Prevalence of malnutrition and nutritional risk, and associated factors in adults with head and neck cancer: a systematic review and meta-analysis.

Background: Head and neck cancer (HNC) carries a high burden of malnutrition due to tumor location, treatment toxicity, and systemic inflammation. However, prevalence estimates vary across assessment instruments, and associated clinical factors have not been comprehensively quantified. This systematic review and meta-analysis estimated the prevalence of diagnostically defined malnutrition, nutritional risk, and PG-SGA-based nutritional categories and summarized factors associated with these outcomes in adults with HNC. Methods: A systematic literature search was conducted in the databases PubMed, Embase, Web of Science, Cochrane Library, CINAHL, CNKI, Wanfang, VIP, and CBM databases from inception to July, 2026. Research reports on malnutrition, nutritional risk and related factors for unhealthy weight in adults were included. Joanna Briggs Institute (JBI) critical appraisal checklist was employed to assess the methodological quality of prevalence studies, and the Quality in Prognosis Studies (QUIPS) tool was used for associated-factor studies. Pooled prevalence estimates were calculated using random-effects models and restricted maximum likelihood (REML) estimation in the logit scale. Generalized Linear Mixed Models and alternative transformations were used for sensitivity analysis. Results: Twenty studies involving 4,456 participants were included. The pooled prevalence of GLIM-defined malnutrition was 39.4% (95% CI: 30.4-49.1%; five studies, 1,158 participants), PG-SGA-based nutritional categories 57.8% (95% CI: 37.7-75.7%; six studies, 724 participants), and NRS 2002-defined nutritional risk 32.0% (95% CI: 18.9-48.7%; six studies, 1,423 participants). Considerable heterogeneity was observed across all assessment-specific analyses. Frequently reported associated factors included advanced age, lower body mass index, reduced serum albumin or prealbumin, treatment-related burden, radiotherapy exposure, and disease characteristics. Most studies had low-to-moderate or moderate methodological limitations, primarily related to observational designs, residual confounding, and incomplete reporting of analytical methods. Conclusion: Malnutrition and nutritional risk are common in patients with HNC, although estimates depend on the outcome definition and assessment instrument. GLIM provides a diagnostic definition of malnutrition, PG-SGA classifies nutritional status, and NRS 2002 identifies nutritional risk rather than confirmed malnutrition. Because the evidence was predominantly observational and geographically heterogeneous, the findings should be interpreted as descriptive and associative rather than causal. Risk-stratified nutritional screening and multidisciplinary assessment should therefore be considered in clinical practice. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/, identifier CRD420261418334.