Globally, the incidence of colorectal cancer is growing. Over the lifespan, approximately 1 in 23 men and 1 in 25 women will develop colorectal cancer.1 Despite this, if detected early, surgery alone or in combination with chemotherapy or radiotherapy can provide excellent survival outcomes.2 However, colorectal cancer surgery carries significant postoperative morbidity, consequently increasing the length of hospital stay, slowing recovery and increasing health care costs.3 Therefore, colorectal cancer treatment is associated with a significant burden on patients and the healthcare system. There is a need to reduce morbidity in this population.

Recently, preoperative modifiable risk factors, including poor physical, nutritional and psychological aspects, have been associated with increased risk of postoperative morbidity.4,5,6,7 This has resulted in the development of many prehabilitation randomised controlled trials aimed at optimising preoperative patient health in an attempt to reduce postoperative morbidity. Recent randomised trials have focused on unimodal or multimodal interventions, including exercise, nutrition and/or psychological support. In other cancers, there is strong evidence suggesting that prehabilitation is effective in reducing postoperative complications and length of hospital stay.8,9

In colorectal cancer, previous systematic reviews have focused on specific populations (e.g., frail patients), introduced high risk of bias with the inclusion of non-randomised trials, included trials with active controls (i.e., rehabilitation after surgery), explored the effectiveness of single preoperative interventions only (i.e., exercise), did not follow recommendations on appraising and synthesising the evidence and/or are outdated.8,10,11,12,13 In addition, two major randomised trials have been published in the last 12 months (i.e., PHYSSURG-C and PREHAB).14,15 Thus, further analysis is warranted.

This study aimed to systematically review the effectiveness of prehabilitation modalities on reducing postoperative morbidity and length of hospital stay in patients undergoing colorectal cancer surgery. Improved understanding on the effectiveness of prehabilitation interventions will provide better recommendations for the management of colorectal cancer patients, future prehabilitation guidelines and on the development of future research.

Methods

Protocol and Registration

This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, and methods recommended by the Cochrane Handbook for Systematic Reviews of Interventions.16,17 The review protocol was registered on the Open Science Framework platform (https://osf.io/dashboard; https://doi.org/10.17605/OSF.IO/VW72N).

Study Selection

Studies meeting the following eligibility criteria were included: (1) randomised controlled trials describing the effectiveness of prehabilitation (including exercise, nutrition and/or psychological interventions) in patients undergoing colorectal cancer surgery, when compared with control (i.e., usual care, minimal intervention, or an active intervention not affecting the outcomes of interest [e.g., delivered 30-days postoperatively]); and (2) reported postoperative complications and/or length of hospital stay outcomes. Trials reporting on mixed populations (e.g., >5% of patients not having colorectal cancer) and studies published as abstracts from conference proceedings were excluded.

Data Sources and Searchers

A comprehensive search strategy was developed with the support of an experienced librarian from the University of Sydney. The search included a combination of text words and Medical Subject Headings for ‘randomised controlled trials’ AND ‘preoperative’ AND ‘cancer’ AND ‘prehabilitation’ (including ‘exercise’ OR ‘nutrition’ OR ‘psychological’ interventions) AND ‘postoperative outcomes’ (including ‘complications’ or ‘length of hospital stay’). Citation tracking of the included trials and previous literature reviews were also conducted. The search was employed in the CINAHL (Ovid), Cochrane Library, Medline (Ovid), PsycINFO (Ovid), AMED (Ovid) and Embase (Ovid) databases in April 2023.

Two review authors (DS and FN or WJ) independently screened study titles, abstracts and full text of all identified studies using the Covidence systematic review software (www.covidence.org). Any disagreements between the review authors were resolved by discussion with a third author (MS or CK).

Data Extraction and Risk-of-Bias Assessment

Two independent review authors (DS and FN or WJ) performed data extraction using a standardised data extraction sheet. Any disagreements between the review authors were resolved by discussion with a third author (MS or CK). Data extracted included study characteristics, details of prehabilitation intervention and control groups, and outcomes of interest. Data reported as median (and interquartile range, 95% confidence interval [CI], range or p-value) were converted to mean and standard deviation using the recommendation strategies of the Cochrane Handbook.17 When appropriate, for trials presenting three arms (e.g., two active interventions and one control), the two active interventions were combined.

Risk of bias was assessed using the revised Cochrane risk-of-bias tool for randomised controlled trials (RoB 2).18 Two review authors (DS and FN or WJ) independently assessed risk of bias for all included trials. Disagreements between the review authors were resolved by discussion with a third author (MS or CK). Overall risk of bias was judged as ‘low risk’, ‘some concerns’, or ‘high risk’ of bias.

