Surgical outcomes and quality of life following exercise-based prehabilitation for hepato-pancreatico-biliary surgery: A systematic review and meta-analysis

Any Deprto , , Kevin Verhoeff,,, Kiern Purih , Jnie Y Kung , Dvi L Bigm ,Khle Z Djni

a University of Alberta, 116 St & 85 Ave, Edmonton, Alberta T6G 2R3, Canada

b Department of Surgery, University of Alberta Hospital, 8440 112 Street NW, Edmonton, Alberta T6G 2B7, Canada

c John W. Scott Health Sciences Library, University of Alberta Hospital, 8440 112 Street NW, Edmonton, Alberta T6G 2B7, Canada

d Department of Surgery, Division of General Surgery, HPB Transplant and Oncology, University of Alberta Hospital, 8440 112 Street NW, Edmonton, Alberta T6G 2B7, Canada

ABSTRACT

Keywords:Exercise-based prehabilitation Preoperative exercise Hepatobiliary Pancreaticoduodenectomy Hepatic resection Pancreatectomy

Introduction

Prehabilitation involves anticipatory treatment administered to patients aiming to optimize their preoperative condition before the insult of operation [1–4] . The term prehabilitation and its principles were first reported by military physicians during World War II,with diet and exercise modifications aimed at improving the functional fitness of substandard recruits prior to the stress of deployment [5] . Modern applications have developed from these practices, with various preoperative interventions aimed at improving patient outcomes following surgery [1–4] .

Exercise-based prehabilitation involves physical fitness assessment and targeted interventions to improve cardiovascular capacity, physiologic reserve, strength, and postoperative outcomes. The patient’s active role in exercise interventions and sense of involvement may also improve psychosocial outcomes [ 6 , 7 ]. The direct pathophysiological pathway through which exercise prehabilitation improves outcomes is not well described. However, patients with hepato-pancreatico-biliary (HPB) and gastrointestinal malignancies often suffer significant loss of strength within the waiting period for surgery [ 8 , 9 ]. During this time, exercise prehabilitation programs help patients increase their physiologic reserve [9–11] . Offsetting frailty, sarcopenia, and malnutrition in these patients may improve postoperative outcomes [ 8 , 11–14 ]. Directed physical activity alleviates sarcopenia [ 8 , 12 , 14 ], which is a significant contributor to postoperative length of stay (LOS) and morbidity across various surgeries [ 11 , 13 , 15–20 ]. Even limited volumes of exercise in sedentary populations have a substantial dose-response to resist cardiorespiratory and musculoskeletal declines and improve patient physiologic capacity [ 21 , 22 ]. Studies have demonstrated the benefits of diet, exercise, and psychosocial prehabilitation modalities before surgery, both as independent [ 1 , 3 , 6 , 23 ] and combined modalities [ 2 , 24 , 25 ]. Improved physical performance with exercisebased prehabilitation has demonstrated reduced LOS, morbidity,mortality, and improved quality of life (QoL) in thoracic, colorectal, and abdominal surgery patients [ 1 , 24–26 ].

Prehabilitation is difficult to evaluate in HPB patients, who differ from other surgical populations due to their frequent baseline frailty, substantial weight loss, brief preoperative timeline, marked sarcopenia, and malnutrition due to malabsorption [ 8 , 10 , 13 , 14 , 23 , 27–31 ]. An HPB patient’s preoperative condition may also further deteriorate during neoadjuvant chemoradiation or after procedures to relieve biliary obstruction prior to resection [ 10 , 23 , 28 , 29 ]. These factors make implementing and studying exercise-based prehabilitation in this patient cohort difficult but exceedingly important. Given these considerations, most prior studies have only risk-stratified HPB patients based on their exercise capacity [ 29 , 32–34 ], where only a few small randomized control trials (RCTs) and case series have evaluated outcomes following exercise prehabilitation for this population [ 8 , 10 , 23 , 27 , 28 ].Even less is known about the psychosocial aspects of exercise prehabilitation for these patients [ 23 , 28 , 35 , 36 ]. This systematic review and meta-analysis evaluated the surgical outcomes of patients undergoing major HPB related surgeries following exercise-based prehabilitation. Our study aimed to provide a quantitative assessment of postoperative outcomes including LOS, complications, physical performance, and psychosocial outcomes following prehabilitation before HPB surgery. We hypothesized that exercise-based prehabilitation reduces postoperative LOS and postoperative complication rates, and improves physical performance and psychosocial outcomes after HPB surgery compared to standard care.

