Effect of Exercise Training on Quality of Life after Colorectal and Lung Cancer Surgery: A Meta-Analysis

Simple Summary Surgery is the treatment modality associated with better long-term survival in patients diagnosed with lung cancer and colorectal cancer. However, as a result of surgery, patients experience a substantial decline in health-related quality of life (HRQoL) and increased fatigue symptoms. The purpose of this systematic review was to investigate the effect of pre- and/or postoperative exercise training on HRQoL and fatigue after surgical resection for lung and colorectal cancer. Our results showed that exercise training interventions improve the physical domain of HRQoL and reduce fatigue levels after lung cancer surgery, supporting its use to optimize patients’ recovery. No benefits were found on HRQoL and fatigue after colorectal cancer surgery. Abstract Surgical treatment affects health-related quality of life (HRQoL) and increases fatigue symptoms in patients with lung cancer (LC) and colorectal cancer (CRC). We aimed to systematically review the effect of exercise training on HRQoL and fatigue after LC and CRC surgery. Randomized controlled trials published before 21 March 2021, were searched in PubMed, Scopus, Web of Science, SPORTDiscus and PEDro. Eligible trials compared the effect of exercise interventions initiated preoperatively or in the first 3 months after surgery versus usual care on postoperative HRQoL and fatigue. Standardized mean differences (SMD) were pooled using random-effects models. Twelve studies with a total of 777 patients were included. In LC patients (10 studies, n = 651), exercise training in general led to a moderate improvement in the physical domain of HRQoL (0.68: 95% CI: [0.47; 0.89]) and a small reduction in fatigue levels after surgery (SMD = 0.28: 95% CI: [0.02; 0.53]), while no effects were found in other HRQoL domains. In CRC (two studies, n = 126), exercise training showed no effects on HRQoL and fatigue after surgery. Exercise training is an effective intervention to improve physical function and fatigue after LC surgery. Further studies are necessary to clarify the effects of exercise on HRQoL and fatigue after CRC surgery.


Introduction
The epidemiologic relevance of cancer is growing worldwide, with an incidence of 19.3 million new cancer cases and almost 10.0 million cancer deaths in 2020 [1].

Selection of Studies
The selection of studies started with an independent screening of titles and abstracts by two independent reviewers (PM and SP). If there were doubts about a potential article following the inclusion criteria or if there was incomplete information to make a clear inclusion or exclusion decision, that article was kept for the following phase (analysis of its full text). The second screening phase was also carried out independently by the same reviewers. Studies that were identified by mutual consent were included in the systematic review. In case of disagreement, a third reviewer (JC) was consulted and the final decision was based on the combination of the three opinions. A record of the excluded articles as well as the reasons for their exclusion was kept.
The Cohen's kappa coefficient was calculated to evaluate interrater reliability in the full text screening [48]. The kappa values can be interpreted as follows: values ≤ 0 as indicating no agreement and 0.01-0.20 as none to slight, 0.21-0.40 as fair, 0.41-0.60 as moderate, 0.61-0.80 as substantial, and 0.81-1.00 as almost perfect agreement [48].

Data Extraction
Data extraction was independently performed by two reviewers (PM and SP) with any discrepancies being resolved through discussion with a third reviewer (JC). Relevant extracted data were organized using standardized tables, that included the following topics: (1) Study characteristics; (2) Participants' demographic and clinical characteristics; (3) Exercise training dose based on the FITT principles (frequency, intensity, time, and type) [15] and adverse events during exercise interventions; (4) HRQoL/fatigue measures and results. When information regarding any of the above topics was unclear, the authors of the papers were contacted to provide details.

Quality Assessment
Methodological quality of the included studies was assessed by two reviewers (PM and SP), using the Physiotherapy Evidence Database scale (PEDro scale) [49]. Any discrepancies in judgements were resolved by consensus, with a third reviewer (JC) acting as a mediator if necessary. The PEDro scale comprises 11 items: Eligibility criteria, randomized allocation, hidden allocation, baseline comparison between groups, participants, physiotherapists and blind assessors, adequate follow-up, intention to treat the analysis, comparison between groups and point estimate and variability. Based on these items, a score of 0 to 10 is attributed to the RCTs [50]. According to the PEDro scale, studies with a score of 0 to 3 have a "poor" methodological quality, between 4 to 5 "reasonable", 6 to 8 "good" and 9 to 10 "excellent" [50].

