Length of stay in hospital and all-cause readmission following elective total joint replacement in elderly men

George Mnatzaganian,¹ Philip Ryan,^1,2 Paul E Norman,³ David C Davidson,⁴ Janet E Hiller<sup>1,5

1</sup>School of Population Health and Clinical Practice, Discipline of Public Health, ²Data Management and Analysis Centre, The University of Adelaide, South Australia; ³School of Surgery, University of Western Australia, Western Australia; ⁴Royal Adelaide Hospital, South Australia; ⁵Faculty of Health Sciences, Australian Catholic University, Fitzroy, Victoria, Australia

Background: We retrospectively assessed the independent effects of patient and clinical factors on length of stay and all-cause 90-day, one-year, and two-year readmission following elective total joint replacement. We also evaluated the independent association between length of stay and readmission with postoperative five-year mortality.
Methods: Longitudinal data from 819 men who had had total joint replacement were integrated with validated hospital morbidity data and mortality records. Length of stay, readmission, and mortality following total joint replacement were each modeled using multivariable proportional hazards regression.
Results: Mean age at surgery was 76.3 ± 4.3 years. Median length of stay following total joint replacement decreased by 25% from 12 days in 1997–1998 to nine days in 2005–2007. Age, inhospital complications, total knee replacement, private hospital, and increased body weight were significantly associated with longer stay. A dose-response relationship between weight and length of stay was observed (P = 0.003). Crude 90-day, one-year, and two-year readmission rates were 17.3%, 47.4%, and 65.0%, respectively. Ninety-day readmission was associated with comorbidity, inhospital complications, and treatment in public hospitals. Age, comorbidity, socioeconomic disadvantage, and increased weight significantly increased the risk of one-year and two-year readmission. Patients with a body mass index ≥ 30 kg/m2 were 34% more likely to be readmitted within two years compared with those having a body mass index of 18.5–24.9 kg/m2. All-cause one-year and two-year readmission rates were significantly higher in patients undergoing a total knee replacement than those undergoing a total hip replacement. No independent associations between length of stay and all-cause readmission or mortality were observed. All-cause 90-day and one-year readmission independently increased the risk of long-term postoperative mortality.
Conclusion: After adjusting for confounding factors, including age, comorbidity, obesity, and inhospital complications, length of stay was not independently associated with post-discharge readmission or death. We report that the obese, those having total knee replacement, and the socioeconomically disadvantaged are the highest consumers of hospital resources following elective total joint replacement in elderly men. Our findings may assist clinicians in better selection of elderly patients for surgery, and informing them about their individual level of risk.

Keywords: elderly, total joint replacement, obesity, length of stay, all-cause readmission