Risk Analysis Index Compared with Modified Frailty Index for Lower Extremity Amputation

Frailty can be used as a preoperative indicator for those patients at higher risk for morbidity and mortality. Patients undergoing major lower extremity amputation (LEA) with critical limb ischemia have relatively high short- and long-term postoperative mortality related to their medical comorbidities. The objective of this study was to compare mortality and adverse events after major LEA using frailty scoring systems Risk Analysis Index (RAI) and Modified Frailty Index-11 (mFI). A retrospective review of all patients undergoing major LEA over a 6-year period was conducted at a South Texas Veterans Affairs hospital. The mFI was scored using 11 preoperative risk factors, and RAI was calculated using a 14-point system previously published. After calculating the frailty score for both indices, the patients were separated into frail (RAI-F: 10-29, mFI-F: 0.2-0.4) or very frail (RAI-VF: >30, mFI-VF: >0.4) groups. The short-term 30-day outcomes for the frail and very frail groups were mortality, readmission, reoperation, and major adverse event; the long-term outcome was 1-year mortality. Kaplan-Meier survival curves with log-rank analysis was performed over 24 months for both indices. Receiver operating curves were used to assess predictability for 1-year mortality for each index. The RAI score divided the 298 patients undergoing LEA into 90 for the frail group and 196 for the very frail group. The mFI score had 90 patients in the frail group and 196 in the very frail group. The remainder of the patients were not frail. There was no statistically significant difference in 30-day outcomes between frail and very frail groups for mFI or RAI, although mortality for both indices trended towards a higher mortality in the very frail group. By 1 year, both RAI and mFI had a statistically significant greater mortality in the very frail group compared with the frail group (Table). In the survival analysis, the log-rank test demonstrated a statistically significant greater survival for both indices in the frail group compared with the very frail group (Fig). The receiver operating curve indicated a better predictability for 1-year mortality by C-statistic with RAI (0.701) compared with mFI (0.653) The frailty indices for major LEA demonstrated improved survival in the patients categorized as frail as compared with very frail, and RAI had better mortality predictability compared with mFI. The utilization of frailty scoring should be considered when counseling medically complicated patients on their survival before undergoing a major LEA.TableShort- and long-term outcomes with RAI and mFI for frail and very frail patientsRAI frail (n = 153), No. (%)RAI very frail (n = 111), No. (%)P valuemFI frail (n = 90), No. (%)mFI very frail (n = 196), No. (%)P value30-day mortality5 (3.3)10 (9.0).0601 (1.1)13 (6.6).07230-day readmission39 (25.5)24 (21.6).55919 (21.1)41 (20.9).99930-day unplanned operating room19 (12.4)9 (8.1).31413 (14.4)16 (8.2).13830-day adverse event58 (37.9)46 (41.4).61038 (42.2)72 (36.7).4331-year mortality25 (16.3)46 (41.4).000115 (16.7)59 (30.1).020mFI, Modified Frailty Index-11; RAI, Risk Analysis Index. Open table in a new tab