Meta-analysis of the impact of prehabilitation on patients undergoing upper gastrointestinal tract tumor surgery

Abstract

BACKGROUND

Upper gastrointestinal cancer (UGIC), including esophageal and gastric cancers, poses a major global health challenge due to its high morbidity and mortality. During the preoperative period, patients often face functional decline, malnutrition, and psychological stress, which can impair recovery. Prehabilitation, a multidisciplinary preoperative intervention, shows promise in optimizing patients' physical and mental status.

AIM

To evaluate the impact of prehabilitation on patients undergoing UGIC surgery and provide a basis for implementation of the prehabilitation compound plan.

METHODS

A computerized search of databases including Web of Science, PubMed, EMBASE, The Cochrane Library, Cumulative Index to Nursing and Allied Health Literature, China National Knowledge Infrastructure, Wanfang, and Chinese Science and Technology Journal Database was used to collect clinical trials on the impact of prehabilitation on patients undergoing UGIC surgery. After screening, a meta-analysis was conducted using Review Manager 5.0 software, and linear regression analysis was performed on the prehabilitation duration and outcome indicators.

RESULTS

A total of 13 clinical trials were ultimately included, with 8 literature quality evaluations at A level and 5 literature quality evaluations at B level. The meta-analysis results showed that compared with conventional nursing, the prehabilitation group had higher six-minute walk distance, lower postoperative complications and mortality rates, and shorter hospital stays, with statistically significant differences; there were no statistically significant differences in intensive care unit monitoring time and albumin levels between the two groups; regression analysis between prehabilitation duration and outcome indicators showed no significant relationship.

CONCLUSION

Prehabilitation can improve the perioperative functional ability of patients with UGIC and promote postoperative recovery, but its impact on nutrition, psychology, and quality of life needs to be further explored through more high-quality trials; in addition, further research is needed on the prehabilitation time, location, and specific plan.

Key Words: Prehabilitation; Upper gastrointestinal cancer; Meta analysis; Clinical trials; Quality of life; Nutrition

Core Tip: This meta-analysis evaluates the impact of prehabilitation-a multidisciplinary intervention including exercise, nutrition, and psychological support-on patients undergoing upper gastrointestinal cancer surgery. Prehabilitation significantly improved functional status, reduced postoperative complications and mortality, and shortened hospital stays. No significant improvement was observed in nutritional indicators or intensive care unit stay. These findings support integrating prehabilitation into perioperative care for upper gastrointestinal cancer, although further high-quality trials are warranted to refine protocols and assess psychological and nutritional outcomes.

INTRODUCTION

Upper gastrointestinal cancer (UGIC) primarily includes esophageal cancer, gastric cancer, and tumors at the gastroesophageal junction. In 2020, the incidence and mortality rates of esophageal and gastric cancer were among the top ten globally and in China, imposing significant medical and economic burdens on society and families[1-3]. Currently, surgery remains the primary treatment modality. However, there is a “window period” between tumor diagnosis and surgery, during which patients often experience a decline in physical activity, increased nutritional risk, and pronounced psychological issues - problems associated with living with cancer[4-6]. This period also presents individual challenges such as insufficient preoperative preparation, increased surgery-related complications, and higher perioperative mortality rate[7,8]. Therefore, targeted preoperative interventions that address functional decline, nutritional depletion, and psychological distress are urgently needed to improve perioperative outcomes in UGIC patients.

In recent years, prehabilitation - defined as a series of interventions implemented before surgery, including exercise, nutritional support, and psychological counseling - has emerged as a promising approach for enhancing surgical readiness and promoting recovery[9]. These interventions are often based on enhanced recovery after surgery (ERAS) protocols and delivered by multidisciplinary teams. While some studies show no difference between prehabilitation and standard care[10], others report significant improvements in physical function, postoperative outcomes, and even survival among patients receiving prehabilitation programs[10,11]. Despite growing international attention, prehabilitation for UGIC remains underexplored, and most existing studies vary widely in intervention type, timing, and evaluation standards. Moreover, no comprehensive meta-analysis has yet focused exclusively on UGIC surgical patients. The Chinese guidelines for ERAS (2021 edition) also include prehabilitation as an important part of preoperative preparation[12], although the evidence supporting it remains relatively weak. Currently, there is a lack of standardized protocols for the timing and methods of prehabilitation, both domestically and internationally. Furthermore, there has been no meta-analysis specifically focused on prehabilitation for UGIC.