Data Synthesis and Analysis
Meta-analyses were conducted if data from at least three studies or 100 patients could be combined, using standardized mean differences (SMDs) and 95% confidence interval (CI), to allow comparison of data from different instruments [51]. A random-effects model was used in the meta-analysis, as it combines sampling error and between-study variance to estimate effect size [52]. The following thresholds were used to interpret the effect sizes: <0.2 = trivial effect; 0.2-0.5 = small effect; 0.5-0.8 = moderate effect; >0.8 = large effect [53].
The statistical heterogeneity among studies was assessed using the I-squared (I 2 ), that represents the percentage of variation across studies that is attributable to heterogeneity rather than chance [54]. We adopted the following thresholds: I 2 = 25%: low heterogeneity; I 2 = 50%: moderate heterogeneity; I 2 = 75%: high heterogeneity [54]. If substantial statistical heterogeneity was detected, sensitivity analysis was undertaken by pooling the data of high-quality studies only (PEDro score ≥ 6). When a HRQoL domain was assessed by generic and cancer-specific questionnaires, we performed a subgroup analysis to examine if the exercise training effect was influenced by the type of instrument used.
When not enough data was provided in a study to estimate the exercise training effect, we contacted the authors to provide the required data (mean and standard deviation (SD)). When the post-intervention SD was not reported in a study and not provided by the authors, the pre-intervention SD was used.
All statistical analyses were conducted using the statistical software Comprehensive Meta-Analysis (CMA) (Biostat, Englewood, NJ, USA, version 3.3.070) [55]. A p-value of < 0.05 was considered statistically significant.

Publication Bias
The publication bias was calculated using the software CMA [55], generating a funnel plot by the standard error (SE) and the standard difference in means to determine whether the plot was balanced. The risk of publication bias was assessed by the visual inspection of the funnel plots and using Egger's test to provide a more objective and accurate assessment of funnel plot asymmetry than subjective visual assessment [56].
The kappa statistics of the agreement between the independent reviewers' screening of the full-text was 0.87, showing a strong agreement. The flowchart of the literature search, screening and selection process is presented in Figure 1. The kappa statistics of the agreement between the independent reviewers' screening of the full-text was 0.87, showing a strong agreement. The flowchart of the literature search, screening and selection process is presented in Figure 1.
Control groups received usual care without exercise training, consisting in routine physiotherapy treatments plus airway clearance techniques [24,58], routine outpatient appointments after discharge plus pain medication [64], and no advice about exercise training [57,60]. In addition to usual care, the control groups received general instructions on daily activities and weekly phone calls [67], a pedometer with instructions on how to record the total number of steps per day [61], a postoperative exercise intervention initiated 14 weeks after surgery (late exercise intervention) [63,66], and advice to perform physical activity [62]. Two studies reported the levels of physical activity, with no differences between exercise and control groups in light intensity activity and moderate-to-vigorous intensity activity [58,67].
Control groups received usual care, that consisted in no exercise training and instructions for the continuation of usual activities [59,65]. Levels of physical activity were measured in both studies conducted in CRC patients [59,65]. One study reported that at the of the exercise intervention, 88% of the participants in the exercise group performed at least 210 min of moderate to vigorous activity per week, in contrast with 56% in the control group [65]. In the other study [59], the authors reported a significant problem with exercise contamination, because both the exercise and control groups increased their levels of moderate to vigorous activity per week, with no appreciable differences between the two groups. At the end of the intervention, 76% of the participants in the exercise group and 51.6% of the participants in the control group reported more than 60 min of moderate to vigorous activity per week [59]. Table 1 describes the characteristics of the included studies. Significant differences between the exercise group and the control group; No between-group differences. * Measurement at 14 weeks after surgery (end of the early initiated exercise intervention), BFI (Brief fatigue inventory); C.G

