Browse the corpus

Bladder cancer (BC) is the 10 th most common malignancy globally, with about 614,000 new cases worldwide in 2022. It ranks 13 th in mortality (, ), and the number of new deaths has been rising annually, with an increase up to 151.2% (). Based on the tumor invasion depth into the bladder wall, BC includes muscle invasive and non-muscle invasive BC (). Due to prolonged hematuria, BC patients often experience hypoalbuminemia, anemia, and malnutrition. Moreover, protein breakdown caused by surgery and discomfort resulting from chemotherapy (e.g., nausea, vomiting, diarrhea) all raise the nutritional risk. In particular, the postoperative incidence rates of gastrointestinal complications and intestinal obstruction are 20% and 14%, respectively, in patients undergoing radical cystectomy (RC), which further exacerbates the nutritional risk (, ). Research suggests that the incidence of malnutrition in BC is 16%–55%, and patients with preoperative malnutrition have an approximately 22% higher rate of postoperative complications, with an infection rate of 15% (). Collectively, impaired nutritional status is a serious threat to the quality of life of BC patients. Nutritional management can improve patients’ nutritional status and facilitate postoperative rehabilitation (, ). However, only 33.7% of cancer patients with malnutrition are given nutritional interventions ().

Nowadays, nutritional management of BC is conducted primarily in specific stages within the hospital, and a whole-course nutritional management bridging hospital, community, and home settings remain scarce. Consequently, it is difficult to monitor the nutritional status in real time post-discharge and promptly adjust follow-up measures, creating a management gap (). Besides, the study population is limited, i.e., studies focus solely on patients undergoing RC, without covering radiotherapy, chemotherapy, and palliative care (). In clinical practice, patients and their families lack sufficient knowledge regarding BC nutrition and hold some misconceptions, further compromising the efficacy of nutritional interventions. It is obvious that BC patients are faced with serious nutritional issues, and scientific and systematic nutritional management protocols are lacking.

Maximizing the efficacy of interventions via whole-course nutritional management is recommended by guidelinecs (). Whole-course nutritional management is a systematic, dynamic, and individualized precision management pattern, where interventions are adjusted promptly based on the dynamic changes in nutritional status during hospitalization and post-discharge, continuously improving nutrition (). Available studies have applied the whole-course nutritional management to patients undergoing surgery for urological tumors, and the results confirmed its effects of ameliorating postoperative nutritional status and outcomes, and facilitating recovery (). Due to the prolonged course and complexity of BC, patients at different stages and undergoing different treatments have varying nutritional needs. Therefore, individualized precision nutritional interventions should be adopted based on disease progression, nutritional status, and treatment methods, necessitating a scientific, systematic, and precision nutritional management protocol.

Currently, relevant domestic and international guidelines and expert consensus focus primarily on treatment and medication and less on nutritional management (, ). Clinical medical workers’ capability in nutritional management and treatment remains insufficient. Therefore, scientific guidance for nutritional management and prognosis improvement is urgently needed. Using an evidence-based approach, this study intends to summarize the evidence for nutritional management at different stages of treatment for BC from high-quality domestic and international studies. The findings can provide medical workers with scientific and reliable evidence and references for the selection and formulation of interventions in clinical practice. Unlike previous research focusing primarily on perioperative nutritional management in BC patients, we adopted a whole-course perspective covering from admission to post-discharge care, including surgery, radiotherapy, chemotherapy, and palliative care. Stratified management was implemented based on nutritional status, and patients were classified for the first time into four strata according to both treatment modality and nutritional status and received precise and individualized nutrition support. Additionally, we incorporated traditional Chinese medicine (TCM) techniques as adjunctive interventions to facilitate gastrointestinal functional recovery.

We defined the clinical question by the PIPOST framework. P (Population): BC patients who undergo RC, radiotherapy, transurethral resection of the bladder, chemotherapy, or palliative care; I (Intervention): dietary guidance, nutrition education, enteral and parenteral nutrition, oral nutritional supplements (ONS); P (Professional): clinicians, dietitians, rehabilitation physicians, clinical pharmacists, and specialized nurses; O (Outcome): incidence of malnutrition; S (Setting): urology clinics, inpatient wards, and home. T (Study type): expert consensus, expert recommendations, guidelines, evidence summaries, systematic reviews, and randomized controlled trials.

