Veteran patients with Diabetes and/or Arterial Peripheral Vascular Disease are at significant risk for lower extremity amputation. There is no diagnostic test that can accurately or reliably predict which amputation level will heal, and what the patient specific, mortality risk, re-amputation risk or mobility outcome will be at each major amputation level. Decisions are therefore often made empirically based upon clinical experience. This has led to extreme variability in the relative frequencies of lower extremity amputation across VHA, as surgeons prioritize the risks and benefits in different ways. Cross sectional research has shown that there is a relationship between amputation level and the outcomes of mobility, re-amputation, and mortality. If an amputation is performed at the transfemoral level (TF) there is a greater risk of mortality, a greater likelihood of reduced mobility, but a very good probability tht primary healing will take place. In contrast an amputation at the transmetatarsal level (TM) is associated with very high revision rates, low mortality and for those that heal, a good mobility outcome. The outcome after transtibial (TT) amputation lies between these two extremes. These general and non-specific outcomes are likely strongly influenced by patient specific factors, but there is no evidence available at this time to guide decisions about which is the most appropriate amputation level, nor for the individual patient to provide input based upon their own priorities. This uncertainty leads to challenges for both the surgical team, and the patient. The literature indicates there is an important need for predictive models to be developed to better inform surgical recommendations and at the same time allow for patient centric priorities to be considered.
Specific Aims Statement: We will use both clinical and administrative VA databases to develop and validate two patient-specific multivariable prediction models.
Specific Aim 1 : To develop and validate a patient-specific multivariable prediction model for 1-year mortality risk after incident dysvascular transmetatarsal (TM), transtibial (TT), and transfemoral (TF) amputation.
Specific Aim 2 : To develop and validate a patient-specific multivariable prediction model that predicts 1-year ipsilateral re-amputation (same or higher amputation level) surgery risk after incident TM, TT, and TF amputation. The successful development of these predictive models will complement a mobility outcome prediction model whose development is currently funded by VA RR&D. Together they will enable surgeons to better quantify patient specific outcome probabilities associated with performing an amputation at each major lower extremity amputation level. Communication of these probabilities to the patient will enable them to provide input to the decision as they balance their personal priorities of mobility, mortality and re-amputation.
One of the major consequences of diabetes and peripheral vascular disease is lower extremity amputation. Based upon cross sectional data the choice of amputation level will have a significant impact on multiple critical outcomes including mobility, mortality, and need for re- amputation. Currently there is inadequate scientific data that informs the surgeon or individual patients about the probability of each of these outcomes. This is a significant contributor to the variability in the relative frequency of various amputation levels in different regions of the VA Health Care System. This study proposes to develop patient specific predictive models of mortality and re-amputation risk in Veterans. The long term goal is to enhance surgical decision making and patient participation in decision making regarding this complex and life altering surgical procedure.
