Determining Predictors for Unsuccessful Total Knee Arthroplasty Outcomes

Background: Total knee arthroplasty (TKA) is the second most commonly performed orthopedic procedure in the United States (US). However, decreased functional ability and chronic pain remains a significant problem for 6% - 30% of patients after TKA. An understanding of factors affecting TKA outcomes may lead to better methods of patient management and selection, which would improve surgical outcomes. Therefore, this quality improvement (QI) project analyzed whether pre-surgery Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) sub scores for pain, stiffness, and function are reliable predictors of total WOMAC scores six months post-surgery. Objective: The purpose of this QI project was to identify factors associated with post-TKA outcomes in patients at Banner – UMC’s orthopedic clinic in Arizona during a retrospective five-year period (2010-2015) Methods: This QI project analyzed data collected from 131 patients, 63.4% (n=83) were female. Average age was 65 years (range= 40-89). Patient’s WOMAC scores were assessed pre- and post-surgery.WOMAC scores range from ‘0’ to ‘96’ with higher scores representing worse pain, stiffness, and functional ability. Unsuccessful TKA was defined as a post-surgery total WOMAC score greater than 24.3. Data were analyzed by multiple linear regression using IBM Statistical Package for the Social Sciences statistical analysis software version 25. Results: Patients’ average WOMAC scores pre-surgery and post surgery were 48.56 + 16.898 and 15.24 + 15.210 respectively. Twenty-six of the 131 patients in this QI project had a post-surgery total WOMAC score greater than 24.3, indicating that 20% of patients had unsuccessful TKAs. This regression model predicted 7.6% (R2 =.076) of the variance in total WOMAC scores six months after TKA and was statistically significant at alpha = .05 (F1,125 =9.923, p=.002). None of the other variables in the model were significantly associated with total WOMAC scores six months after TKA. Conclusions: The results of this QI project support previous studies indicating pre-surgery pain as a significant predictor of TKA outcomes, though it accounted for only a small percent of the variance. Further research is needed to determine if other factors such as body mass index and preexisting co-morbidities are more predictive of TKA outcomes. Also, objective methods for assessing pre- and post-TKA outcomes may be more reliable than the subjective WOMAC scores obtained