Knee arthroplasty: hospital length of stay and co-morbidity

Introduction Total knee replacement (TKR) is one of the most common orthopaedic procedures performed in Italy with an increase practice of 11 % from 2001 to 2009. The current need is therefore to reduce hospital length of stay in order to meet the increased demand and lower the costs of hospitalization. Materials and methods Post-operative outcomes of patients who underwent TKR at the 1st Orthopaedic Division of Pisa University between April 2008 and September 2009 were evaluated retrospectively. The sample consisted of 100 patients with a mean age of 70 years. All patients diagnosed with knee osteoarthritis undertaking a primary TKR were included in the study. Referring to fast-track protocols we have considered: age, sex, BMI, co-morbidities and ASA score, number of blood units transfused, post-operative analgesia protocol, physiotherapy protocol, day of the week when surgical procedure was performed, surgical access route. Each parameter was correlated with length of stay in order to identify whether, and how it would influence the post-operative period. The physiotherapy protocol for all patients began the first post-operative day. Post-operative analgesia was based on the use of ketorolac and morphine in elastomeric pump for 48 h, with paracetamol as rescue treatment. The surgical route was Midvastus in all patients and all surgical procedures were performed by the same operator. Results In our series (2008–2009), the average length of hospital stay was 9.25 days. Among analyzed factors, the following resulted to be predictive of length of stay (p\0.05): ASA score C3 (+1.5 days), BMI C 30 (+2.3 days), need for blood transfusion C2 (+3.26 days), age C75 (+2.12 days). Discussion Bleeding was shown to be the most important amendable variable that if controlled could lead to a significant reduction in length of stay. Obesity and consequently reduced mobilization could be addressed with a more aggressive and early rehabilitation protocol. ASA and age which had a significant impact with length of stay are not modifiable. Conclusions The identification of the variables influencing length of stay is important to reduce costs and deliver a better care for patients. In view of the results of this audit, we have monitored more closely bleeding by introducing the use of the tranexamic acid and promotingearly mobilization in obese patients, in agreement with the most recent literature. The preliminary results obtained, following this new approach, appear to be satisfactory (TKR mean hospital length of stay reduced of 2 days).