Actual physical activity after major abdominal cancer surgery

Background: Enhanced recovery after surgery protocols emphasize the importance of early postoperative mobilization. However, literature quantifying actual physical activity after major abdominal cancer surgery is scarce and inconclusive. Material and methods: A single-center prospective cohort study was conducted at the University Medical Center Groningen from 2019 to 2021. Patients ’ postoperative physical activity was measured using an accelerometer, with the primary aim of assessing daily physical activity. Secondary aims were identifying patient-related factors associated with low physical activity and studying the consequences of low physical activity in terms of complication rate and length of hospital stay. Results: 143 patients included (48 % male; mean age 65 years), 38.5 %, 24.5 %, 19.6 %, and 14 % underwent pancreatic, hepatic, colorectal, or cytoreductive surgery with hyperthermic intraperitoneal chemotherapy, respectively. Median daily step count was low; from median 71 steps on the first to 918 steps on the seventh postoperative day. An association between physical activity and age (OR 3.597, p = 0.013), preoperative weight loss ≥ 10 % (OR 4.984, p = 0.004), Eastern Cooperative Oncology Group performance status ≥ 2 (OR 4.016, p = 0.001), midline laparotomy (OR 2.851, p = 0.025), and operation duration (OR 1.003, p = 0.044) was found. An association was observed between physical activity and the occurrence of complications (OR 3.197, p = 0.039) and prolonged hospital stay ( β 4.068, p = 0.013). Conclusion: Postoperative physical activity is low in patients undergoing major abdominal cancer surgery and is linked to postoperative outcomes. Although physical activity should be encouraged in all patients, patient-specific risk factors were identified that can aid early recognition of patients at risk of low physical activity.


Introduction
Initiation of early mobilization is essential to postoperative physical and functional recovery [1].However, in clinical practice, evaluating daily mobilization relies primarily on patient reports.Limited studies have used real-time postoperative physical activity data to examine the effects of early mobilization following major abdominal surgery [1].
Since the widespread implementation of enhanced recovery after surgery (ERAS) protocols, focus on early postoperative physical activity has increased [2,3].Initiation of early postoperative mobilization is embedded in ERAS protocols and is promoted by optimizing postoperative analgesia and reducing the use of drains and tubes [4][5][6][7].Early mobilization has been found to contribute to a decrease in postoperative pulmonary complications, thromboembolic events, and overall postoperative morbidity, leading to shorter hospital stays [8][9][10].Given the significant physiological role of skeletal muscle mass in recovering from disease and its correlation with improved survival rates, preserving skeletal muscle mass is crucial.This involves promoting physical activity and ensuring adequate nutritional intake to support overall health and resilience [11].
The principles of ERAS have been broadly embraced, and numerous protocols have been published across multiple surgical disciplines [1].Nevertheless, patient adherence to specific daily mobilization goals outlined in these ERAS protocols is often low [12].Additionally, data on mobilization compliance are scarce for patients who underwent highly complex abdominal surgery.
Postoperatively, physical therapists play an important role in promoting patients' physical activity by setting daily mobilization goals and enhancing self-efficacy.If physical therapists were able to accurately track physical activity and identify patients at risk of not meeting targets, patients could experience better rehabilitation outcomes.Therefore, in this study we first aimed to quantify daily postoperative physical activity.Our second objective was to identify patient-related factors associated with low postoperative mobilization and examine the consequences of insufficient postoperative physical activity as measured by complications and length of hospital stay.

Study design
This prospective cohort study is part of the MUSCLE POWER study, which explores the risk factors and clinical impact of surgery-related muscle loss after major open-abdominal cancer surgery [13].The enrollment period for the study extended from May 2019 to June 2021.Eligible participants were adult patients diagnosed with a suspected malignant disease, scheduled for cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC), (sub)total pelvic exenteration, (sub)total colon resection, pylorus-preserving pancreatoduodenectomy (PPPD), Whipple procedure, (sub)total pancreatectomy, or major liver resection involving three or more liver segments.Exclusion criteria were an existing inability to walk and insufficient proficiency in the Dutch language necessary to fully comprehend the study's goal and purpose.Additionally, patients with incomplete data collection during the first five postoperative days were excluded.To minimize bias in measured postoperative physical activity, the study excluded patients admitted to the intensive care unit (ICU) for 3 or more days during the initial postoperative week and those undergoing (reconstructive) surgery mandating postoperative bedrest.
This study was approved by the Medical Ethics Committee of the University Medical Center Groningen (UMCG; METc2018/361, version 3.0, January 21, 2019), and written consent was obtained from all patients prior to inclusion.The study protocol is registered with the Netherlands Trial Register (NTR; NTR NL7505, version 1.0, February 7, 2019).

