One year after a below-knee amputation, one in four patients has died

Design

national population study using mortality-linked Hospital Episode Statistics, 25 years

Population

25,176 people with diabetes undergoing 28,232 below-knee amputations in England

Primary outcome

Kaplan-Meier all-cause mortality

Effect

23.5% at 1 year, 60.0% at 5 years, 82.3% at 10 years; no improvement since 2012

Mortality-linked Hospital Episode Statistics covering 25 years of English practice captured 25,176 people with diabetes who underwent 28,232 below-knee amputations. Kaplan-Meier all-cause mortality was 23.5% at one year, 60.0% at five years and 82.3% at ten. None of that has improved since 2012. Older age (HR 1.62, 95% CI 1.56-1.69), greater comorbidity (HR 1.04 per Charlson point) and female sex (HR 1.06, 95% CI 1.03-1.10) each independently marked higher risk.

These numbers change the conversation that happens before the operation. A below-knee amputation is routinely framed as a limb-salvage failure with a rehabilitation pathway attached. On this data it is closer in prognosis to an advanced malignancy, and the patient in front of you deserves to hear the real figure rather than a reassurance about prosthetic fitting. The absence of any improvement over fourteen years also tells you the gain is not going to come from better surgery.

In practice that means two things. The referral for a threatened foot is an urgent medical event, not a surgical waiting-list item, and the months after the operation are a cardiovascular and renal problem as much as a wound problem. Ninety-day myocardial infarction ran at 3.74% and readmission at 20.25%. Contralateral amputation reached 6.43% (95% CI 6.14-6.73) at five years, so the remaining foot needs a named follow-up plan before discharge.

• Give the one-year figure honestly when consenting for a below-knee amputation, alongside the rehabilitation plan.
• Treat the 90 days after surgery as a high cardiovascular risk window — check statin, antiplatelet and blood pressure before discharge, not at the next clinic.
• Book surveillance for the contralateral foot at discharge; it is the limb most likely to be lost next.
• Record Charlson comorbidity burden — it identifies who most needs early palliative and social input.
• Escalate the threatened foot to a multidisciplinary team early; the mortality signal here is about how late people arrive, not about the operation.

Why it matters

A below-knee amputation is being consented for as a limb operation when its prognosis is closer to that of an advanced cancer.

Don't overread it

This is observational coded hospital data — it identifies who dies, not why the operation fails to help.

The statistics, in plain English

These are observed survival curves from a whole-population dataset, not estimates from a model, so the mortality figures are as solid as the coding behind them. The hazard ratios come from a Cox model and describe association, not cause: female sex at HR 1.06 (95% CI 1.03-1.10) is statistically clear only because the cohort is very large, and an excess of that size is not something to act on for an individual patient. Age at HR 1.62 is the one carrying real weight.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.