Objective. To compare regional body composition, bone mineral density (BMD), and clinical outcome in patients with two different shoulder arthroplasty designs. Materials and Methods. This cross-sectional study included 54 patients with a total of 63 shoulder arthroplasties. There were 18 men and 45 women with a mean age of 68.9 years SD ± 10.5. Mean follow-up time was 39.2 months SD ± 14.4. The patients were divided into three groups according to their history: 22 patients were diagnosed with a proximal humeral fracture and treated with a stemmed hemi arthroplasty, 11 patients were diagnosed with osteoarthritis and treated with a stemmed hemi arthroplasty, and 30 patients were diagnosed with osteoarthritis and treated with a resurfacing arthroplasty. All patients underwent a one-day protocol: Regional Dual X-ray Absorptiometri (DXA) was used to measure BMD of the distal third of humerus and regional body composition of the upper arm. The clinical outcome was measured using Western Ontario Osteoarthritis of the Shoulder index (WOOS) and Constant-Murley score. Results. All three groups were comparable regarding demographic data. Mean BMD of the distal third of humerus was 1.029 g/cm2 SD ± 0.204, mean tissue mass of the upper arm was 2.6 kg SD ± 0.7 and mean muscle mass was 1.4 kg SD ± 0.6. BMD of the distal third of humerus, tissue mass and muscle mass were significantly higher in the group diagnosed with osteoarthritis and treated with a resurfacing arthroplasty compared to the group diagnosed with a proximal humeral fracture and treated with a stemmed hemi arthroplasty, P = 0.03, P = 0.01 and P = 0.02 respectively. Median Constant score was 45.0, range 6–89, median WOOS 633, range 28–1824, and median strength 7.0 units, range 0–25. There were no significant differences between the three groups. Nevertheless, there was a trend towards a higher Constant-Murley score and muscle strength in the group of patients diagnosed with osteoarthritis and treated with a resurfacing arthroplasty compared to the group of patients diagnosed with a proximal humeral fracture and treated with a stemmed hemi arthroplasty. Conclusion. After shoulder arthroplasty BMD of the distal third of humerus and body composition are apparently more dependent on diagnosis rather than arthroplasty design. However, a larger number of patients diagnosed with osteoarthritis and treated with a stemmed hemi arthroplasty are needed to support this

Introduction. Obesity has been associated with higher complication rates and poorer outcomes following joint replacement surgery. Body mass index (BMI) is a simple index of body composition and forms part of preoperative assessment. It does not take into account the proportion of lean mass and body fat and can give a false impression of body composition in healthy manual workers. A more accurate measure of body composition is available using non-invasive bioimpedance methods. This study aims to identify whether BMI provides an accurate measure of body fat composition in patients awaiting lower limb arthroplasty surgery. Methods. Consecutive patients attending for pre-assessment clinic prior to total knee and hip replacement surgery were examined. All patients had their BMI calculated and underwent bioimpedance testing using a bedside Bodystat 1500 scanner (Bodystat, UK). Results. 83 patients (28 male) were included. Mean age was 68 years (range 16 to 92). All were awaiting lower limb arthroplasty surgery (39 primary total hip replacement, 4 revision total hip replacement, 38 primary total knee replacement, one unicompartmental knee replacement and one patellofemoral joint replacement). Mean BMI was 30.8 (range 20.8 to 48.9). Mean body fat percentage was 37.4% (range 17% to 53.9%). A weak correlation was seen between the calculated BMI and the measured body fat percentage (r=0.42, Pearson's correlation coefficient). Mean body fat percentage in obese patients (BMI > 30; mean BMI 34.9; n=42) was 42% while in the non-obese patients (BMI < 30; mean BMI 26.6; n=41) was 32.8%. This difference was significant (p<0.001). Conclusion. In patients undergoing lower limb arthroplasty the calculated BMI has a weak correlation with the measured body fat percentage. Bedside, non-invasive bioimpedance analysis provides a quick and accurate measure of body composition and can be used during preoperative assessment. Future correlation of outcome against body composition and BMI will validate the use of body composition in these patients. Care should be taken when relying on BMI alone to assess body fat composition