Patients and setting
Surgery-specific ERAS protocols are implemented across all surgical wards at the UMCG.Furthermore, all surgical wards (i.e., the hepatobiliary surgical ward and oncological surgical ward) are equipped with a dedicated physical therapist to assist patients throughout the postoperative rehabilitation process.Additionally, doctors and nurses actively encourage all surgical patients to engage in daily physical activity.

Data collection
During the first consultation with the surgeon at the outpatient department, eligible patients were informed of the study and invited to participate.One week later, the study investigator contacted patients to inquire about their interest in participating and to schedule an appointment for baseline assessments.During the baseline assessments, various study measurements were obtained, as detailed in a previously published study protocol [13].For this study, preoperative weight loss in the six months prior to surgery was assessed using a validated questionnaire for nutritional assessment called the patient-generated subjective-global-assessment short form (PG-SGA SF) [14].A PG-SGA SF score of ≥4 was used to indicate an increased risk of malnutrition [14].In our hospital, prehabilitation is offered to patients undergoing hepatopancreatobiliary surgery [15].This intervention could potentially render patients more aware and inclined to be physically active postoperatively.To explore this hypothesis, we analyzed whether patients who underwent prehabilitation were more active postoperatively.Patients were classified as having undergone prehabilitation if they had attended at least one appointment at the outpatient prehabilitation clinic in the UMCG.Next, Eastern Cooperative Oncology Group (ECOG) performance status was assessed [16].Additionally, demographic data were prospectively collected from electronic patient charts.Postoperative surgical details, including operation duration, blood loss, pain management, complications (Clavien-Dindo score and comprehensive complication index (CCI)) [17,18], length of ICU stay, length of hospital stay, and readmission rate, were obtained from operative reports and electronic patient charts.

Measurement of physical activity
Physical activity was measured using a mobility tracker (Actigraph wGT3X-BT; Pensacola, FL, USA) worn on the ankle [19,20].All patients were equipped with the device immediately after surgery and were instructed to wear it throughout the first postoperative week.The device did not provide direct feedback to the patient nor researchers while measuring physical activity.Afterwards, the data was uploaded to ActiLife, a software platform which analyzes recorded levels of physical activity.A valid monitoring day was defined as one in which the device was worn for 10 or more hours [21].The median daily step count was evaluated for each postoperative day, along with the sum of the median number of steps for the first 5 postoperative days.Physical activity was defined as "lowest" for patients who walked fewer steps than the lowest tertile (33 %) per day for a minimum of 3 days during the first 5 postoperative days.

Study endpoints
The primary endpoint of this study was the actual postoperative physical activity measured as median steps per day during the first 5 days after major abdominal surgery.Secondary endpoints included the association between patient-related factors (i.e., age, BMI, type of surgery, duration of surgery, and type of analgesia) and lowest postoperative mobilization level.Moreover, the association between the lowest postoperative physical activity and surgical outcomes in terms of complication rates and length of hospital stay was analyzed.

Statistical analysis
The statistical analysis was conducted using the Statistical Package for the Social Sciences Statistics version 28.0 (IBM Corp., Armonk, NY, USA).Continuous data distribution was assessed visually using QQ-plots and complemented by the Shapiro-Wilk test.Continuous variables were expressed as mean with standard deviation (SD) or as median with interquartile range (IQR), depending on the normality of the distribution.Categorical data were presented as the number and percentage.Subsequently, univariate logistic regression analyses were conducted to examine the relationship between potential determinants of low physical activity (e.g., age, sex, ASA score ≥3, PG-SGA SF score, preoperative weight loss, ECOG performance status, presence of comorbidities, preoperative prehabilitation, type of incision, type of analgesia, operation duration, and blood loss).Next, variables eligible for inclusion in the multivariate regression analysis were chosen if univariate analyses R.N.M. Hogenbirk et al.
between the outcome variable and predictor variable yielded a p value < 0.10.Furthermore, interactions between lowest physical activity and postoperative surgical outcomes were evaluated using univariate and multivariate logistic and linear regression analyses.The dependent outcome variables in the models were the presence of any complications, pulmonary complications, severe complications, and length of hospital stay.All models provided estimated regression coefficients and odds ratios with a 95 % confidence interval.Finally, a p value < 0.05 was considered statistically significant in all analyses.

