SRS283 - Preoperative prognosticators of length of hospital stay following total shoulder arthroplasty: a systematic review

Abstract Background Length of stay (LOS) after total shoulder arthroplasty (TSA) is a key marker of recovery and resource utilisation. Predictors of prolonged LOS remain incompletely defined. This review aimed to identify preoperative prognosticators of LOS following TSA. Methods Systematic searches of major databases identified 988 records. 141 studies were included, encompassing >1.1 million TSA patients. Studies reporting LOS in relation to preoperative demographic, clinical, biochemical, psychosocial, or hospital-level variables were eligible. Results Prolonged LOS was consistently associated with advanced age, frailty, and higher ASA grade, with octogenarians staying longer than younger patients (3 versus 2 days, P < 0.001). Nutritional and haematological factors had strong effects: anaemia increased LOS >2 days from 13.8% to 25.6%, malnutrition (GNRI) raised extended LOS from 13.0 to 41.6%, and hypoalbuminaemia more than doubled risk (40.4% versus 15.9%). Comorbidities including morbid obesity, chronic kidney disease, COPD, congestive heart failure, and Parkinson’s disease were significant predictors. Psychosocial and socioeconomic disadvantage also prolonged stay, with depression, limited health literacy, Medicaid insurance, and lower income independently increasing LOS. Medication-related and procedural factors influenced LOS: high perioperative opioid use increased LOS stepwise (1.37 versus 2.33 days), whereas tranexamic acid modestly reduced it (1.5 versus 1.6 days). Reverse TSA, fracture indications, and prolonged operative time were also linked to longer LOS. Conclusions LOS after TSA is influenced by frailty, comorbidity burden, nutritional and psychosocial disadvantage, and surgical complexity. Several factors, including anaemia, malnutrition, hydration status, and opioid exposure, are potentially modifiable, supporting integrated preoperative optimisation and risk prediction.