Yoeri Bemelmans

Physical activity after outpatient and fast-track knee arthroplasty 143 9 Table 2. Patient characteristics and operative data presented as mean (±SD) or absolute numbers for both pathways. ERP OS p-value Patient characteristics Age (years) at index surgery 66.9 (7.9) 64.1 (3.5) n.s. Male/ female 5/5 8/2 n.s. BMI kg/m 2 29.2 (6.0) 27.7 (4.7) n.s. ASA, I/II/III 2/8/0 3/7/0 n.s. Operative data Blood loss, mL 237.5 (106.1) 233.3 (136.9) n.s. OR time, min 65.8 (13.8) 58.8 (18.9) n.s. Outcome Physical activity in the daily life of patients was measured in a non-invasive way using triaxial accelerometer (GCDataconcepts LLC,Waveland, USA). TheAMwas attached onto the lateral side of the non-affected upper leg. Based on previously published principles [27, 28], raw accelerometer data were post-processed and analysed using self-developed algorithms for feature detection and activity classification written in Matlab (MATLAB R2010a, TheMathworks Inc.,Natick,Massachusetts,USA) [25].Moredetailed informationof theAMand itsclinical applicationaredescribed inaprevious study [34]. Activityparameters calculatedwere thedailynumberof events (counts) and total time (duration) spent sittingor lying (inactive), standing, walking or cycling (active) and the number of steps and sit–stand transfers. Inaddition to thesequantitativeparameters, qualitativeaspectsof activities could also be calculated, such as walking cadence (steps/min), time-wise distribution of walking bouts (e.g. numberwalkingbouts consistingof less ormore thananamount of steps), sitting and standing. Physical activity was measured for four consecutive days [13, 26]. The AM was worn only during waking hours with a minimum of 8 h a day and removed at night and during showering. The daily physical activity was measured at three time points. First, patients wore the monitor during four days before surgery until the day of surgery. During the surgery hours, the AM data were extracted. Second, once the surgery was completed, theAMwas returned to thepatients and theywore the sensor until the fourthpostoperative day. Within the first 4–6 h after surgery, the patient started with early mobilization (Table 2). Patients following the OS pathway were discharged later the same day, while the ERP patients were evaluated twice a day to check whether theymet discharge criteria (e.g. safe mobilization, able to climb stairs and able to perform sit–stand transfers, adequate pain/ PONV control). The third activity assessment was during the 6th postoperative week and again measured for four consecutive days. The time between hospital admission and discharge and the time between the start of anaesthesia until the first mobilization were recorded in hours in the patient’s clinical report. PROMs were obtained pre- and 6 weeks