Related Experiment Videos
Age-Adjusted Charlson Comorbidity Index Outperforms Other Comorbidity Measures in Predicting 90-Day Readmissions
Shang-Wen Tsai1, Cheng-Yang Chang1, Yueh-Ting Hung1
1Department of Orthopaedics and Traumatology, Taipei Veterans General Hospital, Taipei, Taiwan; Department of Orthopaedics, School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan.
Background:
Simultaneous bilateral total knee arthroplasty (TKA) is associated with increased perioperative risk compared with unilateral procedures, making accurate risk stratification essential. Several comorbidity indices are widely used in arthroplasty; however, their comparative performance in simultaneous bilateral TKA remains unclear. This study aimed to compare the predictive ability of four commonly used indices, the age-adjusted Charlson Comorbidity Index (aCCI), the Elixhauser Comorbidity Index (ECI), the 5-factor modified Frailty Index (mFI-5), and the health-related quality of life comorbidity index (HRQoL-CI), for postoperative outcomes following simultaneous bilateral TKA.
Methods:
This study included 1,785 patients who underwent simultaneous bilateral TKA between 2010 and 2020. The cohort had a mean age of 71 years (range, 51 to 91), 81.0% were women, and the mean body mass index was 28.2 (range, 17.1 to 42.4). The primary outcomes were 90-day readmissions, including medical, surgical, and surgery-related infection complications. The secondary outcomes included 1-year reoperations and prolonged lengths of stay (≥ eight days). Predictive performance of each index was assessed using receiver operating characteristic analyses, with area under the curve (AUC) values compared using DeLong tests.
Results:
The aCCI demonstrated the highest discriminative ability for overall readmissions (AUC: 0.778), outperforming ECI (AUC: 0.680), mFI-5 (AUC: 0.654), and HRQoL-CI (AUC: 0.698). All indices showed excellent performance for predicting medical readmissions (AUC range, 0.806 to 0.843). For surgical and surgery-related infection readmissions, aCCI consistently demonstrated superior discrimination, with higher AUC values compared with other indices. For 1-year reoperations and prolonged lengths of stay, all indices showed poor to acceptable predictive ability.
Conclusions:
Among commonly used comorbidity indices, aCCI demonstrated the most clinically predictive performance for 90-day readmissions following simultaneous bilateral TKA, particularly for surgical and surgery-related infection complications. These findings support the use of aCCI as a practical tool for perioperative risk stratification in this population.