PREDICTORS OF LENGTH OF STAY FOLLOWING TOTAL HIP REPLACEMENT: A MULTIVARIATE ANALYSIS OF 2302 PATIENTS

Abstract

Introduction: Increasingly, clinical pathways and fast-track protocols are reducing hospital in-patient stay following elective joint replacement surgery. In order to improve efficiency in our unit, we undertook a prospective observational study to identify pre–and peri-operative factors associated with increased length of stay.

Methods: From our prospective primary hip arthroplasty database we analysed data from 2678 consecutive patients over a 9-year period from 1998–2007. Patients were excluded who had bilateral hip replacement, died within 30 post-operative days, or had surgery for a diagnosis other than primary osteoarthritis. This left 2302 patients who were analysed using multiple logistic regression analysis

Results: Length of stay varied from 3 to 58 days, with a mean of 8.1 days, and median 7 days. After multivariate analysis, factors that were found to be significantly associated with shorter length of stay were younger age (p< 0.001), male sex (p< 0.001), more recent year of admission (p=0.008), regular non-steroidal anti-inflammatory medication (p< 0.001), lower Harris Hip Score (p< 0.001), and higher General Health Perception dimension score on SF-36 questionnaire (p< 0.001). In addition, the absence of blood transfusion during admission (p< 0.001) and absence of post-operative urinary catheter (p< 0.001) were also associated with shorter length of stay. The following factors, in particular, were not found to be significantly associated with increased length of stay: obesity, diabetes, smoking, medical comorbidity, other disabling joint condition, use of wound drain post-operatively.

Conclusions: We have identified a number of pre-operative factors that predict likely length of stay in a large cohort of patients undergoing primary hip replacement. This data could be used in the future for resource allocation and to improve efficiency in this significant area of healthcare.