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) [57]. The remaining studies implemented exercise training in the postoperative phase (n = 755), starting between the first postoperative day [24] and 73 days after surgery [59].
The exercise training sessions were performed with on-site supervision in eight studies, that were conducted at the hospital [57,62,64,65,67], in a rehabilitation center [63,66] and in a fitness center [60]. In two studies, the exercise program was initiated at the hospital for one week with on-site supervision (inpatient sessions), and continued at home for a period of 12-20 weeks [24,58], with three home visits [24] or weekly telephone supervision [58] by the research staff. In the remaining two studies, the exercise program was performed at home, with a weekly telephone supervision [61,64].
With respect to CRC patients, in one study (n = 93) [59] the exercise intervention consisted of aerobic exercise performed at home, with patients being allowed to choose the mode of exercise they preferred (e.g., swimming, cycling or walking). The frequency of  Exercise capacity EORTC-QLQ-C30 No between-group differences observed in HRQoL nt differences between the exercise group and the control group; No between-group differences. * Measweeks after surgery (end of the early initiated exercise intervention), BFI (Brief fatigue inventory); C.

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) [57]. The remaining studies implemented exercise training in the postoperative phase (n = 755), starting between the first postoperative day [24] and 73 days after surgery [59].
The exercise training sessions were performed with on-site supervision in eight studies, that were conducted at the hospital [ Exercise capacity EORTC-QLQ-C30 No between-group differences observed in HRQoL Significant differences between the exercise group and the control group; No between-group differences. * Measurement at 14 weeks after surgery (end of the early initiated exercise intervention), BFI (Brief fatigue inventory); C.G

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) [57]. The remaining studies implemented exercise training in the postoperative phase (n = 755), starting between the first postoperative day [24] and 73 days after surgery [59].
The exercise training sessions were performed with on-site supervision in eight studies, that were conducted at the hospital [57,62,64,65,67], in a rehabilitation center [63,66] and in a fitness center [60]. In two studies, the exercise program was initiated at the hospital for one week with on-site supervision (inpatient sessions), and continued at home for a period of 12-20 weeks [24,58], with three home visits [24] or weekly telephone su-   Significant differences between the exercise group and the control group; No between-group differences. * Measurement at 14

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) Exercise capacity EORTC-QLQ-C30 No between-group differences observed in HRQoL nt differences between the exercise group and the control group; No between-group differences. * Measweeks after surgery (end of the early initiated exercise intervention), BFI (Brief fatigue inventory); C.

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) [57]. The remaining studies implemented exercise training in the postoperative phase (n = 755), starting between the first postoperative day [24] and 73 days after surgery [59].
The exercise training sessions were performed with on-site supervision in eight studies, that were conducted at the hospital [57,62,64,65,67], in a rehabilitation center [63,66] and in a fitness center [60]. In two studies, the exercise program was initiated at the hospital for one week with on-site supervision (inpatient sessions), and continued at home for a period of 12-20 weeks [24,58], with three home visits [24] or weekly telephone supervision [58] by the research staff. In the remaining two studies, the exercise program was performed at home, with a weekly telephone supervision [61,64].
With respect to CRC patients, in one study (n = 93) [59] the exercise intervention consisted of aerobic exercise performed at home, with patients being allowed to choose the mode of exercise they preferred (e.g., swimming, cycling or walking). The frequency of

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) [57]. The remaining studies implemented exercise training in the postoperative phase (n = 755), starting between the first postoperative day [24] and 73 days after surgery [59].
The exercise training sessions were performed with on-site supervision in eight studies, that were conducted at the hospital [57,62,64,65,67], in a rehabilitation center [63,66] and in a fitness center [60]. In two studies, the exercise program was initiated at the hospital for one week with on-site supervision (inpatient sessions), and continued at home for a period of 12-20 weeks [24,58], with three home visits [24] or weekly telephone supervision [58] by the research staff. In the remaining two studies, the exercise program was performed at home, with a weekly telephone supervision [61,64].
With respect to CRC patients, in one study (n = 93) [59] the exercise intervention consisted of aerobic exercise performed at home, with patients being allowed to choose the mode of exercise they preferred (e.g., swimming, cycling or walking). The frequency of Exercise capacity EORTC-QLQ-C30 No between-group differences observed in HRQoL nt differences between the exercise group and the control group; No between-group differences. * Measweeks after surgery (end of the early initiated exercise intervention), BFI (Brief fatigue inventory); C.

