol level, smoking status, history of microvascular and macrovascular complications, and Charlson comorbidity complex.

The researchers constructed a series of Cox proportional hazards models for each of the three primary predictors of all-cause mortality, including unadjusted models to determine whether each predictor was associated with the outcome and fully adjusted models, including covariates selected for their known association with foot complications. The results were stratified by age, race/ethnicity and smoking status in a sensitivity analysis. The researchers also conducted sensitivity analyses to evaluate any lower extremity amputation as a predictor of mortality, regardless of Charcot’s neuro-osteoarthropathy or diabetic foot ulcer status, and evaluation of major and minor lower extremity amputation as predictors of mortality.


Results for the 6,992 patients showed between 1999 and 2003, 55 patients (1%) had a diagnosis of Charcot’s neuro-osteoarthropathy; 205 (3%) had a diagnosis of at least one diabetic foot ulcer with active debridement; and 101 (1%) underwent lower extremity amputation. Overall, 6,631 patients (95%) had none of these foot conditions.

After 10 years of follow-up, 1,956 participants had died, with a mean ± standard deviation (SD) time until death of 7.7 years ± 2.3 years. Those who died were more likely to be older, male, of non-Hispanic white race/ethnicity and have a lower educational level and income. They were more likely to have type 2 diabetes, longer duration of diabetes and to be treated with insulin. They also were more likely to have a lower BMI; to have higher systolic blood pressure; to be smokers; and to have histories of retinopathy, nephropathy, neuropathy, hypertension, coronary heart disease, congestive heart failure and peripheral heart disease. In addition, they were more likely to have higher Charlson indices.


The results showed an 84% higher risk of all-cause mortality for those with lower extremity amputation and no other foot complications after adjusting for demographic and clinical variables (hazard ratio [HR] 1.84). Further, the risk of mortality associated with lower extremity amputation was significant for every age group and was significant in men (HR 1.96), but not in women (HR 1.05). Risk of mortality associated with lower extremity amputation was highest for Hispanic patients (HR 5.17), followed by non-Hispanic white patients (HR 2.18).

Unadjusted all-cause mortality rates were higher for patients with Charcot’s neuro-osteoarthropathy and diabetic foot ulcer, but the associations did not remain statistically significant after adjustment.

“We were not entirely surprised that diabetes-related amputation remained in the mortality models; however, we were surprised that foot ulcer and Charcot neuroarthropathy did not remain,” McEwen said.

Among the 1,956 people who died within the 10-year follow-up period, 72 (4%) had lower extremity amputation. Of the 5,036 people who remained alive, 36 (1%) had lower extremity amputation. People with lower extremity amputation had a 5-year mortality of 39% and a 10-year mortality of 67%, in comparison to a 5-year mortality of 15% and 10-year mortality of 27% for people without lower extremity amputation. The researchers found an association between lower extremity amputation and mortality remained statistically significant in fully adjusted models (HR 1.67). Minor amputations were not a significant predictor of mortality (HR 1.48), but major amputations were (HR 1.89). However, survival at 10 years for people who had a minor lower extremity amputation was similar to patients with a major lower extremity amputation.

Specialized care

“We know that approximately 50% of patients with diabetes undergoing lower extremity amputation will ultimately be habitual prosthetic users,” McEwen said. “This will increase their likelihood for remaining out of their home environment in extended care facilities. This may put them at risk for premature mortality, despite aggressive management of cardiovascular risk factors.”

According to McEwen, health care practitioners should urge diabetic patients to aggressively manage their cardiovascular risk factors and pay attention to patients’ feet to identify risk early and manage it appropriately. They also should be aware of the high risk for Hispanic patients, which was not offset by high quality care as hypothesized.

“There may be a need to target patients with Hispanic ethnicity and their caregivers, with culturally sensitive educational materials,” McEwen said. In addition, “Further study should focus on elucidating specific areas that describe increased rate of mortality in the Hispanic population and test interventions to impact the increased risk.” – by Amanda Alexander

Disclosure: McEwen reports no relevant financial disclosures.

Leave a Reply

Your email address will not be published.