Methods

Study design and research question

We conducted a systematic review and meta-analysis in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [37] . The PICOS (population, intervention, comparison, outcome, and study design) framework was applied to outline the study question [38] . The population of interest was adult patients undergoing any HPB surgery excluding transplantation. The intervention was exercise-based prehabilitation as compared to standard care [21] . The primary study outcome was postoperative LOS, with secondary outcomes evaluating postoperative complications, major complications (Clavien-Dindo grade ≥III), mortality, physical performance, and QoL. All peer-reviewed studies were eligible, while case reports, abstracts,letters, narrative, and non-peer-reviewed articles were excluded.

Search strategies

A medical librarian (Kung JY) developed and conducted comprehensive searches in MEDLINE (via Ovid), Embase (Ovid), Scopus, Web of Science Core Collection, Cochrane Library (Wiley), and ProQuest Dissertations and Theses Global in May, 2020. The search was updated in August, 2021. Searches included an exhaustive list of keywords and controlled vocabulary to ensure that all relevant literatures related to HPB-specific procedures and exercise-based prehabilitation were retrieved for the review. Only English studies were included due to language editing limitation and no date limits were applied. Transplant patients were excluded to better focus the investigation.

In addition to searching subscription databases, the research team reviewed the first 200 results from Google Scholar for inclusion. This has been demonstrated to be a reasonable number of results to screen since there is high overlap between Web of Science and Google Scholar [39] . Bibliographies from included studies were also considered for inclusion. Ethics approval was not required as all data collected were previously published. We did not have a registered protocol; however, our research question was well defined prior to the search and no alterations were made to our proposed question and outcomes.

Study inclusion and exclusion criteria

Studies were systematically reviewed and selected based on the following inclusion criteria: patient underwent an HPB surgical procedure; patient underwent an exercise-based prehabilitation program for a minimum duration of 7 days; the study evaluated at least one outcome of interest compared to standard care; and the study enrolled at least five patients. Abstracts, animal studies,non-English studies, studies without postoperative data, and studies with patients＜18 years old were excluded to limit heterogeneity. A minimum of 7-day exercise-based prehabilitation duration was included to ensure that HPB patients were allotted sufficient exercise exposure to receive a physiological training effect[ 21 , 22 , 40 ]. The American College of Sports Medicine definition of exercise was used to determine if certain activities qualified as exercise [21] .

As per PRISMA guidelines, two independent reviewers screened titles and abstracts, assessed full texts, and extracted data from included studies (Deprato A and Verhoeff K) [37] . Disagreements were resolved by re-extraction or third-party adjudication (Purich K).

Study risk of bias and quality assessment

Quality assessment was completed independently by two authors (Deprato A and Verhoeff K), with disagreements resolved by consensus. Included non-randomized studies were assessed for quality using the Methodological Index for Non-Randomized Studies (MINORS) [41] . The MINORS criterion is a validated index for non-randomized comparative studies and consists of a 12-point checklist where each aspect is scored as 0 when not reported, 1 for reported but inadequate, and 2 for reported and adequate. The maximum possible score for non-randomized comparative studies is 24. Comparative study scores can be assessed as: 0-6 = very low quality; 7-12 = low quality; 13-18 = moderate quality; and 19-24 = high quality [42] . RCTs were evaluated with the revised Cochrane risk-of-bias tool for randomized studies (ROB2).The ROB2 tool is recommended by Cochrane Reviews and evaluates studies for bias on six domains with signaling questions used to guide assessment [43] . The assessment of randomized controlled studies was not planned in our initial study protocol since randomized controlled studies were unexpected and were added after data extraction.

Primary and secondary outcomes

All outcomes were determined apriori. Patient demographics were collected including age, sex, prehabilitation intervention, and follow-up. Prehabilitation interventions of included studies were evaluated in terms of the frequency, intensity, time, and type (FITT)characteristics of their exercise-based components. Further data were collected for the prehabilitation modalities including duration of programming and patient adherence. Surgical details collected included patient diagnosis, surgery, and whether patients underwent neoadjuvant chemotherapy or biliary stenting/drainage.

The primary outcome was postoperative LOS, defined as the number of days patients remained in-hospital following surgery before discharge. Secondary outcomes included complication rates,major complication rates (defined as Clavien-Dindo grade ≥III),physical performance (any quantitative assessment of cardiorespiratory function, muscle strength, muscle endurance, flexibility,body composition, or neuro-motor fitness) [21] , and QoL (assessed by any validated measure). Preoperative physical performance outcomes were added as an outcome during the study to investigate patient responses to exercise-based prehabilitation prior to their surgery.