We performed a full search using subject headings plus free-text terms. English terms included but not limited to “Bladder Cancer,” “Urinary Bladder Neoplasms,” “Transurethral Resection of the Bladder,” “Radical Cystectomy,” “Nutrition,” and “Nutrition Management.” Corresponding Chinese terms were utilized in Chinese databases. An information specialist from the university library with expertise in search methodology for systematic review assisted in developing the search strategy. Two reviewers (Xie Hongwen and Zhang Bingbing) who had received formal training in systematic review methodology drafted the initial strategy. To optimize sensitivity and specificity, the draft was subsequently refined by iterative pre-search and discussion with the research team and information specialist. The robustness of the strategy was guaranteed by (1) multiple rounds of pre-searches, (2) consultation with the information specialist, and (3) review and approval by the research supervisor (Xie Hongwen).

Using the “5S” pyramid of evidence-based resources, we performed a full search in multiple websites and databases, including one computerized decision-making system (UpToDate), 11 electronic databases (Embase, BMJ, PubMed, CINAHL, Web of Science, CNKI, Scopus, Cochrane, Sinomed, Wanfang, and JBI), 10 guideline websites (Registered Nurses’ Association of Ontario RNAO, Scottish Intercollegiate Guidelines Network SIGN, National Institute for Health and Care Excellence NICE, World Health Organization WHO, Agency for Healthcare Research and Quality AHRQ, New Zealand Guidelines Group NZGG, Guidelines International Network GIN, Yimaitong, National Comprehensive Cancer Network NCCN, and Australian Research Council NHMRC), and 14 major professional society websites (World Council of Enterostomal Therapists WCET, International Society of Urology SIU, European Associations of Urology EAU, European Society for Clinical Nutrition and Metabolism ESPEN, United Ostomy Associations of America UOAA, Wound, Ostomy and Continence Society WOCN, Canadian Urological Association CUA, Canadian Association for Enterostomal Therapy CAET, American College of Physicians ACP, American Urological Association AUA, American Society of Colon and Rectal Surgeons ASCRS, Chinese Urological Association, Chinese Urologic Surgeon Association CUSA, and the Cancer Nutrition Management Branch of the Chinese Society of Nutrition).

The search was conducted until December 2025, and an updated search was performed in January 2026 to identify any new publications, with one additional guideline included.

Two reviewers independently conducted screening and data extraction. They were both trained in the JBI evidence-based methodology and experienced in systematic reviews. Disagreements were settled by consensus or consultation with a third reviewer when necessary. After duplicate publications were removed by EndNote 20, the title and abstract were read based on the predefined eligibility criteria, and the full text was then examined. The studies excluded and the main reasons for exclusion are listed in . Besides, we checked the reference lists of all included studies to avoid any omissions. Using a standardized form, we extracted key data (title, year of publication, author/publisher, publication source, journal, and country/region).

Two reviewers (Zhang Bingbing and Chen Yuan) independently scanned the titles and abstracts of all records retrieved, and the full texts of potentially relevant studies were obtained and assessed by the two reviewers. Disagreements were settled by discussion with a senior reviewer (Xie Hongwen). When conflicting evidence emerged, high-quality, evidence-based, and recently published studies were prioritized. Clinical guidelines were assessed by the AGREE II covering 23 items in six domains: participant, rigor of development, scope and purpose, clarity of presentation, applicability, and editorial independence. Each item was scored on a seven-point scale (1: a lack of data or extremely poor quality, 7: a high-quality report meeting all criteria). The sum of the scores for all items represented the score for each domain. The raw score was standardized by the standard formula of AGREE II: standardized score (%) = (raw score–minimum possible score)/(maximum possible score–minimum possible score) × 100.

Since the criteria for standardized scores are not defined in the AGREE II User Manual, multiple quality assessment studies were extensively referenced (–). On this basis, the following recommendation criteria were defined: (1) Grade A (recommended): standardized scores ≥ 60% in all six domains; (2) Grade B (recommended with modifications): 30%–60% in most domains (≥3); (3) Grade C (not recommended): <30% in ≥3 domains. We included only guidelines of Grade A or Grade B. To ensure consistency, inter-rater reliability was quantified by the intraclass correlation coefficient (ICC): <0.40, poor; 0.40–0.75, moderate; >0.75, high (). Expert consensus and expert opinions were assessed by the JBI Critical Appraisal Checklist () from six items of credibility and logical development. A lower score corresponds to a higher level of evidence [Level 1 (the highest): evidence from systematic reviews of randomized controlled trials, Level 5 (the lowest): evidence from expert opinions or laboratory studies]. Two reviewers independently assessed each item as “Not Applicable,” “Yes,” “No,” or “Unclear.” Articles were included only if they met a predefined quality threshold, namely affirmative responses (“Yes”) to most items, emphasizing the logical derivation of conclusions. Evidence from UpToDate was directly considered high quality ().