Patient characteristics and surgical details
A total of 287 patients were screened for eligibility, and out of these, 212 (74 %) were included in the MUSCLE POWER study.Data on postoperative physical activity were available for 155 patients (73 %), of whom six (4 %) were admitted to the ICU for 3 days or more during the first postoperative week, three (2 %) were prescribed bedrest after their reconstructive surgery, and three (2 %) had three or fewer days of physical activity data available.Therefore, 143 out of 212 patients (67 %) were included in the final analysis (Fig. 1).No differences were found between the included and excluded patients regarding baseline characteristics.
Abbreviations: BMI = body mass index; ASA = American Society of Anaesthesiologists; PG-SGA SF = Patient-Generated Subjective Global Assessment Short Form; ECOG = Eastern Cooperative Oncology Group.
time to the onset of the first postoperative complication was 5 (3-7) days (Table 2).

Postoperative physical activity
Postoperative physical activity among patients was generally low, commencing at a median of 71  steps on the first postoperative day and gradually increasing to 918 (426-2034) steps on the seventh postoperative day.The median cumulative physical activity over five days for the entire cohort was 1573 (834-2692) steps (Fig. 2).
A total of 42 patients (29.4 %) demonstrated mobility below the lowest tertile (33 %) for three days or more and were categorized as having the lowest physical activity.On the first postoperative day, these patients took a median of 19  steps, in contrast to the rest of the cohort, which took a median of 104 (60-171) steps.It was not until the fourth postoperative day that these patients exceeded a median of 100 steps per day (lowest physical activity on fourth postoperative day: median 110 (45-201) steps versus 537 (307-1451) steps for the rest of the cohort).Patients with the lowest physical activity had a median total of 472 (160-857) steps over 5 days, while the rest of the patients had a median of 2085 (1533-3845) steps.
In multivariate regression analysis, the factors associated with the lowest postoperative physical activity were age, preoperative weight loss ≥10 %, ECOG performance status ≥2, midline laparotomy incision, and operation duration (Table 3).
Lowest physical activity was associated with the occurrence of complications in univariate logistic regression analysis (OR 3.700 [1.330, 10.293]; p = 0.012) and with the length of hospital stay in univariate linear regression analysis (β 5.943 [1.855, 10.031]; p = 0.005).When age, BMI, ASA score ≥3, preoperative weight loss, and operation duration were adjusted for in the multivariate logistic regression analysis, a significant association was observed between the occurrence of complications and low physical activity (OR 3.197 [95 % confidence interval 1.061, 9.628]; p = 0.039).Additionally, when age, BMI, ASA score ≥3, preoperative weight loss, operation duration, and the occurrence of severe complications (Clavien-Dindo score ≥3) were adjusted for in the linear regression analysis, a significant association was found between low physical activity and an increased risk of prolonged hospital stay (β 4.068 [0.868, 7.268]; p = 0.013).However, when the multivariate linear regression model was adjusted for age, BMI, ASA score ≥3, preoperative weight loss, operation duration, and the occurrence of any complications, this association lost its statistical significance (β 3.829 [− 0.332, 7.991]; p = 0.071; Table 4).