Statistical analysis

Whether data were presented as either median(range/interquartile range) or mean and standard deviation was estimated by applying the formula presented by Wan et al. [44] .Patient characteristics and follow-up data were summarized and described as weighted means, percentages, or counts in rarity of data. Meta-analysis was used to compare differences between groups in primary and secondary outcomes when appropriate.Results were reported as mean differences for continuous data and odds ratios for discrete data using 95% confidence interval (CI) and significance level ofP＜0.05.

The estimated effects were calculated using RevMan 5.4 software provided by the Cochrane website. The Mantel-Haenszel random-effects method was used for dichotomous data analysis and the inverse-variance random-effects method for continuous data analysis, assuming the true effect estimates varied among studies. Included studies were tested for heterogeneity, with low,moderate, and highI2values quantified as＜50%, 50%-75%, and＞75%, respectively [45] . Subgroup analysis of LOS for patients undergoing hepatic resection and pancreatic surgery was planned prior to data collection. In addition to quantitative analysis, a narrative review of included studies was performed.

Results

Study selection

The search retrieved a total of 2524 studies (articles and abstracts), where after removing all duplicates, 1778 unique publications underwent title and abstract review with 67 isolated for full-text review. After review, 6 studies met inclusion criteria (RCT,n= 3; prospective cohort,n= 1; retrospective cohort,n= 2)( Fig. 1 ).

Study risk of bias and quality assessment

Results from the risk of bias assessment demonstrated that comparative studies (n= 3) were of high quality (MINORS score = 18.7 ± 0.6, Table S1). The ROB2 analysis completed on the three randomized studies demonstrated that one RCT had a low risk of bias, while the other two had a high risk of bias; this was primarily due to poor assessment of experimental non-compliance with bias in the direction of the control (Table S2).

Characteristics of study population

A total of 957 patients were identified in the six studies, of whom, 536 (56.0%) underwent exercise-based prehabilitation and 421 (44.0%) received standard care in the control group. Prehabilitation patients and those undergoing standard care were similar in terms of age (63.7 vs. 66.1 years) and sex (n= 203, 37.9% female vs.n= 162, 38.5% female), respectively ( Table 1 ).

Exercise-based prehabilitation was unimodal in two studies[ 23 , 46 ], with the rest utilizing multimodal approaches including nutrition and psychosocial counselling ( Table 2 ). Exercise administration occurred under supervision in two studies [ 14 , 23 ], unsupervised and home-based in two [ 8 , 36 ], and a mixture in two [ 27 , 46 ].Exercise modalities typically involved whole-body cardiorespiratory and strengthening exercises, programmed 3 to 5 times a week for a duration of 7 to 32 days. Adherence to prehabilitation was very high (all reported＞95%) for supervised exercise, however was not reported for any home-based components ( Table 2 ).

Diagnoses were similar between groups for pancreatic, hepatic,biliary, and “other” diagnoses ( Table 3 ). Prehabilitated patients underwent 370 (69.0%) pancreaticoduodenectomies, 152 (28.4%) hepatic resections, and 10 (1.87%) hepatopancreaticoduodenectomies compared to 291 (69.1%), 119 (28.3%), and 6 (1.43%) respective procedures for the control group ( Table 3 ). The numbers of patients who received neoadjuvant chemotherapy were similar between the groups (n= 33, 36.7% vs.n= 19, 36.5%), as was the rate of biliary stenting (n= 61, 64.9% vs.n= 66, 67.3%) ( Table 3 ).

Two studies reported results for in-hospital postoperative complications [ 36 , 46 ], two reported follow-up durations (12 and 24 weeks after the date of surgery) [ 8 , 14 ], and two did not report follow-up duration ( Table 1 ) [ 23 , 27 ].

Outcome results

Exercise-based prehabilitation resulted in a statistically significant postoperative LOS reduction with a mean difference of 5.20 days (95% CI: -9.89 to -0.51,I2= 94%,P= 0.03; Fig. 2 A). When the study by Kitahata et al. [46] . was excluded from analysis, this mean difference decreased to 1.85 days (95% CI: -5.70 to 1.99,I2= 92%,P= 0.34). Analysis was completed with this study excluded as it represented an outlier and because it was the only study that included non-oncologic resections.