Four reviewers independently assessed the included guidelines and best practice guidelines, while two reviewers independently assessed the included expert consensus and recommendations.

Recommendations regarding whole-course nutritional management of BC were acquired from the included studies by two reviewers. The results were subsequently cross-checked. Disagreements were settled by discussion or adjudication by a third reviewer. Both reviewers independently translated recommendations published in English into Chinese and compared them to reach a consensus. All recommendations (including both translations and original ones) were summarized according to the following principles: (1) for identical recommendations, the version with the clearest wording was selected (based on predefined criteria for completeness, accuracy, and clinical applicability); (2) for conflicting recommendations, their sources were traced, and the most recent and highest-quality evidence was prioritized; (3) for complementary recommendations, they were merged into a synthesized statement; and (4) for independent recommendations, the original statements were retained. This method was adapted to the established practice of evidence synthesis (). During summary, genuine conflicts were rare. When conflicts emerged, they were typically settled by reviewing publication dates, the level of evidence, and consistency with broader public health principles.

Based on the best evidence for BC across treatment methods and nutritional statuses, a first draft of the whole-course nutritional management intervention protocols and an expert survey questionnaire were prepared. Inclusion criteria for experts: intermediate professional title or higher; Bachelor’s degree or higher; work experience of at least 10 years in urology, nutrition, or TCM, or less than 10 years but currently teaching at a university with a solid theoretical foundation; with specialized knowledge in relevant fields; with informed consent and active participation. Questionnaires were issued and collected via paper or email. A two-round consultation was conducted at an interval of 8 weeks, with a 1-week collection period for each round. The evidence was revised and refined based on expert opinions.

We initially retrieved 2,602 records, of which 978 duplicates were removed. A total of 1,473 records were excluded following title and abstract reading, and another 135 were excluded after full-text review. A total of 16 studies were included finally, comprising one clinical decision, seven guidelines, three expert consensus articles, one expert recommendation, and four evidence summaries. The selection process is displayed in the PRISMA flowchart (), and the basic study characteristics are listed in .

Four guidelines were assessed as Grade A and three as Grade B. The inter-rater reliability was high, with an ICC of 0.905–0.979 (>0.75) for all guidelines. The intermediate professional title or higher; Bechelor’s in .

All three expert consensus articles and one recommendation met the predefined quality threshold and were therefore incorporated to extract evidence. The results of quality assessment are detailed in .

We summarized the evidence for whole-course nutritional management of BC across seven key dimensions: nutrition team formation and management processes, three-tier nutritional diagnosis, principles of nutrition support therapy, content of nutritional interventions, nutritional surveillance, health education, and nutritional follow-up. This generated 91 distinct statements of best practice evidence, and they were scored by the JBI Evidence Pre-grading System (2014) () (section “2 Materials and methods). The evidence for nutritional management is displayed in .

(1) Analysis of general information of experts: Experts’ academic backgrounds and professional titles were described by proportions and frequencies. Quantitative data were described by mean ± standard deviation. Two-round expert consultation was conducted, involving 17 experts in the first round, and 15 experts in the second round. General information of experts is shown in .

(2) In the first round of the survey, thirty questionnaires were issued, of which 17 were collected and confirmed as valid, with a valid response rate of 57%. A total of 13 experts provided feedback. In the second round, 17 questionnaires were issued, of which 15 were collected, with a valid response rate of 88%. Three experts provided feedback. Expert authority was measured by the authority coefficient (Cr), which was calculated by the coefficient of expert familiarity (Cs) and the coefficient of judgment basis (Ca): Cr = (Ca + Cs)/2, with Cr ≥ 0.7 considered acceptable. The Cr was 0.88 (>0.7), suggesting high expert authority and reliable consultation conclusions. The Kendall’s coefficients of concordance were 0.250 and 0.313, respectively, for the two rounds, indicating that the concordance in the second round improved over the first. The expert opinions were summarized, similar items were consolidated, and duplicates were removed. The research team discussed and responded to the expert opinions, then provided anonymous feedback to all experts for further consultation until opinions had no significant changes. The Ca, Cr, Cs, and summary and revision of expert opinions are tabulated in . The importance scores and feasibility scores for the 91 evidence statements are presented in .