This study aims to conduct a meta-analysis of published clinical trials examining the effects of prehabilitation on UGIC surgical patients. By evaluating outcome indicators such as six-minute walk distance (6MWD), complication rate, hospital stay duration, intensive care unit (ICU) monitoring time, mortality, and nutritional status, we hope to provide robust evidence to guide clinical implementation of prehabilitation protocols in this high-risk patient group.

MATERIALS AND METHODS

Literature search strategy

A comprehensive computerized search was performed across multiple databases, including Web of Science, PubMed, EMBASE, The Cochrane Library, Cumulative Index to Nursing and Allied Health Literature, China National Knowledge Infrastructure, Wanfang, and Chinese Science and Technology Journal Database, to identify clinical trials investigating the effects of prehabilitation in patients with UGIC. The search covered all records available from the inception of each database up to December 31, 2022. In English-language databases, the search strategy included the following keyword combinations: (1) Upper gastrointestinal, esophageal, esophagus, esophagogastric, gastric, stomach, duodenal, duodenum; (2) Cancer, neoplasms, carcinoma, tumor, malignancies; and (3) Prehabilitation, pre-rehabilitation, preoperative, perioperative care. The final search string was constructed as: (1), (2), and (3). For Chinese-language databases, equivalent and context-appropriate Chinese search terms were used.

Selection criteria

Inclusion criteria: (1) Population: UGIC patients with a willingness for prehabilitation and stable medical conditions; (2) Intervention and comparison: The experimental group received prehabilitation management, using at least one prehabilitation method such as exercise, nutrition, or psychology; the control group received conventional preoperative care as recommended by the ERAS guidelines; (3) Outcomes: 6MWD, incidence of complications, patient hospitalization duration, ICU monitoring time, postoperative mortality rate, patient nutritional indicators: Albumin (ALB) (g/L); and (4) Studies: Published clinical trials.

Exclusion criteria: (1) Duplicate publications; (2) Animal experiments, mechanistic studies, pharmacology or drug synthesis research; (3) Intervention measures not classified as prehabilitation; (4) Incomplete outcome measures; and (5) Low-quality score in the Joanna Briggs Institute (JBI) literature quality assessment.

Literature screening and data extraction

This study involved two researchers independently screening the literature, extracting data, and cross-checking. Discrepancies in literature evaluation were resolved through discussion, third-party consultation, or by emailing the original authors. For literature screening, non-clinical trial literature was excluded based on Medical Subject Headings terms and search strategy. Initially, titles were read for preliminary screening to exclude irrelevant literature. Abstracts and full texts were further reviewed for final inclusion. If necessary, study authors were contacted via email or phone to obtain important information for this research. Data extraction included: (1) Basic information of included studies: Study title, first author, publishing journal, impact factor, whether it was a clinical trial, etc; (2) Baseline characteristics of the study subjects and intervention measures; (3) Key elements for bias risk assessment; and (4) Outcome measures of interest, measurement methods, and result data.

Bias risk assessment in included studies

Two researchers independently assessed the risk of bias in the included studies and cross-checked the results. The risk of bias was evaluated using the Australian JBI tools for intervention studies randomized controlled trials (RCTs)[13], quasi-experimental studies[14]), categorizing each study’s quality into three levels: A (low bias, fully meeting standards), B (moderate bias, partially meeting standards), C (high bias, fully meeting standards; such studies were excluded). The detailed results of the risk of bias assessments for the included randomized controlled trials and quasi-experimental studies are presented in Supplementary Tables 1 and 2, respectively.

