Patients with Dementia Undergoing Major Limb Amputation Have Poor Outcomes
Samir K Shah1, Lingwei Xiang2, Rachel R Adler2, Clancy J Clark3, John Hsu4, Susan Mitchell5, Emily Finlayson6, Dae Hyun Kim5, Kueiyu J Lin7, Joel S Weissman2
1University of Florida, Gainesville, Gainesville, FL;2Center for Surgery and Public Health, Boston, MA;3Wake Forest Department of Surgery, Winston Salem, NC;4Massachusetts General Hospital, Boston, MA;5Beth Israel Deaconess Medical Center, Boston, MA;6University of California, San Francisco, San Francisco, CA;7Brigham and Women's Hospital, Boston, MA
Patients with dementia undergoing major limb amputation have poor outcomes
Background. Major limb amputation fundamentally affects patientsí function and independence. These impacts may be even more pronounced in patients with dementia. Despite this, there is little information on the frequency of amputation and postoperative outcomes, which is important for decision-making. As such, we sought to characterize the frequency and outcomes of major limb amputation for diabetic foot ulcers (DFUs) and chronic limb-threatening ischemia (CLTI) and to compare outcomes in patients with and without dementia.
Methods. We analyzed Medicare fee-for-service claims from 01/01/2016 to 12/31/2020 in beneficiaries 66 years or older undergoing amputations at or proximal to the ankle for DFUs and CLTI. We used generalized estimating equations logistic or linear regression as appropriate. The dementia codes we used have been validated using chart review. All analyses were performed using SAS 9.4.
Results. We compared 11,879 patients with dementia against 37,231 patients without dementia. Patients with dementia were older (79.7 vs. 73.8, y, p<.0001), more likely to be female (43.6% vs. 33.4%, p<.0001), and more likely to be non-white Hispanic or Black (41.9% vs. 27.8%, p<.0001) (Table). Most patients were community-dwelling (82.8% overall). Patients with dementia underwent more above-knee amputations (AKAs) compared to those without dementia (58.9% vs 39.0%). Patients with dementia were more likely to have urgent/emergent cases compared to those without (68.2% vs 62.6%, p<.0001). Multivariable analysis showed that mortality at 30 days (15.7%) and 1 year (54.3%) were worse in patients with dementia (Figure). The likelihoods of major complications (22.8%) and amputation revision (17.5%) were better in those with dementia and 30-day readmission (28.6%) was similar between the groups.
Conclusion. Major limb amputation for DFUs and CLTI is common among patients living with dementia. Patients with dementia are more likely to undergo urgent procedures and AKAs. Outcomes are sobering (for example, >50% mortality at one year). Outcomes compared to those without dementia are nuanced and include a higher 30-day and 1-year mortality but lower rates of complications and amputation revision and similar 30-day readmission. These are important data that can support goal-concordant decision-making. Additional work is needed to measure patient- and caregiver-centered outcomes to support decision-making.
|Table. Patient and Procedural Characteristics by Dementia Status|
|Feature, No. (%)||Overall(N=49,110)||No Dementia(N=37,231, 76%)||Dementia(N=11,879, 24%)||P-value|
|Age, years (median)||75.0||73.8||79.7||<.0001|
|-Asian or Pacific Islander||1.2%||1.1%||1.4%|
|Elixhauser Index (median)||16||15||20||<.0001|
|Congest. Heart Failure||48.3%||48.3%||48.4||0.98|
|Chronic Kidney Disease Stage V/Dialysis||29.9%||31.2%||26.1%||<.0001|
|Hospital (e.g., transfer)||9.7||10.1||8.7||<.0001|
Back to 2024 Abstracts