Discussion
This study demonstrates that patients undergoing major abdominal surgery generally display low levels of postoperative physical activity.The risk factors related to the lowest levels of physical activity were age, preoperative weight loss, ECOG performance status ≥2, and midline laparotomy incision.Multivariate regression analysis revealed a relationship between the lowest physical activity in the first 5 days and the occurrence of complications and prolonged length of hospital stay.
Low levels of physical activity are generally associated with negative consequences, such as thromboembolism, pneumonia, and muscle atrophy [1,22].Although ERAS protocols focus on initiation of early postoperative physical activity to promote functional recovery [2,[4][5][6][7], actual physical activity in patients seems to be low among various surgical and nonsurgical populations [12,[23][24][25][26][27].Hussey et al. reported on low postoperative physical activity, with post-esophagectomy patients (n = 30) spending more than 96 % of each day sedentary during the first 5 days postoperatively [27].In our study, we identified a correlation between low physical activity and an extended length of hospital stay.This finding aligns with previous research by Daskivich et al. and Browning et al. who each reported a reduced risk of prolonged hospitalization for patients who achieved a step count median of 1000 steps on the first postoperative day or spent more time in an upright position [23,28].Although Low et al. reported a somewhat higher average daily step count over the total length of hospital stay (median of 968.22 [48.86-3185.71]steps) compared to our patients, they reported a similar association between advanced age and a decreased daily step count [24].Additionally, they reported reduced 30-and 60-day readmission rates in patients with a higher postoperative median daily step count [24].
Researchers have discussed whether the association between low physical activity and the occurrence of complications is a causal relationship, or the other way around.Physical activity is commonly assumed to decrease due to complications causing physical discomfort.However, our results demonstrate a median time of 5 days between surgery and the onset of complications but low physical activity from the first postoperative day onward.This suggests a causal relationship between low physical activity and the occurrence of complications.To underpin this finding, Engel et al. (n = 199) reported a significant correlation between the extent of postoperative ambulation and decreased complications in patients after caesarean section [29].Generally, postoperative complications develop over the course of several days, and since our patients demonstrated low physical activity directly after surgery, this may have contributed to postoperative complications.
The results of this study indicate more emphasis should be placed on physical activity of patients after major abdominal surgery.Therefore, any modifiable barriers that prevent patients from starting early postoperative mobilization should be identified.Notably, both patient-and hospital/treatment-related factors are modifiable barriers.First, we identified advanced age, preoperative weight loss, ECOG performance  status ≥2, and midline laparotomy incision as risk factors associated with low postoperative physical activity.These risk factors may serve as indicators to identify patients requiring extra attention during the postoperative period.Furthermore, both preoperative weight loss and ECOG performance status are also considered risk factors that can be accounted for to achieve perioperative optimization in prehabilitation programs [30,31].Moreover, as indicated in previous studies, the presence of medical attachments (e.g., urinary catheters) impede patients during mobilization and hamper physical activity [28,32].Although these medical attachments are often necessary during the postoperative period, practitioners should be aware of their potential negative consequences.
Nurses play an important role in promoting physical activity of patients by providing them with physical and motivational support.However, the workload of nurses is considered a key barrier to improving the physical activity of patients, because supporting patients during mobilization is a time-consuming task [1,33].Although physical therapists can help nurses improve the physical activity of their patients, limited staffing in wards remains a critical concern.Another barrier identified by Geelen et al. is that a patient's hospital stay is predominantly restricted to their hospital bed [33].Because most in-hospital activities occur in or near the patient's bed (e.g., ward rounds), the hospital environment does not encourage physical activity.Finally, although ERAS protocols across all surgical domains advocate for promoting physical activity among patients, no specific activity targets are defined.A lack of predefined activity targets for patients may limit clarity for healthcare staff and patients and consequently lead to insufficient physical activity.
New methods to stimulate and improve physical activity could be implemented for patients after surgery.One source of support may be from medical students, who can relieve the nursing staff of some of their duties.In our hospital, for instance, a project involving medical students watching over patients with delirium was recently executed successfully [34].These trained medical students helped calm down patients during periods of agitation, thereby decreasing the nurses' workload.If properly instructed, medical students might also be employed to support patients during mobilization.Patients with identified risk factors associated with low postoperative physical activity may especially benefit from daily activity support.As an alternative, activity monitors have been proposed as an inexpensive platform for monitoring the daily step count of patients [1,23,[35][36][37][38].If the physical activity of patients, as registered by their activity monitors, is included in the daily ward round as an additional vital parameter, patients with low physical activity can be easily detected at an early stage, allowing for timely interventions aimed at promoting physical activity.In this study, we used Actigraph accelerometers, which are increasingly being used in research settings and proven to be feasible for monitoring physical activity in patients [39,40].Most studies have only focused on the diagnostic function of these activity trackers, but Wolk et al. reported that using an activity tracker with continuous patient feedback in a group of patients (n = 132) substantially increased the mean step count on postoperative days 1-5 compared to a control group [35].Hence, activity trackers with direct feedback might be used as a therapeutic modality to help motivate patients to increase their daily activity [35,37].More than half of the participants in this study underwent prehabilitation.We hypothesized that these individuals might be more aware of the importance of early postoperative physical activity, making them more likely to engage in such activities.Furthermore, prehabilitation programs commonly encompass cardiovascular conditioning, respiratory exercises, and mobility training, all contributing to enhanced functional capacity [41].This enhanced physiological preparedness aids in navigating immediate surgical challenges and potentially facilitates a smoother transition to postoperative rehabilitation.Additionally, psychological aspects are integral to prehabilitation and encompass factors like surgery-related anxiety.A positive mental state fostered through prehabilitation can influence a patient's attitude toward postoperative recovery, promoting active engagement in physical activities [42].Although our study did not find a statistically significant difference in immediate postoperative physical activity levels between patients who underwent prehabilitation and those who did not, it is crucial to acknowledge that the impact of prehabilitation may extend beyond the immediate recovery phase.Prehabilitation is designed to enhance baseline physical fitness and resilience before surgery, and its benefits, including improved muscle strength and cardiovascular fitness, may contribute to a more robust and sustained recovery trajectory.Thus, the enduring impact of prehabilitation on postoperative physical activity levels may become more apparent in later recovery stages.Further investigation and longitudinal follow-up are warranted to explore these lasting benefits.
This study has some limitations.First, although we speculated about the importance of specific targets for ambulation, the study design was not suited to providing such targets.However, because patients with the lowest physical activity did not have a step count over 100 until the fourth postoperative day, whereas the rest of the participants already had a median step count above 100 on the first postoperative day, we cautiously suggest setting a target that minimally increases by 100 steps per day after major abdominal surgery.The next limitation lies in the study design, which did not allow us to assess the causality of low physical activity; hence, we can only speculate about possible explanations.Also, 61 of the included patients had an ECOG performance status of ≥2, introducing a potential risk of bias since these patients already displayed lower physical activity levels preoperatively.However, we suggest that these patients in particular require additional attention in the postoperative period.Finally, we included different types of abdominal surgery, but each surgical procedure is associated with a specific postoperative recovery pattern.Nonetheless, as depicted in the population graphs in Fig. 2, physical activity did not considerably differ between these surgical populations.
To conclude, this study found evidence of low postoperative physical activity in patients who undergo major abdominal cancer surgery.Although physical activity should be encouraged in all patients, we identified specific individual-level risk factors that should alert medical staff to encourage these patients to increase their postoperative activity.We herein propose several methods to increase physical activity in patients.