Complication rates (OR = 0.70, 95% CI: 0.39 to 1.26,I2= 46%,P= 0.23; Fig. 2 B) and major complication rates (OR = 0.83, 95%CI: 0.60 to 1.14,I2= 0%,P= 0.24; Fig. 2 C) trended towards nonstatistically significant decreases with prehabilitation. Table S3 reports all complications reported by included studies.

Mortality decreased in patients treated with exercise-based prehabilitation, though this was again not statistically significant(OR = 0.67, 95% CI: 0.17 to 2.70,I2= 0%,P= 0.57; Fig. 2 D). Only 4 deaths for each group were reported out of the 919 included patients.

Subgroup analysis demonstrated that those undergoing pancreatic surgery had a non-significant 15.77-day LOS reduction (95% CI:-38.11 to 6.57,I2= 95%,P= 0.17; Fig. 2 E) and those undergoing hepatic resection had a non-significant 0.29-day increase (95% CI:-3.69 to 4.27,I2= 80%,P= 0.89; Fig. 2 F).

Fig. 1. PRISMA summary of the systematic review selection process identifying studies evaluating exercise-based prehabilitation in hepato-pancreatico-biliary patients. RCT:randomized control trial.

Table 1 Characteristics of included studies and patient demographics reported from included studies.

Physical performance and QoL outcome measures were highly heterogeneous limiting the ability to pool data. Individual study results are presented in Table 3 . Ausania et al., Dunne et al.,and Nakajima et al. all evaluated the effect of prehabilitation from baseline to preoperatively. They all demonstrated improved cardiopulmonary function (e.g., VO 2 at peak exertion, VO 2 at anaerobic threshold, walk tests) and improved strength (e.g.,knee extension strength, handgrip strength) ( Table 4 ) [ 8 , 23 , 27 ].Additionally, Nakajima et al. demonstrated alleviated sarcopenia, with decreased fat mass, increased muscle mass, and increased fat-free mass from baseline to preoperatively ( Table 4 ) [8] .Kaibori et al. demonstrated that similar cardiopulmonary function, strength, and sarcopenia benefits are also recognized postoperatively [14] .

Only Dunne et al. and Wang et al. reported QoL outcomes( Table 4 ). Wang et al. reported the prehabilitation group had better Functional Assessment of Cancer Therapy-Hepatobiliary (FACT-HEP)preoperatively than the control group [36] , whereas the Dunne et al. demonstrated statistically significant improvements in postoperative 36-Item Short Form Survey (SF-36) scores for the prehabilitation group, with no change for the control group [23] .

Discussion

Despite ongoing improvements to surgical techniques and perioperative treatment, morbidity following HPB surgery remains substantial. We demonstrate that optimizing patient condition prior to surgery through exercise-based prehabilitation may lead to improved postoperative outcomes with decreased postoperative LOS,and trends suggesting decreased complications, major complications, and mortality. We also provide a pathophysiological pathway for these improvements with improved cardiopulmonary capacity,strength, lean body mass, and QoL following exercise prehabilitation.

The 5.20-day mean reduction in LOS, and 1.85-day reduction when one outlier study was removed [46] , found in our study are consistent with the 2.2-day LOS decrease for colorectal surgery patients and 2.86-day LOS decrease for abdominal surgery patients after exercise-based prehabilitation [ 25 , 26 ]. Given the aberrant results presented by Kitahata et al. [46] and their non-oncologic patient population, the 1.85-day reduction is likely more representative of the true LOS reduction. Despite the 1.85-day decreases LOS being non-statistically significant, we believe that this result does represent a clinically-significant reduction. Patients undergoing pancreatic surgery (15.77-day decrease) seemed to have a greater effect from prehabilitation than those undergoing hepatic resection (0.29-day increase), though the former results were again skewed by results from Kitahata et al. [46] .

Table 2Exercise-based prehabilitation interventions administered by included studies.

Fig. 2. Forest plots evaluating postoperative outcomes between patients treated with exercise-based prehabilitation compared to those undergoing standard care. A: Postoperative length of stay; B: complications; C: major complications; D: mortality; E: length of stay subgroup analysis of patients undergoing pancreatic surgery; F: length of stay subgroup analysis of patients undergoing hepatic resection.

Table 3 Surgical details and perioperative procedures reported from included studies for patients treated with exercise-based prehabilitation compared to standard care.