Statistical analysis

Statistical analysis was performed using RevMan 5.3. For continuous outcomes, mean difference (MD) was used as the effect measure; for dichotomous outcomes, odds ratio (OR) with 95% confidence intervals (CI) was calculated using the Mantel-Haenszel method. Heterogeneity was assessed using the χ² test (significance level α = 0.1) and quantified with the I² statistic, with values of 25%, 50%, and 75% indicating low, moderate, and high heterogeneity, respectively. A fixed-effect model was applied when I² < 50%, and a random-effects model was used if I² ≥ 50% after excluding substantial clinical heterogeneity. Subgroup analysis, sensitivity analysis, or descriptive analysis was conducted to explore sources of heterogeneity where appropriate. Publication bias was assessed using funnel plots. To evaluate whether the duration of prehabilitation influenced outcomes, linear regression analysis was performed using SPSS 26.0.

RESULTS

Literature screening process and results

A total of 4519 potentially relevant articles were initially identified through database searches. After a rigorous, multi-step screening process, 13 studies encompassing 1322 patients met the inclusion criteria and were ultimately included in the meta-analysis. Among them, 8 were RCTs[10,11,15-21], while the remaining 5 were quasi-experimental studies[22-26]. The detailed screening process is illustrated in Figure 1, and the basic characteristics of the included studies are summarized in Table 1. Based on the JBI quality assessment, all five quasi-experimental studies were rated as A-level (low risk of bias), whereas the eight RCTs were rated as B-level (moderate risk of bias). A detailed summary of the quality assessment results is provided in Table 1.

Figure 1 Literature screening flowchart. CNKI: China National Knowledge Infrastructure; VIP: VIP Chinese Scientific Journals Database.

Table 1 Basic characteristics of the included studies.