Ethical approval and consent to participate
The study protocol was approved by the Medical Ethics Committee of the UMCG, the Netherlands (METc2018/361, version 3.0, January 21, 2019), and registered within the International Clinical Trials Registry Platform (201 800 445, NL7505, version 1.0, February 7, 2019).This study was performed in accordance with the ethical standards set by the Declaration of Helsinki.All patients provided written consent before participating in the study.
Fig. 1.Title: Inclusion flowchart Legend: Overview of the assessment of eligibility, inclusion, and follow-up of patients included in the MUSCLE POWER study with available data on postoperative physical activity.

Fig. 2 .
Fig. 2. Title: Postoperative physical activity after major abdominal surgery Legend: Graphical overview of the actual physical activity in steps per day after (a) major abdominal surgery, (b) hepatic surgery, (c) pancreatic surgery, (d) colorectal surgery, and (e) CRS and HIPEC.The boxes indicate the first quartile (lower border of the box), median (horizontal line within the box), and third quartile (top border of the box).The whiskers extending from the top and bottom borders of the boxes indicate the range of nonoutlier values.The individual points outside the whisker boundaries indicate outlier values that are less than the first quartile or greater than the third quartile by 1.5 times the IQR.CRS = cytoreductive surgery; HIPEC = hyperthermic intraperitoneal chemotherapy.

Table 1
Patient characteristics.

Table 2
Surgical details.

Table 3
Univariate and multivariate analysis of factors associated with low postoperative physical activity.Data are presented as continuous variables.Abbreviations: ASA = American Society of Anaesthesiologists; PG-SGA SF = Patient-Generated Subjective Global Assessment Short Form; ECOG = Eastern Cooperative Oncology Group; PCEA = patient-controlled epidural analgesia.N = number of patients; OR = Odds ratio; CI = confidence interval; aOR = adjusted odds ratio.