While non-statistically significant, our results also suggest small trends toward reduced complication rates, major complication rates, and mortality. Hughes et al. [26] found very similar complication reductions (OR = 0.63,P= 0.005) following prehabilitation for major abdominal surgery, which increases our confidence in these results despite non-statistical significance and moderate heterogeneity. Additionally, included studies demonstrated improved preoperative and postoperative physical performance, providing a pathophysiological mechanism to support improved outcomes. Of the three studies reporting the effects of prehabilitation on preoperative function, all of them showed improvements in cardiopulmonary capacity prior to surgery [forced expiratory volume in 1 s (FEV1), hand-grip, forced vital capacity (FVC), or walk tests]. Dunne et al. demonstrated that improvements were contrasted with physiologic declines for patients treated with standard care [23] , and others have specifically demonstrated improved preoperative strength and alleviated sarcopenia [ 8 , 27 , 47 ].Improvements such as these, rather than functional decline suggest that prehabilitation may limit the development of preoperative frailty [48] ; however, it remains unclear whether these interventions can achieve equal benefit for patients who are already frail.A recent study by Tsukagoshi et al. suggests that frail HPB patients may confer an even greater benefit from prehabilitation, but they evaluated only laboratory values, not clinical outcomes [49] . Clinical trials are ongoing to evaluate prehabilitation focused specifically for frail patients with gastrointestinal malignancy (Clinical-Trials.gov: NCT04715581) and results of this study will be helpful in determining benefits of prehabilitation for this high-risk group.Nevertheless, improvements in preoperative physical performance such as those found in the included studies have previously been associated with postoperative morbidity and mortality in frail populations when assessed using similar cardiorespiratory [50–53] and body composition [ 18 , 19 , 47 ] measurements to those used in this study. This provides additional confidence in our results, as we demonstrate improved outcomes and a pathophysiological mechanism for these changes.

QoL improvements with prehabilitation were also recognized,with both studies evaluating these outcomes demonstrating benefits ( Table 4 ). This is an important consideration for HPB patients who experience significant cancer stigma, which is a self-inflicted hopeless feeling that leads to decreased QoL [54–56] . Several recent studies have demonstrated an association between decreased mobility after HPB surgery and worse QoL [ 56 , 57 ]; our results are in keeping with that showing improved QoL measures with better preoperative mobility and strength. Engaging and empowering these patients in their recovery with exercise prehabilitation may provide them a sense of hope and reduce that stigma leading to an improved QoL.

While our results suggest benefits of prehabilitation, an optimized prehabilitation protocol has yet to be demonstrated. Several studies have reported pre-established protocols [ 4 , 36 , 58 ], and the American Cancer Society and American College of Sports Medicine have developed exercise guidelines for cancer survivors which could potentially benefit HPB patients perioperatively [ 59 , 60 ]. De-spite these guidelines, prehabilitation remains highly heterogeneous. Included studies demonstrate that supervised protocols encourage high rates of adherence and increased adherence with supervision has also been shown by others [ 14 , 23 ]. However, homebased self-administered programs cost less and permit greater training volumes, and thus dosage responses, given that patients adhere to recommendations and possess the prerequisite selfefficacy to independently complete programming [ 8 , 61 ]. Notably,two studies with solely home-based prehabilitation did successfully demonstrate reduced LOS, complications, and major complications with improved strength and alleviated sarcopenia [ 8 , 36 ]. In terms of frequency, most included studies utilized prehabilitation for at least 60 min per session for at least three times a week;however, it should be noted that patients must still be able to tolerate the exercise programming administered. For example, there are studies expecting diseased patients of a median 70 years to complete a program, twice daily, involving 30 min stationary cycling, 20 min on treadmill, 20 min stair-stepping, and 600 squats –a challenging prescription even for healthy, young adults [ 21 , 46 ].Intensity was often described as moderate to high but was rarely objectively defined and future work should attempt to evaluate this topic. In the included studies, cardiopulmonary exercise was primarily stationary bikes or walking while body weight exercises were used for strengthening, which achieved adequate prehabilitation outcomes and were also easily accessible. Future studies creating an optimized and accessible standardized protocol would be of great benefit.

Table 4 Physical performance and quality of life outcomes reported from included studies for patients treated with exercise-based prehabilitation compared to standard care.