Ref.	Quality	Study design	Number (T/C)	Age (T/C, years), mean ± SD	Participants	Intervention		Time	Location	Outcome1
						T	C
Minnella et al[16], 2018	B	RCT	26/25	67.3 ± 7.4/68 ± 11.6	Esophagogastric resections	Aerobic: 30 minutes moderate - Borg: 12-13, continuous training: 3 day/week. Resistance exercise: Moderate intensity; 3 sets of 8-12 reps with resistance band 1 day/week for 4 weeks)	Standard care as per ERAS	36 days (17-73)	Home	1, 2, 3, 4, 5, 10
						Nutrition: Provide dietary advice and whey protein supplementation
Swaminathan et al[11], 2020	B	RCT	29/29	56.03 ± 14.95/56.82 ± 11.27	Gastrectomy	Start inspiratory muscle training based on motivational lung capacity measurement 7 days before surgery, twice a day	Standard care as per ERAS	7 days	Hospital	3, 5
Mazzola et al[22], 2017	A	UCT	41/35	75 (44-90)/75 (59-91)	Frail patients undergoing surgery of the upper gastrointestinal tract	Exercise: Inspirational lung capacity measurement 3 times per day, 30 minutes moderate intensity walking 3 times per week, lasting for 2 weeks	Receiving no special preoperative care	5-7 days	Hospital	3, 4, 5, 10
						Nutrition: 5-7 days prior to surgery, oral immune nutrition supplementation ± nasal jejunum feeding ± total parenteral nutrition
Valkenet et al[17], 2018	B	RCT	121/120	63.7 ± 7.5 /62.7 ± 8.9	Esophagectomy	Inspiratory muscle training ≥ 2 weeks, twice a day	Standard care as per ERAS	≥ 2 weeks	Home	3, 4, 5, 8, 11
Allen et al[10], 2022	B	RCT	26/28	65 ± 6/62 ± 9	Esophagogastric resections	Exercise: 2-3 times a week, with 1 hour of aerobic, resistance, and flexibility training each time	Standard care as per ERAS	15 weeks	Home + hospital	4, 5, 8, 9, 11
						Nutrition: Calorie and protein intake under the guidance of a nutritionist
						Psychological: Targeted medical guidance 6 times
Yamana et al[18], 2015	B	RCT	30/30	68.33 ± 7.64/65.9 ± 9.5	Esophagectomy	Inspiratory, abdominal and lower limb muscle training ≥ 1 week	Receiving no special preoperative care	≥ 7 days	Hospital	4, 5
Christensen et al[23], 2019	A	UCT	21/29	63.9 ± 8.2/65.5 ± 7.3	Surgery for adenocarcinoma of the gastro-esophageal junction	Approximately 75 minutes of high-intensity aerobic exercise and resistance training twice a week	Standard care as per ERAS	9 weeks	Home + Hospital	4, 5, 8
Liu et al[19], 2020	B	RCT	26/24	62.04 ± 5.12/64.58 ± 5.87	Esophagectomy	Preoperative nutritional support one week in advance and home enteral nutrition one month in advance	Standard care as per ERAS	5w	Home+ Hospital	(3, 4, 5, 6, 7, 8, 10, 11)
Dai et al[20], 2022	B	RCT	48/46	55.21±9.87/ 54.73±9.26	Gastrectomy	Exercise: Aerobic exercise for about 20 minutes three times a week, a group of resistance exercise per day, 2-3 respiratory muscle training sessions per day	Standard care as per ERAS	6-8 days	Hospital	1, 2, 3, 4, 5, 6, 7, 8
						Nutrition: Calorie and whey protein intake under the guidance of a nutritionist
						Psychology: Personalized Psychological Counseling
Li et al[24], 2021	A	UCT	84/84	59.23 ± 6.13/58.1 ± 4.76	Surgery for esophageal gastric junction adenocarcinoma	Exercise: Approximately 30 minutes of moderate intensity aerobic exercise three times a week, 30 minutes of resistance training once a week, and 2-3 respiratory muscle training sessions per day	Standard care as per ERAS	9-11 weeks	Home + Hospital	1, 2, 4, 5, 6, 7
						Nutrition: Medication treats symptoms, adjusts body mass, and balances diet
Guo et al[21], 2018	B	RCT	50/50	62.94 ± 6.35/64.88 ± 5.85	Esophagectomy	Exercise: Climbing stairs, walking, using an inhaler to assist in inspiratory muscle training, and abdominal breathing training	Standard care as per ERAS	7-9 days	Hospital	4, 5, 11
						Nutrition: Enteral nutrition after admission, carbohydrate intake 12 hours before surgery
						Psychology: Pre health compound case education
Xie et al[25], 2019	A	UCT	128/128	61.54 ± 8.12/62.82 ± 7.93	Esophagectomy	Exercise: Coughing, blowing balloons, walking, and training with respiratory function exercise equipment	Standard care as per ERAS	Hospitalization until surgery	Hospital	5, 6, 7, 11
						Nutrition: Enteral nutrition 3 days before surgery, carbohydrate intake 12 hours before surgery
						Psychology: Pre health compound case education
Li et al[26], 2019	A	UCT	34/30	60.43 ± 10.21/61.30 ± 11.12	Esophagectomy	Exercise: Climbing stairs, 6-minute walk test, and respiratory function training	Standard care as per ERAS	7 days	Hospital	4, 5, 6, 7
						Nutrition: High protein diet should be given 7 days before surgery, semi liquid and enteral nutrition powder should be given 3 days before surgery, and carbohydrates should be supplemented 1 day and on the same day before surgery

¹Outcome: 1: Preoperative six-minute walk distance (m); 2: Six-minute walk distance at 4 weeks after surgery (m); 3: Length of stay (day); 4: 30-day mortality after surgery; 5: Post-operative complications (Clavien-Dindo ≥ III); 6: Albumin at 1 day before surgery (g/L); 7: Albumin at 7 days post-surgery (g/L); 8: Quality of life; 9: VO₂ max at anaerobic threshold: Maximal oxygen uptake at anaerobic threshold; 10: Re-admission rates; 11: ICU stay.

