Accurate detection of body composition changes is essential for evaluating nutritional interventions. In a previous 12-wk randomized controlled trial, adults with excess body weight who consumed a powdered meal replacement (PMR) under weight-stable conditions increased lean soft tissue assessed by dual-energy x-ray absorptiometry. This secondary analysis evaluated whether comparable changes were detected using other body composition techniques and whether hydration-related variation impacted tracking of fat-free mass (FFM) and related compartments. We hypothesized that multiple techniques would detect increases in FFM and hydration markers. Participants (n = 63) were randomized to either a PMR group, which added PMR twice daily to their usual diet, or a control group, which maintained their usual diet. Body composition was assessed using dual-energy x-ray absorptiometry, bioelectrical impedance analysis, air-displacement plethysmography (ADP), and an adapted 4-compartment (4C) model. Changes (95% confidence intervals) in the PMR group from baseline to week 12, compared with the control group, included increases in ADP-derived FFM (1.31 [0.51; 2.12] kg, P < 0.001), 4C-derived FFM (1.01 [0.25; 1.78] kg, P = 0.002), and total body protein (TBpro: 0.60 [0.14; 1.07] kg, P = 0.007). No significant changes were observed in fat mass or hydration-related compartments. In sex-stratified analysis, the PMR group increased ADP-derived FFM among males (1.70 [0.28; 3.12] kg, P = 0.021) and females (1.12 [0.05; 2.18] kg, P = 0.031). Overall, ADP was the most sensitive method for detecting longitudinal changes and performed comparably to the 4C model. Increases in FFM and TBpro appeared independent of hydration changes, suggesting true accretion of metabolically active tissue, likely muscle mass.
Objectives Artificial intelligence–assisted systems increasingly produce personalized meal plans, but their value depends on whether clinicians judge those plans credible and usable. We evaluated complete plans from a constrained artificial intelligence–assisted pipeline under blinded professional review. Methods In a cross-sectional exploratory web study, 54 eligible anonymous evaluator records from Brazilian dietitians contributed 130 evaluations of five matched plan pairs, with one pipeline-generated and one dietitian-designed plan per persona. Evaluators rated nine items across three domains and were offered an optional pairwise plan preference task. The overall score required all nine items and was available for 124 evaluations. Results Pipeline-generated plans received a mean rating of 3.67 (standard deviation, 0.57) (n = 59) compared with 3.11 (standard deviation, 0.73) for comparator plans (n = 65) (d = 0.86). Among 38 evaluator records with complete scores for both origins, 31 favored pipeline-generated plans, 6 favored dietitian-designed plans, and 1 had equal mean scores for both origins (paired t[37] = 5.32, P < 0.001). Pipeline-generated plans received higher ratings in all five persona pairs. This pattern was descriptive because evaluator records overlapped across pairs. Domain differences were +0.76 for nutritional adequacy, +0.69 for technical quality, and +0.21 for practical applicability. In the optional task, 32 of 42 eligible pairwise choices selected the pipeline-generated plan as more appropriate. Conclusions Under blinded presentation, practicing dietitians gave plans from the constrained pipeline a mean rating above the neutral anchor of the 5-point scale. These findings support prospective evaluation of a structured first draft workflow, with required clinician edits and patient-relevant outcomes as endpoints.