Finally, prehabilitation may confer unique benefits to two specific HPB patient populations due to their adequate preoperative delay prior to resection: those undergoing neoadjuvant chemotherapy and preoperative biliary stenting. The utility of neoadjuvant chemotherapy in HPB oncology remains uncertain, although current research suggests that it may benefit those with borderline resectable pancreatic adenocarcinoma [ 62 , 63 ]. A significant concern for these patients is the toxicity and physical decline experienced during neoadjuvant treatment [ 12 , 62 , 64 ]. However, neoadjuvant treatment creates a logistically fortuitous window after diagnosis and before surgery to conduct prehabilitation. Numerous previous studies have demonstrated the benefits of prehabilitation during neoadjuvant therapy for other malignancies [ 9 , 10 , 12 , 28 , 35 , 65 ].For HPB patients, Dunne et al. [23] demonstrated improvements for all outcome measures following prehabilitation with the majority of included patients receiving neoadjuvant therapy. Parker et al. demonstrated alleviated sarcopenia in patients undergoing prehabilitation during neoadjuvant therapy as compared to standard care and Baimas-George et al. also recently demonstrated a significant reduction in frailty index after prehabilitation during neoadjuvant therapy [ 47 , 48 ]. Similarly, for patients receiving preoperative biliary stenting, delaying surgery for 4 weeks following could reduce perioperative complications, again providing time for prehabilitation [66] . Nakajima et al. [8] demonstrated impressive improvements following a home-based prehabilitation program, for a patient population where a majority received biliary stenting;in numerous studies other beneficial results may also be recognized [ 23 , 27 , 35 , 36 , 46 ]. These studies provide initial experience and outcome support for prehabilitation during neoadjuvant therapy or after biliary stenting and future studies evaluating this unique patient subgroup would be of interest.

The limitations to our study should be considered. Prehabilitation is a novel topic and research in this area has only recently begun to be published, with all included studies being released within the past 8 years. Because of this, only 957 patients were evaluated in this study. Additionally, only 385 patients are included in subgroup analyses with Kitahata et al. [46] excluded. This sample size is limited by current data and additional studies are needed to further evaluate prehabilitation for HPB patients. Also because of the novelty of prehabilitation in HPB, exercise protocols vary tremendously and this likely explains significant heterogeneity of our results. To limit this, we applied a random-effects model in this meta-analysis and have taken this into consideration within our conclusions. Considering these sample size and heterogeneity limitations, conclusions from this study should be evaluated with caution. In most included studies, multimodal prehabilitation was utilized, making it difficult to determine the independent effect of specific exercise interventions on outcomes. This is a limitation that is consistently noted across other prehabilitation studies [ 25 , 26 ]. Other limitations should also be recognized. HPB patients with excessive frailty and poor function are likely underrepresented as they were often excluded from most studies. These frail patients may potentially benefit most from prehabilitation but are often excluded [67] . Significant differences also exist between healthcare systems and discharge practices producing LOS variability [ 8 , 46 ].

In conclusion, this systematic review and meta-analysis evaluates surgical outcomes following exercise-based prehabilitation for various HPB patients. Prehabilitation was associated with a statistically significant 5.20-day LOS reduction compared to standard care, and non-statistically significant but clinically-significant 1.85-day reduction after outliers were removed. Analysis of postoperative complication rates, major complication rates, and mortality demonstrated no statistically significant differences, but trended towards benefit from prehabilitation. Exercise prehabilitation also improved cardiopulmonary reserve, alleviated sarcopenia, and improved QoL in individual study results. However, prehabilitation was heterogeneous and often multi-modal, limiting any definitive conclusions. Despite limitations, the potential benefits of exercisebased prehabilitation remain high with limited patient risk and should be considered for preoperative HPB patients. Future studies focusing on an optimized, feasible, and well-defined exercise program are needed to allow for more informed conclusions in evaluating this treatment option and help enable implementation into standard practice.

Acknowledgments

None.

CRediT authorship contribution statement

Andy Deprato : Data curation, Formal analysis, Methodology,Writing - original draft. Kevin Verhoeff: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Supervision, Writing - review & editing.Kieran Purich : Data curation, Methodology, Supervision, Writing- review & editing. Janice Y Kung : Data curation, Methodology,Resources, Software, Writing - review & editing. David L Bigam :Conceptualization, Methodology, Resources, Supervision, Writing -review & editing. Khaled Z Dajani : Conceptualization, Methodology, Resources, Supervision, Writing - review & editing.

Funding

None.

Ethical approval

Not needed.

Competing interest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.hbpd.2022.02.004 .