Data Synthesis and Analysis

Postoperative complication rates were reported as the number of patients presenting with at least one complication and were used to calculate the pooled treatment effect (relative risk and 95% CIs). Relative risk < 1 favoured prehabilitation interventions. Length of hospital stay was reported as mean and standard deviation and was used to calculate the pooled treatment effect (mean difference and 95% CIs). Positive mean differences favoured prehabilitation interventions. Data were pooled using random effects meta-analysis when there was acceptable homogeneity across outcomes and prehabilitation interventions. All meta-analyses were performed using Comprehensive Meta-Analysis software (Biostat Inc., Englewood, NJ, USA). When data could not be included in a meta-analysis, descriptive summary tables were performed.

The quality of evidence for each outcome was evaluate using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach and rated as ‘high quality of evidence’, ‘moderate quality of evidence’, ‘low quality of evidence’ or ‘very low quality of evidence’.19 The quality of evidence was downgraded by one level accordingly to the following criteria: (1) risk of bias (≥25% of included trials presenting one or more domains classified as high risk of bias); (2) inconsistency (statistically significant heterogeneity [I2 >50%] or ≤75% of trials with findings in the same direction); (3) imprecision (dichotomous outcomes with sample size <300 participants, or for continuous outcomes with sample size <400 participants); and (4) publication bias (publication bias identified by visual inspection of funnel plots if >10 trials were included). The indirectness criterion was not considered as we only included the colorectal cancer population with relevant outcomes and direct comparisons. For single trials with <400 participants, inconsistency and imprecision (i.e., sparse data) were downgraded and rated as ‘low quality’ evidence. The quality of the evidence could be further downgraded to ‘very low quality’ of evidence if risk-of-bias limitations were found.

Results

Study Selection

Of the 3963 studies identified in the initial search, 23 trials (including 2475 patients) met the eligibility criteria and were included (Fig. 1).14,15,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41

Figure 1
figure 1

PRISMA flow diagram

Full size image

Study Characteristics

The identified trials investigated the effectiveness of multimodal (exercise, nutrition and psychological interventions) and unimodal prehabilitation (exercise, or nutrition or psychological interventions). The sample size of the included trials ranged from 20 to 668 (average = 105 participants). Only eight trials had a sample of ≥ 100 participants.14,15,20,21,22,23,28,37,38 The average age of the included participants was 66 years, and the duration of the intervention for multimodal, exercise, nutrition and psychological trials was 4 weeks, 6 weeks, 1 week and 1 h, respectively. Detailed information on the included trials can be found in Table 1.

Table 1 Characteristics of the included trials
Full size table

Risk of Bias

Information on risk of bias of the included trials can be found in Table 2. Risk of bias due to ‘deviation from the intended intervention’ was one of the domains with increased high risk of bias, while the domain risk of bias in ‘measurements of the outcomes presented’ had the lowest risk of bias. Overall, all prehabilitation trials presented at least some risk of bias.

Table 2 Risk-of-bias summary of the included studies
Full size table

Multimodal Interventions

The effect of multimodal interventions was explored in three trials, including exercise, nutrition, and psychological support (N = 381).15,23,29 One trial reported the effectiveness of multimodal intervention on postoperative complications (N = 251), including any complications, ileus, severe (Charlson Comorbidity Index [CCI] < 20) complications, medical complications, surgical complications, and surgical and medical complications.15 No effect of multimodal prehabilitation on postoperative complications (i.e., any complication, ileus, or surgical) was observed (Table 3). However, low quality of evidence of a significant effect favouring multimodal prehabilitation over control was observed on severe (CCI <20) complication rate (relative risk 0.57, 95% CI 0.35–0.92), medical complication (relative risk 0.56, 95% CI 0.34–0.93) and medical and surgical complication (relative risk 0.39, 95% CI 0.16–0.96). Length of hospital stay was reported in three trials (N = 381).15,23,29 No effect of multimodal prehabilitation on length of hospital stay, when compared with control (mean difference 0.62, 95% CI − 0.87 to 2.11), was observed. The quality of evidence was rated as very low for the length-of-stay outcome (Fig. 2 and Table 3).

Table 3 Summary of findings and quality of evidence assessment (GRADE)
Full size table
Fig. 2
figure 2

Mean difference for postpoperative length of hospital stay (days) in contrilled trials on efficacy of prehabilitataion for patients undergoing colorectal cancer surgery. Positive values favours prehabilitation interventions.