Objective This study aimed to investigate the incidence and clinical impact of hospital-acquired malnutrition (HAM) and develop a machine learning model for early risk prediction. Methods This secondary analysis included prospectively collected data from a multicenter observational study involving 34 tertiary hospitals in China. Adult patients hospitalized for 7 to 30 days were enrolled. Patients without malnutrition at admission were evaluated for HAM at discharge according to the Global Leadership Initiative on Malnutrition (GLIM) criteria. Machine learning models were developed using clinical and nutritional variables available at admission, and model performance was evaluated using discrimination, calibration, and clinical utility analyses. Results Among 4746 hospitalized patients, 1364 (28.7%) had malnutrition at admission. Of the 3382 patients without baseline malnutrition, 731 (21.6%) developed HAM during hospitalization. Patients with HAM had significantly higher rates of hospital-acquired complications (9.0% versus 2.9%), longer hospital stays (14.93 versus 13.53 days), and higher healthcare costs (¥48 980 versus ¥20 640) compared with those without HAM (all P < 0.001). Male sex, older age, pre-admission weight loss, reduced calf circumference, decreased handgrip strength, reduced muscle mass phenotype according to GLIM, and malignant tumors were associated with HAM. The XGBoost model demonstrated the best predictive performance, achieving an area under the receiver operating characteristic curve of 0.882 (95% CI: 0.857–0.906) in the independent test set. Conclusions HAM is common among hospitalized patients without malnutrition at admission, and the developed machine learning model may enable early identification of high-risk patients for timely nutritional intervention.
Background Low skeletal muscle (SM) is common at ICU admission and is associated with worse prognosis. Early identification of at-risk patients is essential, but determinants of low SM in critically ill patients remains unclear. Objective To identify sociodemographic, nutritional, and clinical characteristics associated with low SM at ICU admission. Methods This cross-sectional analysis included critically ill patients ≥18 years admitted within 24 hours. Sociodemographic and nutritional data were obtained through anamnesis, while clinical data were extracted from records. Quadriceps muscle layer thickness (QMLT) was measured using a B-mode US. Lower QMLT was defined as values below the sex-specific 25th percentile of the sample. Binary logistic regression was performed. Results 399 patients were included (age 61.4 ± 14.9 years; 56.6% male). Older age (OR = 1.02; CI 95% 1.0–1.04), previous unintentional weight loss (OR = 1.64; CI 95% 1.03–2.61), reduced food intake before admission (OR = 2.32; CI 95% 1.44–3.74), and pressure injuries (OR = 1.83; CI 95% 1.05–3.20) were associated with low QMLT. However, in multivariate analysis, only older age (OR = 1.02; CI 95% 1.00–1.04) and reduced food intake (OR = 2.26; CI 95% 1.22–4.18) remained independent determinants. Conclusion Older age and reduced food intake were independently associated with lower QMLT at ICU admission. These findings highlight the need for early screening and targeted strategies to mitigate muscle loss and improve outcomes in critically ill patients.
Objectives The Global Leadership Initiative on Malnutrition (GLIM) was established to standardize malnutrition diagnosis. Despite growing research, few studies have investigated its application in outpatients. This study assessed the prognostic value of GLIM combinations in colorectal and/or lung cancer outpatients. Methods Ninety-two outpatients with colorectal and/or lung cancer at different treatment stages were assessed between December 2022 and July 2024. Nutritional status was classified using eight GLIM combinations. Phenotypic criteria (weight loss, reduced muscle mass, reduced muscle strength) remained constant across all models, while etiologic criteria (reduced intake, inflammation, cancer-related burden) varied. Classifications were compared with the Subjective Global Assessment. The prognostic value was evaluated for 12-mo mortality. Logistic regression, Kaplan–Meier analysis, and Cox regression were adjusted for age, sex, cancer stage, tumor site, and Eastern Cooperative Oncology Group status. Results The mean patient age was 64 y (interquartile range: 57–74 y). Malnutrition prevalence ranged from 30.4% to 89.1% depending on the GLIM combinations. When reduced intake was combined with any phenotypic criteria, 12-mo mortality risk increased 5.6–19.8-fold. The combination including all phenotypic and etiologic variables independently predicted shorter survival. Conclusions Malnutrition prevalence and prognostic performance varied according to the GLIM combination, influencing predictive accuracy for mortality. As evidence grows regarding which models achieve the greatest diagnostic accuracy in oncology, combinations including phenotypic criteria and reduced intake should be prioritized to optimize prognostic evaluation and clinical decision-making.
