TITLE: Mobilization Started Within 2 Hours After Abdominal Surgery Improves Peripheral and Arterial Oxygenation: A Single-Center Randomized Controlled Trial RUNNING HEAD: Effects of Immediate Mobilizations After Surgery

Objective. The aim of this study was to investigate if mobilization out of bed, within 2 hours after abdominal surgery, improved participants´ respiratory function and whether breathing exercises had an additional positive effect. Methods. Participants were 214 consecutively recruited patients who underwent elective open or robot-assisted laparoscopic gynecological, urological, or endocrinological abdominal surgery with an anesthetic duration of >2 hours. They were recruited to a randomized controlled trial. Immediately after surgery, patients were randomly assigned to 1 of 3 groups: mobilization (to sit in a chair) and standardized breathing exercises (n = 73), mobilization (to sit in a chair) only (n = 76), or control (n = 65). The interventions started within 2 hours after arrival at the postoperative recovery unit and continued for a maximum of 6 hours. The primary outcomes were differences in peripheral oxygenation (oxygen saturation [Sp O 2 ], as a percentage) and arterial oxygenation (arterial oxygen pressure [Pa O 2 ], in kilopascals) between the groups. Secondary outcomes were arterial carbon dioxide pressure, spirometry, respiratory insufficiency, pneumonia, and length of stay. Results. Based on intention-to-treat analysis (n = 214), patients who received mobilization and breathing exercises had significantly improved Sp O 2 (mean difference [MD] = 2.5%; 95% CI = 0.4 to 4.6) and Pa O 2 (MD = 1.40 kPa; 95% CI = 0.64 to 2.17) compared to the controls. For mobilization only, there was an increase in Pa O 2 (MD = 0.97 kPa; 95% CI = 0.20 to 1.74) compared to the controls. In the per-protocol analysis (n = 201), there were significant improvements in Sp O 2 and Pa O 2 for both groups receiving mobilization compared to the controls. Secondary outcome measures were not different between groups.

Early mobilization is recommended after abdominal surgery and is commonly considered to be one of the factors that enhances patients' recovery. In the fast-track-concept, Enhanced Recovery After Surgery (ERAS), it is recommended that patients are mobilized on the day of surgery. 1 However, adherence to these guidelines is unclear. 2 Postoperative pulmonary complications, such as atelectasis or pneumonia, are common after surgery, and may cause unnecessary discomfort, prolonged hospital stay, and increased health care costs. 7, 8 Usually, postoperative pulmonary complications are consequences of reduced functional residual capacity (FRC) following surgery, anesthesia, and immobilization. [8][9][10][11][12][13] Mobilization or being in an upright position and breathing exercises are assumed to prevent a reduction in FRC, [14][15][16] and are frequently recommended after surgery. 8, [17][18][19] Previous studies have evaluated mobilization combined with other treatments, 1,20 with different breathing exercises 21,22 , mobilization starting at postoperative day 1 (POD1), 3,18,23,24 or compared to standard of care. 6,25 To date, the isolated effect of immediate postoperative mobilization on respiratory function has not been thoroughly investigated. 3,18 To our knowledge, no study has compared the immediate respiratory effect of mobilization out of bed to sitting in a chair plus breathing exercises, starting on the same day as abdominal surgery, to an untreated control group.
Therefore, a randomized controlled trial was conducted, hypothesizing that mobilization, within 2 hours after elective open or robot-assisted laparoscopic gynecological, urological, or endocrinological abdominal surgery, would improve patients´ respiratory function and that breathing exercises would have an additional positive effect.

[H1] Methods
A 3-arm randomized controlled trial was conducted at a postoperative recovery unit at a university hospital in Stockholm, Sweden. The study was approved by the Regional Ethical Review Board in Stockholm and prospectively registered at ClinicalTrials.gov (NCT02929446).
Immediately after surgery, when the patients entered the postoperative recovery unit, a nurse at the recovery unit, independent of the trial, randomly assigned the spontaneously breathing patients to mobilization and breathing exercises, mobilization only, or control.
A computer-generated randomization in blocks of 9 was used to allocate patients to the different trial groups (1:1:1). Treatment allocation was concealed by random selection of opaque sealed envelopes prepared by an investigator with no further involvement in the trial.
It was not possible to mask patients or health care professionals at the recovery unit to treatment allocation. The surgeons who performed the operations, the anesthesiologists at the postoperative recovery unit, the research nurses and the physical therapists who performed the postoperative spirometry were all masked with regard to group allocation. All data were coded, and group assignment was masked for analyses.