Full size image

Exercise Interventions

Three trials investigated the effect of preoperative exercise on postoperative complications and length of hospital stay (N = 729).14,26,32 Preoperative exercise was not effective on reducing postoperative complications (i.e., any complication, pneumonia, wound infection, urinary infection, pulmonary embolism, infections complications and anastomotic leak) and length of hospital stay. The quality of evidence ranged from moderate to very low for all outcomes reported (Fig. 3 and Table 3).

Fig. 3
figure 3

Relative risk for number of postoperative complications in controlled trials on exercise prehabilitataion for patients undergoing colorectal cancer surgery. Values < 1 favours excercise prehabilitation interventions

Full size image

Nutritional Interventions

A total of 16 trials investigated the effect of preoperative nutrition interventions on postoperative complications and length of stay.20,21,22,24,25,28,30,31,33,34,35,36,37,38,39,40,41 Pooling estimates from seven trials (N = 679) provided moderate quality of evidence of a significant effect favouring preoperative nutrition intervention over control on postoperative infectious complications (relative risk 0.65, 95% CI 0.45–0.94) (Fig. 4).20,22,28,30,31,33,41 Preoperative nutrition (14 trials, N = 1084) was effective in reducing postoperative length of hospital stay when compared with control (mean difference 0.87, 95% CI 0.17–1.58).20,21,24,25,28,30,31,33,34,35,36,37,38,39,41 The quality of evidence was rated as low for the length-of-stay outcome. No other significant effect was observed (Fig. 2 and Table 3).

Fig. 4
figure 4

Relative risk for number of postoperative complications in controlled trials on efficacy of nutrition prehabilitataion for patients undergoing colorectal cancer surgery. Values < 1 favours nutrition prehabilitation interventions

Full size image

Psychological Interventions

A single trial investigated the effect of preoperative psychological interventions on complications and length of stay (N = 75).27 No effect was observed on complications and length of hospital stay. The quality of evidence was low for all outcomes reported (Table 3).

Discussion

This systematic review and meta-analyses found moderate-quality evidence that preoperative nutrition intervention was effective in reducing infectious complications by 35% and length of hospital stay by approximately 1 day in patients undergoing colorectal cancer surgery. The effect of multimodal, exercise and psychological prehabilitation interventions on postoperative outcomes was uncertain due to the limited number of trials, heterogeneity in reported outcomes, and the low quality of evidence. Currently, there is limited confidence in the effect estimates of prehabilitation following colorectal cancer surgery and the results should be interpreted with caution.

The number of randomised controlled trials investigating the effectiveness of prehabilitation on postoperative outcomes of patients undergoing cancer surgery has increased drastically over the last decade. This has provided a window of opportunity to conduct systematic reviews and meta-analyses on the available evidence. Recently, a number of systematic reviews have been published aimed at synthesising the effects of prehabilitation on preoperative functional capacity, postoperative outcomes and quality of life. A review conducted by Bausys et al. summarised the current evidence on prehabilitation in the management of colorectal cancer patients.11 Of the 21 articles identified, 10 were either non-randomised controlled trials or retrospective studies. That review reported that most of the individual studies demonstrated at least some positive effects of prehabilitation on patients’ physical, nutritional, or psychological status and in reducing postoperative morbidity. Interestingly, in the current review, when postoperative outcomes were pooled within different prehabilitation modalities, most of the meta-analyses performed demonstrated no effect. Multimodal prehabilitation trials would expect to have a synergistic effect on outcome improvement, especially when compared with unimodal interventions such as exercise or nutrition alone. The difference between the results of the two reviews may be due to the bias introduced by the inclusion of non-randomised studies in the previous review. This was further evidenced by the systematic review and meta-analysis conducted by Chang et al., where the effect of prehabilitation on frail colorectal cancer patients was described.10 The initial significant effect of prehabilitation on postoperative complications (odds ratio 0.51, 95% CI 0.34–0.78) and length of hospital stay (standardised mean difference − 0.34, 95% CI − 0.46 to − 0.23) when randomised and non-randomised studies were included disappeared when only randomised trials were pooled (odds ratio 1.04, 95% CI 0.23–4.64; and standardised mean difference − 0.14, 95% CI − 0.44 to 0.16, respectively).10

