The ability to identify those at risk for longer inpatient stay helps providers with postoperative planning and patient expectations. Decreasing length of stay (LOS) in the future will be determined by appropriate patient selection, risk stratification, and preoperative patient optimization. The purpose of this study was to identify factors that place patients at risk for extended postoperative LOSs.
The study cohort included 2009 primary total knee arthroplasty (TKA) patients and 905 total hip arthroplasty (THA) patients. Patient comorbidities were prospectively identified and the LOS for each patient was tracked after a primary arthroplasty. Statistical analysis was performed to correlate which comorbidities were associated with longer inpatient stays.
In the TKA population, gender, smoking status, venous thromboembolism history, body mass index, and diabetes status were not found to be a significant predictor for the LOS. Age was found to be a factor in univariate regression testing (P < .001). In the THA population, univariate testing showed female gender (P < .001), smoking status (P = .002), and age (P < .001) to be factors, but like the TKA population, venous thromboembolism history or diabetes status was not significant. In THA multivariate analysis, age (P < .001) and female gender (P = .018) continued to be factors, but smoking was determined to be a confounding variable.
Age and gender were associated with a longer LOS after THA, whereas only age was a significant factor after TKA. Development of age-adjusted LOS models may help aid patient expectations and risk management.