T: Trial; C: Control; RCT: Randomized controlled trial; UCT: Uncontrolled clinical trial; ERAS: Enhanced recovery after surgery.

Meta-analysis results

Preoperative and postoperative functional ability - 6MWD: A total of 3 studies[16,20,24] compared preoperative and postoperative 6MWD between the two groups. Using a fixed-effect model (I² = 0%), the preoperative 6MWD in the prehabilitation group was higher than that in the control group (MD = 49.80, 95%CI: 33.20-66.39, P < 0.001) (Figure 2A). The postoperative 6MWD in the prehabilitation group was also higher than that in the control group (MD = 71.79, 95%CI: 56.06-87.52, P < 0.001) (Figure 2B).

Figure 2 Forest plot comparing preoperative six-minute walk distance (A) and postoperative six-minute walk distance (B) between the two groups. CI: Confidence interval.

Incidence of complications: A total of 11 studies[10,11,16,18-24,26] reported the incidence of complications, A fixed-effect model (I² = 0%) was used for the meta-analysis. The incidence of complications in the prehabilitation group was lower than in the control group (OR = 0.46, 95%CI: 0.33-0.65, P < 0.001) (Figure 3).

Figure 3 Forest plot comparing postoperative complication rates between the two groups. CI: Confidence interval.

Hospital stay and ICU care duration: Seven papers[10,16,17,19,20,22,23] reported on the length of postoperative hospitalization of patients. Using a fixed-effects model (I² = 44%), hospital stay for the prehabilitation group was shorter than that for the control group (MD = -0.90, 95%CI: -1.28 to -0.52, P < 0.001) (Figure 4A). In a comparison between the two groups, a total of 5 studies[10,17,19,21,25] which reported the ICU care duration were included. Using a random-effects model (I² = 92%), the difference in ICU care duration between the two groups was not statistically significant (MD = -0.47, 95%CI: -1.06 to 0.11, P = 0.11) (Figure 4B).

Figure 4 Forest plot comparing hospital stay durations (A) and intensive care unit care durations (B) between the two groups. CI: Confidence interval.

Subgroup analysis by geographic region (Figure 5) revealed that the Asian subgroup showed a significant reduction in ICU stay (MD = -6.01, 95%CI: -10.27 to -1.73, P = 0.006), while the European subgroup showed no significant difference (MD = -9.18, 95%CI: -35.89 to 14.52, P = 0.45). This suggests prehabilitation may be more effective in reducing ICU stay for Asian patients.

Figure 5 Forest plot for the subgroup analysis by geographic region of intensive care unit care durations between the two groups. CI: Confidence interval.

30-day postoperative mortality rate: Five studies[16,17,22-24] compared postoperative mortality between the two groups of patients. Using a fixed-effects model (I² = 0%), the mortality rate in the prehabilitation group was lower than that in the control group (OR = 0.30, 95%CI: 0.11-0.81, P = 0.02) (Figure 6).

Figure 6 Forest plot comparing postoperative mortality rates between the two groups. CI: Confidence interval.

Nutritional indicator: ALB: Five studies[19,20,24-26] reported preoperative ALB levels in patients. Using a random-effects model (I² = 92%), the difference in preoperative ALB between the two groups was not statistically significant (MD = 1.96, 95%CI: -0.15 to 4.07, P = 0.07) (Figure 7A). Four studies[19,20,25,26] compared postoperative ALB levels between the two groups of patients. Using a random-effects model (I² = 96%), there was no statistically significant difference in ALB levels one week post-operatively between the two groups (MD = 3.55, 95%CI: -0.00 to 7.10, P = 0.05) (Figure 7B).

Figure 7 Forest plot comparing preoperative albumin (A) and albumin levels one week post-operatively (B) between the two groups. CI: Confidence interval.

A sensitivity analysis of the meta-analysis comparing ALB between the two groups was performed. After excluding the study by Xie et al[25], the results indicated higher preoperative ALB in the prehabilitation group, using a random-effects model (MD = 2.58, 95%CI: 0.84-4.33, P = 0.004). The results for ALB one week postoperatively remained stable after sequentially excluding individual studies.