[H2] Interventions
The intervention period started within 2 hours after arrival at the postoperative recovery unit.
Interventions continued for 6 hours, or less if the anesthesiologist in charge considered the patient fit for discharge to the surgical ward. The 3 groups were mobilization and breathing exercises, mobilization only, and control.
[H3]Mobilization and breathing exercises. For this group, mobilization out of bed, assisted by a physical therapist and a nurse, to sit in a chair (or unsupported on the bedside if unable to stand and transfer to a chair) was performed. The patient was instructed to sit for as long as her or she could. If required, sitting was interspersed by 1 hour of bed rest with the bed head at ≥45 degrees. The patient was instructed by a physical therapist to perform 10 × 3 deep breathing exercises every hour with a positive expiratory pressure device (PEP T-piece and Resistor; Intersurgical AB, Danderyd, Sweden) at a pressure of 10 to 15 cm of H 2 O The physical therapist ensured that the breathing exercises were correctly performed.
[H3]Mobilization. For this group, mobilization out of bed, assisted by a physical therapist and a nurse, to sit in a chair (or unsupported on the bedside if unable to stand and transfer to a chair) was performed. The patient was instructed to sit as long as he or she could. If required, sitting was interspersed by 1 hour of bed rest with the bed head at ≥45 degrees.
[H3]Control. For the control group, no mobilization or breathing exercises were performed.

[H2] Data collection
Demographic variables such as weight, height, age, history of obstructive/restrictive respiratory disease, smoking, ASA score, 27 and planned surgery were collected from the patient and his or her medical records. The following treatment-related data were collected from the bedside chart at the recovery unit: time for arrival/discharge, type of surgery, perioperative anesthesia, and total length of anesthesia and surgery. At each hour, patients rated pain and nausea on a numeric rating scale (NRS), where 0 = no pain/nausea and 10 = the worst pain/nausea imaginable. 28 The time point for the start of mobilization and the duration and frequency of mobilization were registered in a bedside chart.

[H2] Outcome measures
The primary outcomes were change (between groups and over time) in oxygen saturation (SpO 2 , as a percentage) measured peripherally with a pulse oximeter and arterial oxygen pressure (PaO 2 , measured in kilopascals) measured via arterial blood gas sample. PaO 2 could not be assessed for 15 patients since they did not receive an arterial line during surgery.
Assessments were performed at arrival to the postoperative recovery unit and every subsequent hour until discharge. The oxygen supply was disconnected 15 minutes before each hourly assessment. The assessments at arrival were performed with all patients in a supine

[H2] Adverse Effects
Possible adverse consequences of mobilization and breathing exercises, such as falls, drainage that accidentally went out (extubated) during or due to the mobilization, and surgical wound ruptures, were monitored according to clinical procedures at the hospital.
[H3] Data Analysis Assuming that SpO 2 would increase by 2% (SD = 4) or that PaO 2 would increase by 0.5 kPa (SD = 1), 30 the number of patients required to establish 80% power at a For respiratory insufficiency 29 and pneumonia, 29 logistic regression was used to assess the association between respiratory insufficiency and the interventions (mobilization and breathing exercises, mobilization only, and control) and the association between pneumonia and the interventions (mobilization and breathing exercises, mobilization only, and control), respectively. Results were presented as odds ratios (ORs) with 95% CIs. The models were adjusted for age (<61 years, 61-74 years, and >74 years), 10,32 sex, body mass index (<30 or ≥30), 10 smoking (never/ever), 10 preoperative lung disease (restrictive and/or obstructive diseases, chronic obstructive pulmonary disease, obstructive sleep apnea, or asthma registered in the medical records) (yes or no), 10,32 duration of anesthesia (<4 hours, ≥4 hours), 7,32 and type of surgery (open or robotic surgery). 10,32 For length of stay at the recovery unit and at the hospital, the 3 treatment groups were compared using analysis of variance. A P value of <.05 was considered statistically significant. The Statistical Package for the Social Sciences Version 24 (SPSS; IBM Corp, Armonk, NY, USA) was used for all statistical analyses.

[H2] Role of the funding Source
The funding source had no role in the design, conduct or reporting of this trial.

[H1] Results
Between The remaining 214 patients were randomly assigned to 1 of the 2 groups receiving mobilization or the control group at arrival to the postoperative unit. Seventy-three patients were assigned to mobilization and breathing exercises, 76 were assigned to mobilization only, and 65 were assigned to control. Six patients had cardiorespiratory instability prior to trial start and could not fulfill mobilization. For 2 patients, the caregivers did not adhere to the study protocol, and 5 patients withdrew from the study (Fig. 1). Table 1 shows the characteristics and demographics of the 214 enrolled patients.
Approximately 60% were women. The most common types of surgery were robot-assisted laparoscopic urologic surgery (36%), such as prostatectomy or cystectomy (radical with construction of orthotopic bladder substitution), and open gynecological surgery (27%), mostly for ovarian cancer stages III and IV, involving bowel surgery (colorectal surgery) and stripping of the diaphragm.

[H2] Intervention compliance
Of the 214 randomized patients, 201 patients completed the allocated intervention. The mean mobilization time was 105 minutes, ranging from 10 to 265 minutes, with no statistically significant difference between the two groups receiving mobilization (P = .93). All patients in the group receiving mobilization and breathing exercises completed the breathing exercises according to protocol.
Four patients, 1 in the group receiving mobilization and breathing exercises and 3 in the group receiving mobilization only, sat unsupported on the bedside. They were unable to mobilize to a chair because they lacked sufficient leg muscle tonus as a result of spinal/epidural anesthesia or low blood pressure and dizziness, which made it unsafe to stand up.

