Lower limb fractures are commonly treated with cast immobilization, and as a main consequence of strict immobilization this typically leads to loss in muscle mass, decrease of bone density and decline in functional abilities. Body-worn sensors are increasingly used to assess outcome in clinical trials by providing objective mobility parameters in a real-world environment. The aim of this study is to investigate the usability aspects and potential changes in mobility parameters in partial-immobilization patients in real-world conditions. Six healthy young males (age 22.2 ± 1.2 years; weight 76.5 ± 6.7 kg, height 185.8 ± 6.1 cm. Mean ± standard deviation) wore a leg cylinder cast with walker boot to immobilize their dominant leg for two consecutive weeks. Subjects were asked to continuously wear a tri-axial accelerometer on the waist (actibelt) during waking hours for 6 weeks including 2 weeks before, during and after cast immobilisation. The total amount of days of continuous recording was 339 days with a total wearing time of 120 days. Software packages which allow to detect steps and to estimate real-world walking speed were used to analyse the accelerometry data. It was suspected that knee immobilization would affect strongly the wave form of the signal with an impact on the accuracy of the speed algorithm, whereas the step detection should be more robust. This effect was confirmed in a preliminary study performed to quantify the accuracy under immobilization conditions. On the other hand, step numbers are known to be sensitive to fluctuations in wearing time which was not uniform throughout the entire study. We concluded that in this setting step frequency is the most reliable parameter. Step frequency showed a systematic decrease in the values during the immobilization period which recovered to pre-immobilisation values after cast removal. This confirms the usability of accelerometry and sensitivity of its mobility parameters for clinical outcome assessment

Total knee replacement in a commonly performed procedure in the United Kingdom with more than 76000 primary procedures performed in 2010. With so many procedures performed there has to be a robust way of assessing the outcome of the procedure. Gait analysis is a valuable tool in objectively assessing the these patients. Inertial movement units (IMU's) are a fairly new development in gait analysis. The aim of our project is to use IMUs to assess the differences in gait profile between a cohort of healthy controls, a group of pre operative knee replacement patients, a group of 8 week post operative patients and finally a group of post operative knee replacement patients at 1 year. We studied a total of 47 patients. We also had data from a previous study done on healthy controls using the same measurement tool. We measured three parameters: peak swing phase flexion, peak stance phase flexion and stride duration. Our findings indicate that pre-operative patients have a significantly reduced peak flexion in swing and stance with increased stride duration. This shows no improvement at the 8 week mark. At the 1 year mark peak flexion in swing returns to pre operative levels but flexion in stance and stride duration are still poor. These findings may not have been identified without gait analysis. Gait analysis using intertial movement units will add much information to radiographs and clinical examination. This information can also be used to tailor individual patients rehabilitation.

Background. In total knee arthroplasty (TKA), patient reported outcome on pain, function or satisfaction fails to differentiate treatment options. Activity, a consequence of pain-free, well functioning TKA and a satisfied patient, may be a discriminative surrogate metric, especially when objectively measured. Methods. Habitual activity was measured in TKA patients (n=32, F/M=20/12, age: 72 ±8yrs) at long-term follow-up (9 ±1yrs) and compared to healthy, age matched controls (n=32, F/M=20/12, age: 71 ±9yrs) using a popular questionnaire (SQUASH) and accelerometry. A small 3D accelerometer (X16-mini, GCD Dataconcepts) was worn for 4 successive days during waking hours at the non-affected lateral upper leg. Data was analysed using validated algorithms (Matlab) counting and timing walking bouts, steps, sitting periods and transfers. Stair climbing events or similar activities such as walking steep slopes were classified using the higher mean hip flexion angle as a feature. Results. SQUASH scores were not sign. different between TKA (mean ±SD: 4551 ±3426) and controls (3659 ±2720, p>0.1). With accelerometry, differences between patients and controls (Median [IQR]) increased from −13% for Time Standing (3.7h [2.6–5.1] vs 4.3h [2.9–5.0], p=0.69), to −26% for Daily Steps (4939 [3796–7910] vs 6731 [5539–8270], p=0.019] and −31% for Sit-Stand Transfers (31.2 [22.1–37.0] vs 45.3 [34.9–58.4], p>0.001). For stair-up events, the difference increased to −74% (6.2 [2.8–22.4] vs 23.9 [10.8–39.1], p>0.001). Conclusions. Self-report activity could not discriminate between patients and controls. With accelerometry, sign. differences increased with rising levels of effort and difficulty. In TKA, walking alone, either by accelerometry or as the major component of self-report may not serve as powerful outcome measure in orthopaedics were energy expenditure from walking is less relevant than functionally demanding but rarer tasks such as transfer or stair climbing

Capturing objective data of the postoperative changes in the mobility of patients is expected to generate a better understanding of the effect of postoperative treatment. Until recently, the collection of gait-related data was limited to controlled clinical environments. The emergence of accurate wearable accelerometers with sufficient runtime, however, enables the long-term measurement and extraction of mobility parameters, such as “real-world walking speed”. An interim analysis of 1967 hours of actibelt data (3D accelerometer, 100 Hz) from 5 patients (planned total 20) with a femur fracture and 5 patients (planned total 20) with a humerus fracture from a geriatric population at two different sites of the university hospital of the Ludwigs-Maximilian-University in Munich was performed. Mobility data was captured during several days of stationary treatment starting directly after surgery and during a short follow-up visit six weeks after the surgery. Preliminary results show an increase of the mean walking speed between the two visits independent of the type of fracture. Patients with a humerus fracture tended to walk faster than patients with a femur fracture during both visits. The data also reveals an unexpected low level of mobility during the stationary stay. Mobile accelerometry can be used to evaluate different postoperative mobilisation strategies and even provide near-time feedback in geriatric trauma patients