Publication bias analysis

Publication bias was tested for the main outcome indicators. Funnel plots for the incidence of complications (Figure 8A) and hospital stay duration (Figure 8B) showed a largely symmetrical distribution on both sides, suggesting a lower likelihood of significant publication bias. For other outcome indicators, the number of studies included was too small, making the construction of funnel plots less meaningful.

Figure 8 Publication bias test for incidence of complications and hospital stay duration. A: Funnel plot of publication bias for the odds ratio estimates; B: Funnel plot of publication bias for the mean difference estimates. OR: Odds ratio.

Linear regression analysis

Finally, we performed a linear regression analysis between the implementation time of prehabilitation and outcome indicators. The results in Table 2 show that no significant correlation was found between any of the indicators and the implementation time of prehabilitation (P > 0.05).

Table 2 Linear regression analysis of the implementation time of pre-rehabilitation and outcome indicators.

Outcome	R2	P value
Preoperative 6MWD	0.61	0.43
6MWD at 4 weeks after surgery	0.72	0.36
Post-operative complications	0.20	0.23
Length of stay	0.03	0.72
ICU stay	0.71	0.16
30-day mortality after surgery	0.22	0.43
ALB at 1 day before surgery	0.49	0.30
ALB at 7 days post-surgery	0.52	0.48

6MWD: Six-minute walk distance; ICU: Intensive care unit; ALB: Albumin.

DISCUSSION

This study conducted a comprehensive systematic review and meta-analysis to investigate the impact of prehabilitation on perioperative outcomes in patients who underwent UGIC surgery. The findings suggest that prehabilitation significantly improves pre- and postoperative functional status, reduces postoperative complications and 30-day mortality, and shortens hospital stay, thereby has substantial clinical value. However, its effects on ICU stay, nutritional status, psychological outcomes, and quality of life (QoL) remain inconclusive and require further analysis considering study heterogeneity and methodological variation.

Prehabilitation enhances physical reserves and postoperative resilience by integrating planned exercise training, nutritional support, and psychological interventions. Our findings indicate that patients in the prehabilitation group had significantly greater 6MWD both before and after surgery compared to controls, demonstrating a clear functional benefit. These results align with previous studies by Piraux et al[27] and Michael et al[28], which reported that structured prehabilitation improves physical capability and accelerates recovery. Additionally, prehabilitation significantly reduced postoperative complication rates and 30-day mortality and shortened hospital stays, underscoring its key role within the ERAS framework.

Although the overall effect on ICU length of stay was not significant, subgroup analysis revealed a statistically significant reduction in ICU stay among Asian populations, but not in European cohorts. These regional differences may be related to disparities in adherence, intervention intensity, or ICU management practices. Given the limited number of studies and the high heterogeneity (I² = 92%), further validation in large-scale studies is warranted.

Prehabilitation showed a trend toward improved preoperative nutritional status but did not significantly affect postoperative ALB levels, which showed high heterogeneity (I² = 96%). This may result from inconsistent timing of assessment, varied nutritional interventions, metabolic changes post-surgery, and patient-specific factors such as age, cancer type, and treatment. Notably, ALB is a nonspecific marker influenced by inflammation and fluid status, limiting its accuracy for nutritional evaluation[29]. Future studies should adopt more reliable tools like the Patient-Generated Subjective Global Assessment or prealbumin[30].

Although some studies, such as the study by Allen et al[10], reported improvements in psychological well-being using the Beck Inventory, there was a lack of consistent psychological interventions and assessment tools across studies, limiting meta-analytic capability. Given the prevalence of psychological distress among UGIC patients, future research should incorporate validated scales like the Hospital Anxiety and Depression Scale, Patient Health Questionnaire-9, or Generalized Anxiety Disorder 7-item Scale to quantify psychological benefits. Likewise, although some trials reported improved QoL, heterogeneity in assessment tools (e.g., European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30, short form 36 health survey, Functional Assessment of Cancer Therapy-General) prevented data pooling[23]. Standardized cancer-specific instruments such as the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30[31] are recommended for future studies.

