Rates of LEA have been shown to increase in the USA between 2008 and 2018 among the Veteran population [7], with limited contemporary information describing the types of LEA and associated disease conditions. To close this gap and understand the disease conditions associated with LEA, the present study examined the demographic and clinical characteristics of Veterans with LEA and the diagnoses associated with amputations. The results showed that few Veterans who underwent LEA had experienced injury to lower extremities (including during combat); rather, the most prevalent diagnoses associated with LEA were chronic disease, specifically diabetes, PAD, and bacterial infections. Moreover, Veterans had high rates of other medical comorbidities, especially cardiovascular diseases. The burden of LEA remains relatively high among US Veterans; the results of this study suggest that in most cases amputation is associated with modifiable risk factors. Effective management, including preventative measures, earlier detection, and using guideline-directed medical therapy, of these medical conditions offers an opportunity to reduce, or prevent altogether, LEA in our Veteran population.

The demographic profile of the study sample was similar to that of previously studied cohorts (mean age > 65 years and mostly male) [7, 10]. The distribution of LEA type was broadly similar to other reports [7]. Only a small proportion of amputations were above the knee (15.4%) and over half (52.9%) were transmetatarsal, which are less disabling as there is better preservation of mobility and limb function with more distal amputations. The number of patients who had LEA procedures declined over the study period, from 6892 in 2019 to 4750 in 2023. This is in contrast to the increasing incidence of LEA among Veterans reported in the decade prior (by 5.23 per 10,000 persons from 2008 to 2018) [7]. One possible reason for the declining number of patients who had LEA procedures observed in our study is the disruption of healthcare services throughout the VHA during the COVID-19 pandemic (2020–2023) [11]. However, it may also reflect the shrinking size of the Veteran population in the USA [8], or be a result of the PAVE program [6]. Additional studies are needed to determine the impact of the pandemic and prevention programs on rates of and factors contributing to LEA in Veterans.

A previous study reported that on average there were 113 LEAs per year among active-duty military personnel in a 16-year period from 2001 to 2017 [12], compared with 5427 LEAs per year among the Veterans in the most recent 5-year period in our study. In addition, unlike in the active-duty personnel whose amputations were typically related to injuries sustained in combat [12, 13], fewer than 1% of procedures in our study population were associated with a diagnosis of combat-related traumatic injury, with the majority being associated with chronic medical conditions. The results of this study, along with incidence data in the active-duty population, dispel any misconceptions as to combat being the primary cause of amputations in the military community; instead, most LEAs in the military population are performed in Veterans and not active-duty service members.

Our study showed that the most common diagnoses associated with LEA were bacterial infections and chronic diseases such as diabetes (81.6%) and PAD (63.3%). Diabetes and PAD are known risk factors for LEA [7]; together, they are estimated to lead to more than half of the LEAs performed in the USA each year [14]. The increased risk of amputation from these conditions may be due to peripheral neuropathy, infection, and ulcers (diabetes) and impairment of blood flow to the extremities (PAD) [14]. Diabetes and other cardiometabolic conditions are more prevalent among Veterans than in the general population [15]. Additionally, non-coronary cardiovascular disease was observed in 76.8% of the study sample; coronary artery disease was present in 43.6%, and chronic kidney disease in 39.4%. Other notable comorbidities were cerebrovascular disease (19.4%) and peripheral neuropathy (17.5%). The high rates of chronic diseases in this study are consistent with the poor overall health of US Veterans [16] and underscores the need for better risk-factor modification strategies for this high-risk population.

The observation that chronic diseases—and not traumatic injury sustained in combat—are the more prevalent diagnoses associated with LEA demonstrates that LEA in Veterans is potentially preventable with appropriate management [2, 17, 18]. While an annual foot examination is required by the PAVE program in all patients with diabetes [19], which has been linked to lower rates of LEA in Veterans with diabetes [20], a recent analysis demonstrated more than 30% of Veterans did not receive vascular assessment in the year prior to LEA [21]. Enrollment in the PAVE program also increased adherence to guideline-recommended therapies such as antiplatelet and lipid-lowering agents in patients with a diagnosis of PAD [22]. Additionally, a dose-dependent relationship was observed between the use of statins and the risk of LEA and mortality in Veterans with PAD [23]. Recent lower extremity PAD guidelines recommend ankle-brachial index screening for those at risk for PAD and low dose anticoagulation is recommended in patients with asymptomatic PAD to prevent amputation [24]. Thus, more effective management of diabetes and, particularly, PAD, where recent guidelines increase focus on detection and prevention, in Veterans can lead to improvements in their overall health and can potentially reduce their risk of LEA.

This is one of few studies to date that has investigated potential reasons for LEA among US Veterans. Besides the large sample size, a strength of this study was that it used current data (from the most recent 5 years) from the nationwide VHA database. The study had certain limitations. First, LEA and diagnoses of medical conditions were identified from the claims data using diagnosis codes, and there may have been misclassification (e.g., of LEA types) based solely on diagnosis codes and limitation of using diagnosis codes that are not readily used (e.g., combat injury-related codes). Similarly, for patients with multiple LEA procedures, the LEA type was identified on the basis of the highest level of amputation in the 30 days post index, which could have led to misclassification if a higher level of LEA occurred outside of the 30-day window. Second, as the observation period was limited to 12 months before and 30 days after the index LEA procedure, the study did not include Veterans with less than 12 months of continuous enrollment in the VHA or Veterans who may have died during index LEA procedure. The trends observed may not be reflective of entire Veteran population. However, as most Veterans remain in the VHA system for a long time, the proportion who did not meet this inclusion criterion was relatively small (3%). This study focused on amputations of lower extremities across all severity levels and age groups. Future studies that focus on a major amputation subgroup or stratified by age groups may be warranted to shed more insight on the heterogeneity of the risk profile for this population.