Do Patient Risk Factors Impact 90-Day Readmission after Total Ankle Arthroplasty?

Category: Ankle,Ankle Arthritis Introduction/Purpose: The Comprehensive Care for Joint Replacement model (CJR) provides bundled payments for in-hospital and 90-day post-discharge care of patients undergoing lower extremity joint replacement including hip, knee, and ankle arthroplasty (THA, TKA, and TAA). Pre-operative risk factors influencing in-hospital and post-discharge costs are, thus, of keen interest. While THA and TKA have been reported to have a 5.3% 90-day readmission rate associated with race, gender, increased BMI, >2 medical comorbidities, increased length of stay, and discharge to inpatient rehab, little is known about factors that influence readmission rates after TAA. The purpose of this study is to identify risk factors associated with 90-day readmission after TAA. Methods: 1,048 patients undergoing TAA (ICD-9 81.56 or ICD-10 0SRF/G) at a single academic institution were prospectively enrolled into an ongoing, IRB-approved longitudinal TAR outcome study between 2007 and 2016. Records were retrospectively reviewed to determine patient, operative, and post-operative characteristics including age, gender, race, risk factors of the Charlson-Deyo comorbidity and Elixhauser indices, post-discharge disposition, BMI, length of stay, and ASA score. Pre-operative Elixhauser and Charlson-Deyo comorbidities were recorded using standardized ICD-9 and ICD-10 codes. Univariate tests of significance (t-tests for continuous inputs and chi-square tests for categorical inputs) were performed to determine the potential relationship between patient characteristics and 90-day readmission using JMP Pro version 13.0.0. The tables display pre-operative cohort-level and outcome-specific patient characteristics as well as the results of significance testing for comorbidities with >1% prevalence. Results: Thirty of 1048 (2.9%) patients were readmitted after TAA during the 90 day post-discharge window. Twenty-two (73%) of the patients were readmitted for surgical wound complication. The majority of the remaining 8 admissions were for medical illnesses not clearly related to the index procedure. Prevalent comorbidities included hypertension, cardiac arrhythmias, depression, obesity, rheumatoid arthritis, diabetes, hypothyroidism, and chronic obstructive pulmonary disease. However, there were no significant differences in patient characteristics between those who were readmitted and those who were not readmitted although patients that were readmitted tended to be slightly older, were less likely to be discharged to SNF or in-hospital rehabilitation, and had higher ASA score and Charlson-Deyo comorbidity index. No individual patient comorbidities were statistically associated with 90-day readmission. Conclusion: The 90-day readmission rate of 2.9% after TAA at our institution is lower than reported rates for THA and TKA nationally (5.3%). Although our patient population had a similar prevalence of risk factors when compared to THA/TKA patients, none of these factors were significantly associated with 90-day readmission. These data suggest that grouping TAA with THA and TKA for CJR may not be advisable. In an emerging era of bundled payments, further work is needed to delineate factors strongly associated with costly readmissions specific to surgical treatment and individualized based on pre-operative patient profile.