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) [57]. The remaining studies implemented exercise training in the postoperative phase (n = 755), starting between the first postoperative day [24] and 73 days after surgery [59].
The exercise training sessions were performed with on-site supervision in eight studies, that were conducted at the hospital [57,62,64,65,67], in a rehabilitation center [63,66] and in a fitness center [60]. In two studies, the exercise program was initiated at the hos-

Intervention Characteristics
One study implemented the exercise training program before surgery (LC, n = 22) [57]. The remaining studies implemented exercise training in the postoperative phase (n = 755), starting between the first postoperative day [24] and 73 days after surgery [59].
The exercise training sessions were performed with on-site supervision in eight studies, that were conducted at the hospital [57,62,64,65,67], in a rehabilitation center [63,66] and in a fitness center [60]. In two studies, the exercise program was initiated at the hospital for one week with on-site supervision (inpatient sessions), and continued at home for a period of 12-20 weeks [24,58], with three home visits [24] or weekly telephone supervision [58] by the research staff. In the remaining two studies, the exercise program was performed at home, with a weekly telephone supervision [61,64].
With respect to CRC patients, in one study (n = 93) [59] the exercise intervention consisted of aerobic exercise performed at home, with patients being allowed to choose the mode of exercise they preferred (e.g., swimming, cycling or walking). The frequency of aerobic exercise was three to five times per week, 20-30 min per session, at an intensity of 65%-75% of predicted maximal heart rate, during 16 weeks [59]. In the other trial conducted in CRC (n = 33) [65], each exercise session was supervised by a physiotherapist, and the exercise training consisted of a combination of resistance exercise (1-2 sets of 10-20 repetitions, 45-75% 1-maximum repetition [1-RM] for all major muscle groups) and moderate-to-high intensity aerobic exercise (cycle-ergometer, alternating intervals at the first ventilatory threshold by three sets of 2 min, with lower intensity intervals by three sets of 4 min), 60 min per session, two sessions per week, over 18 weeks [65].
In LC patients, the total duration of the exercise interventions was approximately 4 weeks in the prehabilitation program (three to five sessions per week, an average of 16 sessions) [57]. In the postoperative phase, the duration of the exercise programs varied from 6 to 20 weeks, with two or three exercise sessions per week in the facility-based programs [60,[62][63][64]66,67] and five sessions per week in the home-based program [61].
The most prescribed type of exercise was aerobic training, which was present in all studies, with the exercise mode consisting of cycling in a cycle-ergometer [57,[62][63][64]66], walking [24,61] and a combination of cycling and walking [58,67]. The duration of aerobic exercise was reported in six studies and varied from 25 to 30 min [57,[61][62][63]66,67]. With respect to the exercise intensity, five studies integrated high-intensity interval training (HIIT), alternating low intensity intervals at 50-60% of peak workload with higher intensity intervals at 80% of peak workload or 85-100% of the maximal heart rate measured by a cardiopulmonary exercise test [57,60,63,66,67]. Six studies included aerobic continuous training with light intensity in the home-based program [61] and moderate-to-high intensity in the hospital-based exercise programs [24,58,62,64].
In two studies, respiratory muscle training was prescribed [60,62], with a dose of five sets of 10 repetitions followed by 1-2 min of unloaded recovery breathing at 50% of maximal inspiratory and expiratory pressures, using a respiratory muscle trainer device at a rate of 15-20 breaths/min [62].

Methodological Quality Assessment
The quality assessment showed a mean PEDro score of 6.3, indicating a good methodological quality. No studies were excluded based on methodological quality.
Five studies conducted postoperatively (initiated between the first postoperative day and 10 weeks after surgery) assessed the global HRQoL using as summary measure the total score of the FACT-L [61], the global quality of life of the EORTC-QLQ-C30 [24,62,67], and the total score of the Saint-George Respiratory Questionnaire (SGRQ) [64].

Discussion
This meta-analysis aimed to investigate the effect of exercise training on HRQoL and fatigue after CRC and LC surgery. An improvement was found in the physical domain of

Publication Bias
For LC, the funnel plot was asymmetrical in the mental domain of HRQoL, indicating the possibility of publication bias. The Egger's test showed an intercept result of −6.49 (SE = 0.88; 95% CI: [−9.28; −3.71]; t = 7.42; p = 0.01), confirming strong evidence of publication bias. No evidence of publication bias was found for fatigue and for the physical, mental, emotional and global domains of HRQoL (see Figure S2. For CRC, due to the limited number of included studies, we were not able to generate funnel plots.