The duration of prehabilitation ranged from 1 week to 4 weeks, with interventions conducted at home or in hospitals and utilizing either single- or multi-modal strategies. Linear regression found no significant association between duration and outcomes, suggesting that the implementation of prehabilitation itself may be more important than the exact length of intervention. In China, short preoperative hospitalization and patient reluctance to delay surgery-especially in rural populations-pose practical challenges[28]. Thus, developing simplified, low-cost, and telehealth-supported home-based programs tailored to China’s healthcare context is essential[32,33].

The ERAS guidelines, including the 2021 Chinese edition, recommend multimodal prehabilitation combining exercise, nutrition, and psychological support[12]. International prospective studies such as the PREHAB trial are exploring standardized structures for implementation[34]. In China, an integrated “hospital-community-home” management model could enhance accessibility and scalability. We propose that future protocols emphasize multidisciplinary collaboration, flexible delivery strategies, and patient-centered design to establish robust, standardized, and context-sensitive prehabilitation pathways.

The included studies had diverse surgical backgrounds, tumor stages, and histories of chemotherapy and radiotherapy. Although all followed guidelines for perioperative care, baseline consistency is unclear. The patient populations, tumor characteristics, and intervention modalities varied significantly among studies, contributing to notable clinical heterogeneity and limiting the consistency and generalizability of the findings. Despite our attempt to perform subgroup analyses based on tumor stage, surgical methods, and types of prehabilitation interventions, these efforts were hindered by insufficient and inconsistent reporting in the included studies. Most studies did not provide detailed stratification information, making subgroup comparisons infeasible. Additionally, there were differences in the evaluation indicators among the studies, and inconsistent use of outcome metrics further limited data comparability. The methodologies of the included studies were not always clearly described. Despite attempts to contact original authors via email or phone, some included literature lacked clear descriptions, potentially leading to implementation and measurement biases. Moreover, the relatively small sample sizes of individual studies weaken the overall statistical power and reliability of the pooled estimates. The limited number and size of the included studies may also introduce bias.

These limitations may affect the internal and external validity of the study findings. For instance, unclear intervention protocols or outcome definitions in original studies could introduce selection and measurement bias, while heterogeneity in patient populations may limit the generalizability of the conclusions to broader clinical settings. Moreover, potential publication bias and the absence of long-term outcome data constrain the ability to assess the durability of prehabilitation benefits. Therefore, further high-quality, large-scale randomized controlled trials are needed to confirm our findings.

Given the limitations identified in this meta-analysis, future research should aim to establish standardized, multimodal prehabilitation protocols tailored to UGIC patients. High-quality, multicenter RCTs are needed to determine the most effective timing and components of prehabilitation. In addition, more robust tools and unified outcome measures should be developed to assess the psychological and QoL effects of prehabilitation, which remain insufficiently explored. Studies focusing on home-based vs hospital-based delivery models will also be essential to guide practical implementation strategies, particularly in different healthcare systems.

CONCLUSION

Prehabilitation management in UGIC patients has high applicability for improving preoperative functional reserves and promoting postoperative recovery. This study indicates that the impact of prehabilitation on nutrition, psychology, and QoL requires further research. Future studies should consider incorporating validated and sensitive tools for nutritional assessment (e.g., Patient-Generated Subjective Global Assessment, prealbumin) and psychological evaluation (e.g., Hospital Anxiety and Depression Scale, Patient Health Questionnaire-9), to ensure robust measurement of these critical dimensions. Additionally, there are no uniform standards for the timing and setting of prehabilitation. More large-sample, high-quality RCTs are needed to verify the effects of different prehabilitation timings and settings on UGIC patients, while also focusing on changes in patients' nutritional status, psychological well-being, and QoL.