Previous systematic reviews investigated the effect of other preoperative interventions, including exercise, nutrition, or psychological support. In the review performed by Gillis et al., pooled outcomes of six nutrition prehabilitation studies (including randomised trials and cohort studies) demonstrated a significant reduction in length of hospital stay by almost 3 days when compared with control.42 In the review performed by Falz et al. short (< 3 weeks) and long-term (≥ 3 weeks) preoperative exercise interventions had no effect on postoperative complications and length of hospital stay following colorectal cancer surgery.43 The evidence from previous psychological prehabilitation reviews is in line with the current findings of this review.44 Despite the number of reviews available in the literature, most applied different methodological approaches, including study designs that would introduce high risk of bias within the pooled estimates. In addition, other reviews included active ‘control’ groups (e.g., exercise prehabilitation) that were introduced early in the postoperative period, potentially influencing postoperative outcomes, such as complication and length of hospital stay.12

The risk of developing a postoperative complication following colorectal cancer surgery is highest in the first 30 postoperative days.45 This is a critical determinant of recovery, long-term outcomes (including quality of life) and treatment costs. Our review found that preoperative nutrition interventions significantly reduced the rates of postoperative infectious complications and length of hospital stay following colorectal cancer surgery. Unfortunately, due to the limited evidence, the effectiveness of other prehabilitation modalities is still lacking. Molenaar et al. conducted a trial investigating the effectiveness of multimodal prehabilitation, and reported a significant effect, when compared with control, on rates of severe complications, medical complications, and combined medical and surgical complications.15 Future multimodal trials will allow for data pooling, which will enhance the quality of the current evidence. Within the preoperative nutrition trials, immunonutrition and other oral nutrition supplementations (including carbohydrate loading) were the most tested interventions, however the dosage used varied across most trials. The duration of the nutrition interventions was also inconsistent, with interventions lasting from a couple of hours to a few weeks (7 days on average). Therefore, determining the prehabilitation standard of care for colorectal cancer patients undergoing surgical treatment is somewhat challenging within the current literature. In an attempt to guide future trials, a recent Delphi study identified key research priorities in prehabilitation.46 Further recommendations on the development of reporting guidelines, including prehabilitation intervention components, and reporting of core set outcomes are warranted.47 Thus, there is a need for the establishment of a core set of outcomes for prehabilitation and the development of prehabilitation guidelines. These steps would enhance the conceptualisation and design of future prehabilitation trials for patients undergoing colorectal cancer surgery. In addition, it is important to acknowledge that the implementation of enhanced recovery after surgery pathways has already led to significant improvements in surgical outcomes, such as reduced length of hospital stay. As a result, it may be more challenging to demonstrate further improvements in complication rates and length of stay when prehabilitation is added to an existing enhanced recovery after surgery program.

Some of the key strengths of this review included the adherence to the Cochrane recommendations; reporting according to the PRISMA statement; inclusion of the latest prehabilitation randomised controlled trials; use of two experienced reviewers to screen studies, extract data and assess risk of bias; use of the Cochrane RoB 2; and use of the GRADE approach to determine the quality of the evidence. Despite this, the current systematic review and meta-analysis has some limitations. Our comprehensive search identified a large number of trials but we may have missed trials stored in the grey literature, therefore publication bias cannot be ruled out. While we pooled trials according to their prehabilitation modalities, the type of intervention, frequency, intensity, duration, mode of delivery, adherence, and progression may vary across the trials. Therefore, identification of the optimal prehabilitation intervention may not be possible. In addition, despite the literature suggesting that prehabilitation should be employed at least 4 weeks before cancer surgery, some of the trials investigated the effectiveness of a single session intervention (e.g., 1 h). Finally, due to the small number of trials identified across each prehabilitation modality, the level of adherence to the interventions was not taken into consideration during the analysis. It is important to note that adherence reporting and definitions are essential for future research and meta-analyses.

The quality of the current prehabilitation literature has been previously described and included deviation from intended interventions, poor outcome reporting and definition, lack of publicly available protocols, underpowered trials, and changes in primary and secondary outcomes.48 Thus, there is an urgent need to further understand the barriers and facilitators to the conceptualisation of higher-quality prehabilitation trials. Furthermore, reporting of postoperative outcomes of the identified trials was somewhat heterogenous. While we pooled outcomes describing a specific complication, for some trials the definition of postoperative complications was either slightly different (e.g., using different complication classification systems) or not available (e.g., not reported within the published article or protocol [if available]). This should be taken into consideration when interpreting the results of this systematic review.

Conclusion

There is moderate quality of evidence that nutrition prehabilitation is effective in reducing infectious complications rates by 35%, and low quality of evidence in reducing length of hospital stay by approximately 1 day in patients undergoing colorectal cancer surgery. The benefit of other prehabilitation modalities, including multimodal, exercise and psychological interventions, is limited due to lack of randomised controlled trials, heterogeneity in reported outcomes, and the low quality of evidence. There are a number of registered prehabilitation randomised controlled trials that may change our confidence in results and effect estimates in the near future.