Discussion
This meta-analysis aimed to investigate the effect of exercise training on HRQoL and fatigue after CRC and LC surgery. An improvement was found in the physical domain of HRQoL and in fatigue symptoms after LC resection, with a moderate and small effect, respectively. No evidence was found on the effects of exercise training in HRQoL and fatigue after CRC surgery.
Regarding LC, our results are in agreement with a previous meta-analysis that included exercise interventions undertaken in the first 12 months after surgery, which found beneficial effects in the physical domain of HRQoL, and no evidence of exercise-induced improvements in the mental component and global quality of life [68]. However, in contrast with that meta-analysis which found no effects of exercise training on fatigue [68], our results showed a significant reduction in fatigue in favor of the exercise groups. These results could be due to the inclusion of two studies in our meta-analysis (283 participants) that implemented the exercise interventions in the first month after surgery [61,63], a period when fatigue levels are more severe [4,5,7]. Therefore, considering that the effect of exercise interventions is greater in cancer patients with higher fatigue levels [69], the early initiation of exercise training after LC resection could be an important factor to mitigate this symptom. This hypothesis is corroborated by a large clinical trial which compared the effect of an exercise intervention initiated 14 days after LC resection in contrast with an intervention initiated at week 14, and found a significant difference in fatigue levels in favor of the early initiated exercise program [63].
Our review thereby contributes to the current literature by providing evidence that exercise interventions initiated preoperatively or in the first 3 months after surgery lead to a significant better physical function (pre-or postoperative exercise training) and reduce fatigue symptoms (postoperative exercise training) when compared to control groups, with no exercise training.
We have also found that the improvements in the physical domain appear to be smaller with cancer-specific questionnaires when compared to a generic instrument (SF-36). This variation in the exercise effect could be partially explained by the low correlation between the physical component summary of the SF-36 and the physical functioning of the EORTC-QLQ-C30, as observed in a prospective analysis of LC patients submitted to surgical resection, suggesting that the two instruments possibly reflet different aspects of the physical domain and may be complementary [70].
The beneficial effects of exercise training in the physical domain of HRQoL could be relevant both to address patients' needs and in terms of survival, because the deterioration in the physical function is perceived by LC patients as an extremely undesirable consequence of surgery [71], and a 10% decrease in this domain during the first 6 months after LC surgery was associated with 18% higher risk of death [37].
Contrary to the substantial deterioration in the physical function and fatigue, the mental/emotional domains tend to return to preoperative levels or even improve after LC surgery [5,7,11,72], showing less need for exercise interventions. With respect to global HRQoL, the only trial that found significant improvements of exercise training was implemented 14 days after surgery and used an LC specific module [66], reinforcing the importance of start exercise interventions earlier after surgery and choose a specific questionnaire, which is more accurate to detect perioperative changes in LC symptoms [70].
It should be emphasized that only one of the included studies was conducted in the preoperative phase [57], achieving significant improvements in the physical domain of HRQoL three months after LC surgery. These results, together with the findings of a previous meta-analysis showing that higher preoperative levels of physical activity were significantly associated with better HRQoL after oncological surgery [73], emphasize the need for further research to investigate if preoperative exercise training can prevent the detrimental impact of LC resection in HRQoL. This could be particularly relevant for the subgroup of patients receiving neoadjuvant therapy, that were excluded from this study, since neoadjuvant chemotherapy was associated with lower preoperative aerobic capacity and thus impact the short-and long-term outcome of tumor resection [74].
The small number of included studies and the heterogeneity of the exercise training prescribed prevent us from providing recommendations about a specific exercise dose to improve HRQoL and fatigue after LC surgery. Nevertheless, consistent improvements in physical function and fatigue were found in four studies combining HIIT plus resistance exercise [57,60,63,66] all of them presenting good methodological quality. Additionally, all these exercise interventions combining HIIT plus resistance exercise improved patients' aerobic capacity, a predictor of better prognosis after LC surgery [32,75,76] and a factor associated with better HRQoL in LC patients who previously completed curative intent treatment [22]. As shown in other cancer types, the therapeutic benefits of HIIT on HRQoL and fatigue may be mediated by improvements in aerobic capacity [77] and in a short time frame as the perioperative phase and the prescription of a higher exercise intensity could be a relevant factor to achieve central and peripheral physiological adaptations [78][79][80]. Furthermore, higher exercise intensities appear to protect against chemotherapy-related inflammation [81], a mechanism involved in the pathogenesis of fatigue [82], which could be clinically relevant for patients eligible to adjuvant treatment after surgery. This rationale is supported by a large clinical trial which found that a combination of high-intensity aerobic and resistance exercise was significantly more effective to reduce physical fatigue compared to low-to-moderate-intensity exercise in cancer patients undergoing (neo-)adjuvant treatment [83].
As for CRC, no effects of exercise training in postoperative HRQoL and fatigue were found. However, these results need to be interpreted with prudence because the effect estimates were only based on two clinical trials with inconsistent findings: a supervised intervention combining moderate-to vigorous-intensity aerobic plus resistance exercise achieved beneficial effects in fatigue symptoms and physical function [65], while a homebased intervention incorporating moderate-intensity aerobic exercise found no benefits in these clinical outcomes [59]. The lack of significant results in the home-based exercise intervention [65] may be partially explained by three factors: (1) High contamination rate (51.6%), with the participants in the control group significantly increasing their levels of moderate to vigorous physical activity, a factor associated with enhanced recovery in self-reported physical functioning after CRC surgery [84]; (2) Adherence rates, that were slightly lower than those observed in the supervised intervention (76% vs. 89%); (3) Type of exercise prescribed, since the clinical guidelines in oncology indicates that combining aerobic and resistance training leads to higher benefits in HRQoL compared with programs involving only aerobic or resistance exercise [15], and the efficacy of resistance exercise programs appears to be superior than aerobic exercise to reduce fatigue levels among cancer patients [85], possibly by the attenuation of muscle wasting and disruptions in muscle metabolism caused by chemotherapy, such as oxaliplatin [82].
In contrast with our results, a previous systematic review concluded that exercise training is effective for improving HRQoL and fatigue following a diagnosis of CRC [84]. However, it did not focus on pre-and/or postoperative exercise interventions, including patients undergoing cancer treatment and long-term survivors [84]. Our review adds knowledge to this field of research by underlining the need to conduct further studies to assess the effect of perioperative exercise interventions in these clinical outcomes.