The primary indicators of this summary (nutrition team formation and management processes, three-tier nutritional diagnosis, principles of nutrition support therapy, content of nutritional interventions, nutritional surveillance, health education, and nutritional follow-up) followed the logic of “organizational support → process guideline → diagnostic grading → treatment principle → intervention implementation →surveillance and follow-up.” The first round of consultation received endorsement from 17 experts, with a Cr of 0.88. The Kendall’s coefficients of concordance were 0.250 and 0.313, respectively, for the two rounds ( P < 0.01), indicating statistically significant but low-to-moderate coordination and acceptable content validity.

The secondary indicators were obtained within the primary indicator framework combined with disease stage and nutritional risk. The three-tier nutritional diagnosis system comprised screening, assessment, and comprehensive evaluation (). Nutritional Risk Screening 2002 (NRS2002) (for general patients) and modified Nutrition Risk in the Critically Ill (m-NUTRIC) (for critically ill patients) were recommended for screening. Patient-Generated Subjective Global Assessment (PG-SGA) was utilized for assessment, with Acute Gastrointestinal Injury (AGI) or AGI Ultrasonography (AGIUS) additionally for critically ill patients. The type and etiology of malnutrition were identified by comprehensive evaluation. Therefore, a strategy of “rapid screening for high-risk individuals → precise assessment of malnutrition severity → selection of treatment protocols” was established. Based on this, BC patients underwent stratified management (nutritional risk and no nutritional risk).

Bladder cancer patients (especially the older ones and those with chronic hematuria) often have developed malnutrition at the time of diagnosis. Screening should be conducted “immediately upon diagnosis” () and “within 24 h post-admission for hospitalized patients” (). For those identified as at no risk in the primary screening, a secondary screening is recommended 1 week later to prevent missed screening due to nutritional deterioration caused by examinations and treatment side effects during hospitalization. NRS2002 covers three dimensions (nutritional status impairment, disease severity, and age), and a total score ≥ 3 indicates nutritional risk. Although Malnutrition Universal Screening Tool (MUST) and Mini Nutritional Assessment-Short Form (MNA-SF) have advantages and disadvantages, and MNA-SF was recommended for older patients in the consultation letter, NRS2002 remained the preferred choice after discussion. It simultaneously assesses disease and nutritional status, and has a strong evidence base, wide global validation, and high sensitivity and specificity in cancer patients, so NRS2002 has been recommended by multiple guidelines (). Additionally, m-NUTRIC can be added for critical illness screening for patients with severe infection or multi-organ dysfunction, with a score ≥ 5 as the threshold for severe nutritional risk ().

Patient-Generated Subjective Global Assessment is preferred for BC patients positive in nutritional screening, as it is specifically designed for cancer patients, covers dimensions such as weight, feeding, symptoms, function, and metabolic needs, and is more sensitive than general tools. Meanwhile, Global Leadership Initiative on Malnutrition (GLIM) criteria are recommended for malnutrition diagnosis, which include three phenotypic criteria and two etiological criteria, and have a clear structure, making it suitable for Chinese patients. For critically ill patients, AGI (four grades based on clinical presentation; higher grades indicate poorer prognosis) or AGIUS may be utilized to dynamically assess gastrointestinal function. These tools may be selected or combined based on clinical conditions to precisely guide nutritional planning ().

The mechanisms of malnutrition in cancer patients are diverse, and individualized interventions are required: Anti-inflammatory nutrition (e.g., ω-3 fatty acids) is required for patients with high inflammation status, ONS or tube feeding for those with inadequate intake, and protein intake and resistance training for those with muscle wasting. The third-level comprehensive evaluation guides individualized treatment based on energy consumption, stress level, inflammatory burden, and metabolic status. Given the difficult operation and high cost of TNF-α, IL-1, and IBI tests in primary hospitals, the expert panel recommended four-level evaluation as an “ideal recommendation” for high-level medical centers or research settings. In routine clinical practice, CRP, albumin, prealbumin, and clinical NIS scores can serve as substitutes, with a note stating that “selected based on the capabilities of the healthcare institution.”