[H2] Primary outcomes
There were no baseline differences in SpO 2 (Tab. 1). In the intention-to-treat analysis (n =  The patients experienced no adverse effects while being mobilized. Earlier trials in other populations and in healthy participants suggest that changing position, to sitting in a chair or to standing, is a simple way to achieve an increased lung volume, provide inspiratory flow in small airways, and prevent and treat atelectasis. 14,15,17,34,35 The effect on volume is instantaneous, but when returning to bed, respiration returns to lower volume.
Nevertheless, there are effects that persist. Recruited airways are less likely to re-collapse when opened, which can affect SpO 2 and PaO 2. outcomes can be improved, as they are associated with increased early postoperative mortality, intensive care unit intake, and total hospital stay.
There was a reduction in FVC, FEV 1 , and PEF from baseline to POD1 in all groups, which supports previous research. 25, 41 However, spirometry measurements were performed the day after the interventions, which might explain why there were no differences between the groups.
Between 20% and 30% of the patients in our study had respiratory sufficiency 29 , which may be considered high compared to other studies. 42 However, all assessments were performed after patients had been without oxygen supply for 15 minutes.
Previous studies have shown that atelectasis occurs already during induction of anesthesia and was greatest 2 hours after surgery. In addition, the atelectasis remained almost the same for 2 days postoperatively. 11,43 In present study, we started the interventions within 2 hours of arriving at the recovery unit. Despite the relatively short intervention, SpO 2 and PaO 2 increased over time in the two groups receiving mobilization but decreased accordingly in the control group. Although the differences between the groups in SpO 2 and PaO 2 can be considered marginal, it may have a greater clinical significance. Our clinical experience is that patients who fall in SpO 2 /PaO 2 , postoperatively initially often receive oxygen as a first treatment, not mobilization. Our opinion is that this risk masking the patient's true respiratory status. In the present study, all assessments of SpO 2 and PaO 2 were without oxygen and this is the first study examining the isolated effect of mobilization (already within 2 hours after abdominal surgery) on respiratory function, such as SpO 2 and PaO 2 . It might be considered too early to mobilize a patient out of bed within 2 hours after abdominal surgery. However, in the present study, we found that it was feasible, and we did not encounter any adverse consequences attributable to the intervention. There were no cases of falls with injury, drainage that went out, rupture of surgical wounds. In addition, mobilization out of bed immediately after abdominal surgery has been shown to affect patients' physical and mental well-being. Patients report that they wake up, feel more alter, regain their autonomy, recover appetite and thirst, and feel that it is easier to breathe when they sit up in a chair, compared to when they lie in bed. 44

[H2] Strengths and limitations
The major strengths of this trial is the design, which limits the risk of selection bias, the inclusion of a relatively large sample of patients from a clinical setting, and the untreated control group which allows investigation of the isolated effect of mobilization. There are also some limitations that need to be considered.
There were some baseline imbalances in patient characteristics. This was adjusted for in the mixed model but should be kept in mind when interpreting the results.
By limiting study inclusion to daytime only, there might be a risk that patients most susceptible for respiratory insufficiency have been excluded. Another limitation is that there was no group that only performed the breathing exercises. Therefore, the isolated effect of breathing exercises cannot be evaluated. Masking of the intervention for physical therapists, nurses and the patients at the postoperative recovery unit was not possible, however the surgeons and the anesthesiologists responsible for patient discharge from the postoperative recovery unit and the investigators who performed the outcome spirometry were masked with regard to group allocation. The patients in the control group were assessed in a supine position, but between the assessments the bed head could sometimes be positioned in 0 to 30 degrees, which is according to the clinical care at our unit.
For clinical reasons, not all patients had an arterial line during surgery. Therefore, there were missing PaO 2 data. However, the analysis of SpO 2 and PaO 2 correlated well. Another important factor is that some patients were discharged before the third hour assessment, which might have reduced power for these analyses. However, there were statistically significant differences between the groups. After discharge from postoperative care, the patients were treated according to standard care, therefore the variables collected during this period were not controlled for and other confounding factors might have influenced the results. Length of stay at the recovery unit is not a precise outcome measure, and many factors influence the discharge decision. Patients were discharged from the postoperative recovery unit based on predefined criteria and not only according to respiratory status. Other important factors such as pain, nausea or prearranged overnight stay also prolonged the stay. Thus, further study is needed to determine the ideal type, level, duration and frequency of mobilization after surgery, and the long-term respiratory effect.

[H1] Conclusions
Mobilization out of bed to sit in a chair, with or without breathing exercises, within 2 hours after surgery, improved PaO 2 and SpO 2 after open or robot-assisted laparoscopic gynecological, urological, and endocrinological surgery, without major side effects.