Implications for Future Research
Considering that the most substantial deterioration in HRQoL and fatigue occurs in the early phase after surgery [5][6][7] and based on the positive association between preoperative physical activity levels and postoperative HRQoL [73], future high-quality trials should explore if prehabilitation exercise programs could prevent the deleterious effects CRC and LC surgery in these clinical outcomes.
Future clinical trials should also target patients with lower physical function and higher fatigue levels, the subgroup of individuals that appears to benefit most from exercise interventions [69], and use cancer-specific modules such as the EORTC-QLQ-LC13/CR29, which provides a more detailed evaluation of cancer-specific symptoms in comparison with generic questionnaires [70]. Finally, more research is warranted to identify the optimal exercise dose to improve HRQoL after CRC and LC surgery.

Strengths and Limitations
Strengths of the current review consist of the use of the PRISMA guidelines [85], the extensive search in multiple databases, the independent and robust screening process, the provision of a detailed description of the exercise interventions based on the FITT principles, and a comprehensive quantitative synthesis of the exercise training effects in the different domains of HRQoL.
There are, however, some limitations that need to be acknowledged. First, only two RCTs including CRC patients were eligible for inclusion, preventing us to provide more precise estimates of exercise effects on HRQoL and fatigue after surgery. Second, the majority of patients in the included studies had early-stage disease, being admitted to curative resection. Therefore, the exercise training effects may not be generalized to patients with advanced-stage disease selected for palliative surgery. Third, although the eligible studies had an overall good methodological quality, in five trials a concealed allocation was not carried out [59,61,63,64,66], which could lead to an overestimation of the exercise training effects [86]. Lastly, the possibility of language bias should not be neglected because only studies published in English were considered to inclusion [87].

Conclusions
The results of our meta-analysis indicate that exercise training is an effective intervention to improve the physical domain of HRQoL and reduce fatigue levels after LC surgery, compared with usual care. Considering that these dimensions are especially affected as a consequence of surgical resection, exercise training could be a relevant supportive intervention to target patients' needs. Further studies are necessary to clarify the effects of exercise training on HRQoL and fatigue after CRC surgery.