Smokers undergoing Total Knee Arthroplasty have Increased Early Morbidity, Mortality and Hospital Costs

Introduction: Smoking is a well-documented modifiable risk factor associated with increased complications following total knee arthroplasty (TKA). There are no recent studies quantifying the early inpatient complication rate and costs associated with smokers and TKA. Given a national push towards bundled payment models, the purpose of this study was to examine the effect of smoking on early inpatient complications and cost following elective TKA in the United States.

Methods: Using the Nationwide Inpatient Sample (NIS) in 2014, all primary elective TKA admissions were identified. Patients were stratified by smoking status through a secondary diagnosis of ‘tobacco use disorder’. Patient characteristics as well as rates of complications and mortality were compared with use of univariable and multivariable analyses.

Results: There were 125,181 primary TKAs identified in the NIS, corresponding to a national estimate of 625,906 cases. Of this group, 19.2% were smokers. Mean age was similar (65.1 vs 66.0 years) for smokers and non-smokers, but smokers were statistically slightly younger (p<0.001). Hospital resources were significantly higher for smokers, with length of stay 3.2 versus 2.7 days, and hospital cost $16,496 versus 15,453 (p<0.001). A multi-variable logistic model adjusting for age, gender, and co-morbidities, showed smokers were at increased likelihood for inpatient mortality (OR 5.2, 95%CI 2.6-10.3), acute renal failure (OR 2.9, 95%CI 2.6-3.3), deep venous thrombosis (OR 2.9, 95%CI 2.3-3.6), urinary tract infection (OR 2.4, 95%CI 2.2-2.7), transfusion (OR 1.5, 95%CI 1.4-1.6), and discharge to a skilled nursing facility (OR 1.2, 95%CI 1.1-1.2).

Discussion and Conclusion: Smoking continues to be common among patients undergoing elective TKA and is associated with higher hospital costs as well as higher rates of immediate inpatient complications. These findings are critical for risk stratification as well as modeling of bundled payment models. Furthermore, this data will allow the orthopaedic community to improve preoperative patient education and management regarding smoking and TKA.