Amputation May Not Independently Predict Mortality in Diabetes

Miriam E Tucker

June 21, 2016

NEW ORLEANS — Development of a foot ulcer and peripheral arterial disease (PAD) are independent predictors of mortality in diabetes patients, whereas amputation is not, new findings suggest.

The data, from a US Veteran's hospital general medicine clinic, were presented June 11 here at the American Diabetes Association (ADA) 2016 Scientific Sessions by Edward J Boyko, MD, professor of general internal medicine and adjunct professor of epidemiology at the University of Washington, Seattle.

In the 22-year study involving 1461 male veterans with diabetes, amputation was a significant predictor of early death in univariate age-adjusted analysis but not in multivariate analysis.

"There was no association with incident amputation….This was due to the strong association between amputation and both incident foot ulcer and peripheral arterial disease," Dr Boyko explained, adding, "Cause-of-death analysis would provide further insights on these associations."

Session moderator Paul J Kim, DPM, told Medscape Medical News that the lack of significance for amputation after cofactor adjustment really isn't surprising.

"You have to think of the diabetic patient as a whole.…Amputation is simply a surgical intervention. It's not a disease. People with diabetes die of myocardial infarctions and strokes and pneumonia. Unfortunately, people have linked amputation directly to death, which is erroneous."

Mortality Predictors

The prospective study reported by Dr Boyko began in 1990 and continued through 2012, although foot-ulcer data were only available through 2002, when the general medicine clinic closed. Inclusion criteria were diabetes, having at least one foot, and being ambulatory. Patients with current foot ulcers were excluded (until healing occurred), as were those with severe physical or mental illness.

As background, Dr Boyko noted that diabetes increases the risk of PAD, neuropathy, foot ulcer, and lower-limb amputation, and all of these complications have been associated with higher mortality.

But what is not known is whether these lower-limb complications are independently associated with mortality or whether they predict mortality in the longer term, he explained.

Patients underwent visual foot examinations (for shape, deformity, onychomycosis), Semmes-Weinstein 5.07 monofilament testing, and transcutaneous oximetry (TcPO2 in mm HG at 44ºC on the dorsal foot). Other measurements, repeated every 1 to 3 years, included vision tests, routine labs, blood pressure, self-reports of a history of lower-limb ulcer and amputation prior to study entry, examination for foot ulcer through 2002, and amputation by examination through 2002 and by medical record search through 2012.

At baseline the 1461 male subjects had a mean age of 62 years, diabetes duration of 15 years, and average HbA1c of 9.6% (which was not unusual for 1990). About 20% had a prior ulcer, and 4.4% had a prior amputation. The majority (82%) were ever-smokers (also typical of the day).

During 13,039 person-years of follow-up, 1103 of the subjects died, 219 developed foot ulcers (through 2002), and 123 had lower-limb amputations (through 2012). Forty-one percent had hallux limitus, and 1.4% had Charcot foot. Average ankle-brachial index (ABI) was 0.9, and TcPO2 at 44ºC was 40.8 mm Hg. Cause-of-death information was not available.

In age-adjusted analysis, significant predictors of mortality included ABI, prior history of foot ulcer, incident foot ulcer during the study, history of amputation, and amputation during the study. Also significant were neuropathy, hallux limitus, mycotic nails, and TcPO2.

Factors found not to be significant mortality predictors were Charcot deformity, weight, systolic blood pressure, diabetes duration, HbA1c, lower-limb ischemia, tinea pedis, and cholesterol or triglyceride levels.

In multivariate analysis, only a few lower-limb–related factors remained significant independent mortality predictors, including ABI above or below the reference range 0.9 to 1.3, incident foot ulcer during the study, and TcPO2.

Prior history of foot ulcer or amputation, amputation during the study, and neuropathy were no longer significant.

Lifetime smoking history, estimated glomerular filtration rate (eGFR), and albumin remained significantly associated with mortality, however.

Quality of Life Is Key: Amputation Can Improve Patients' Condition

Part of the flaw in looking at amputation solely as a risk marker is that, in many cases, it can improve a patient's condition, Dr Kim pointed out.

He noted that there are three main indications for amputation: ischemia, uncontrollable infection, and the third one that is discussed less often, maximization of function.

"Some people do very well postamputation because [the prosthetic] is more functional than the residual limb they would have had.…It's a quality-of-life issue. What patients are really concerned about is function and mobility.…That's something we've neglected to look at very seriously."

But, he said, there is increasing interest now in assessing quality of life as part of cost-effectiveness evaluations: "I think that's where this is going, to see what the most cost-effective intervention is."

Dr Kim cautioned about extrapolating from this single-center study of a veteran population, since there are socioeconomic issues at play. However, he reiterated, "The results don't surprise me."

Neither Dr Boyko and coauthors nor Dr Kim have relevant financial relationships.

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.

American Diabetes Association 2016 Scientific Sessions; June 14, 2016; New Orleans, Louisiana. Abstract 144-OR/144


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: