Dataset Information

Machine Learning-Based Prediction of In-Hospital Complications in Elderly Patients Using GLIM-, SGA-, and ESPEN 2015-Diagnosed Malnutrition as a Factor.

ABSTRACT:

Background

Malnutrition is prevalent in elderly inpatients and is associated with various adverse outcomes during their hospital stay, but the diagnosis of malnutrition still lacks widely applicable criteria. This study aimed to investigate the association of malnutrition diagnosed with the SGA, ESPEN 2015, and GLIM criteria, respectively, with in-hospital complications in elderly patients.

Method

Hospitalized patients over 65 years old who had been assessed with the SGA guideline for malnutrition at admission were retrospectively recruited from a large observational cohort study conducted in 34 level-A tertiary hospitals in 18 cities in China from June to September 2014. Malnutrition was then retrospectively diagnosed using the GLIM and ESPEN 2015 criteria, respectively, for comparison with the results of the SGA scale. The risk factors for malnutrition were analyzed using logistic regression, and the value of the three diagnostic criteria in predicting the in-hospital complications was subsequently explored using multivariate regression and the random forest machine learning algorithm.

Results

A total of 2526 subjects who met the inclusion and exclusion criteria of the study were selected from the 7122 patients in the dataset, with an average age of 74.63 ± 7.12 years, 59.2% male, and 94.2% married. According to the GLIM, SGA, and ESPEN 2015 criteria, the detection rates of malnutrition were 37.8% (956 subjects), 32.8% (829 subjects), and 17.0% (429 subjects), respectively. The diagnostic consistency between the GLIM and the SGA criteria is better than that between the ESPEN 2015 and the SGA criteria (Kappa statistics, 0.890 vs. 0.590). Logistic regression showed that the risk of developing complications in the GLIM-defined malnutrition patients is 2.414 times higher than that of normal patients, higher than those of the ESPEN 2015 and SGA criteria (1.786 and 1.745 times, respectively). The random forest classifications show that the GLIM criteria have a higher ability to predict complications in these elderly patients than the SGA and ESPEN 2015 criteria with a mean decrease in accuracy of 12.929, 10.251, and 5.819, respectively, and a mean decrease in Gini of 2.055, 1.817, and 1.614, respectively.

Conclusion

The prevalence of malnutrition diagnosed with the GLIM criteria is higher than that of the SGA and the ESPEN 2015 criteria. The GLIM criteria are better than the SGA and the ESPEN 2015 criteria for predicting in-hospital complications in elderly patients.

SUBMITTER: Ren SS

PROVIDER: S-EPMC9331502 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background: There are several approaches that can be used for the pre-treatment identification of malnutrition in oncology populations including the Patient-Generated Subjective Global Assessment (PG-SGA), the 2015 consensus statement by the European Society for Clinical Nutrition and Metabolism (ESPEN 2015) and the Global Leadership Initiative on Malnutrition (GLIM). Aims: This study aimed to evaluate whether malnutrition, as defined by these three methods, can be used to predict complications in esophageal cancer (EC) patients after esophagectomy. Methods: We performed a single center, observational cohort study that included 360 EC patients undergoing esophagectomy from December 2014 to November 2019 at Daping Hospital in China. The prevalence of malnutrition in the study population was prospectively defined using the PG-SGA (≥9 defined malnutrition), and retrospectively defined using the ESPEN 2015 and the GLIM. The prevalence of malnutrition and association with postoperative complications were compared in parallel for the three methods. Results: The prevalence of malnutrition before surgery was 23.1% (83/360), 12.2% (44/360), and 33.3% (120/360) in the study population, as determined by the PG-SGA, the ESPEN 2015 and the GLIM, respectively. The PG-SGA and GLIM had higher diagnostic concordance (Kappa = 0.519, P < 0.001) compared to the ESPEN 2015 vs. GLIM (Kappa = 0.361, P < 0.001) and PG-SGA vs. ESPEN 2015 (Kappa = 0.297, P < 0.001). The overall incidence of postoperative complications for the study population was 58.1% (209/360). GLIM- and ESPEN 2015-defined malnutrition were both associated with the total number of postoperative complications in multivariable analyses. Moreover, GLIM-defined malnutrition exhibited the highest power to identify the incidence of complications among all independent predictors in a pooled analysis. Conclusion: Among the PG-SGA, the ESPEN 2015 and the GLIM, the GLIM framework defines the highest prevalence rate of malnutrition and appears to be the optimal method for predicting postoperative complications in EC patients undergoing esophagectomy. These results support the importance of preoperatively identifying malnutrition using appropriate assessment tools, because it can facilitate the selection of management strategies that will optimize the clinical outcomes of EC patients.

Project description:BackgroundThe Global Leadership Initiative on Malnutrition (GLIM) criteria have been recently launched by consensus of the major nutrition societies. GLIM criteria are partly constructed on the previous definition of malnutrition developed by the European Society of Clinical Nutrition and Metabolism (ESPEN). We aimed to assess malnutrition according to the ESPEN and GLIM criteria at baseline and to determine the corresponding risk of mortality during a 4-year follow-up in community-dwelling older adults from the SarcoPhAge (Sarcopenia and Physical Impairment with advancing Age) study. The relationship between malnutrition and incidence of 4-year adverse health consequences (institutionalization, hospitalization, falls, and fractures) was assessed.MethodsThis prospective population-based cohort was part of SarcoPhAge, which included 534 older adults in Belgium, followed up from 2013 to 2019. Community-dwelling healthy volunteers ≥65 years old were recruited. Mortality and adverse health consequences were collected annually by interview or phone call. Baseline malnutrition was defined according to the GLIM and ESPEN criteria. Agreement between the two definitions was reported by Cohen's kappa coefficient. Adjusted Cox regression and Kaplan-Meier survival curves were performed for malnutrition. Logistic regression was used for the other outcomes.ResultsFrom 534 subjects in SarcoPhAge, the records for 411 participants (73.2 ± 6.05 years old; 55.7% women) had all the variables needed to apply the GLIM criteria. Prevalence of baseline malnutrition was 23.4% for GLIM and 7% for ESPEN criteria (k = 0.30, low agreement). The adjusted Cox regression showed a significant increased mortality risk according to malnutrition status as defined by the GLIM [adjusted hazard ratio = 4.41 (95% confidence interval: 2.17-8.97)] and ESPEN [adjusted hazard ratio = 2.76 (95% confidence interval: 1.16-6.58)] criteria. Survival curves differed significantly between malnourished and non-malnourished groups, regardless of the definition used (log rank P < 0.001 for both). No association was found between baseline malnutrition according to these two criteria and 4-year risk of institutionalization, hospitalization, falls, or fractures (all P > 0.05).ConclusionsMalnutrition according to the GLIM criteria was associated with a 4.4-fold higher mortality risk, double that of the ESPEN criteria, during a 4-year follow-up. No association was found between malnutrition according to these two criteria and incidence of other health adverse consequences. GLIM criteria anticipate mortality and might guide interventions, with important implications for clinical practice and research.

Project description:Elderly people and patients with colorectal cancer (CRC) are both at high risk of malnutrition. Therefore, it is of great significance to explore suitable malnutrition screening and diagnostic indicators for elderly patients with CRC. Recently, the Global Leadership Initiative on Malnutrition (GLIM) proposed new diagnostic criteria for malnutrition. The aim of this article was to evaluate the diagnostic value of GLIM criteria for malnutrition in elderly colorectal patients. We explored the relationship between GLIM-malnutrition, post-operative complications and the long-term prognosis of elderly colorectal patients. Elderly patients (aged ≥65 years) who underwent CRC surgery from January 2015 to December 2018 were included. Malnutrition was diagnosed based on the GLIM criteria. The relationships between GLIM-malnutrition and clinical characteristics were analyzed by t-tests, Mann-Whitney U tests, and chi-squared tests. The relationships between GLIM-malnutrition and post-operative complications were analyzed by chi-squared tests, and logistic regression analyses. The relationships between GLIM-malnutrition and the long-term prognosis were analyzed by Kaplan-Meier analyses and logistic and Cox regression analyses. A total of 385 elderly patients were included in this study, and 118 patients (30.65%) were diagnosed with malnutrition according to the GLIM criteria. GLIM-malnutrition was significantly associated with older age, lower body mass index (BMI), lower grip strength, tumor location, higher Nutrition Risk Screening 2002 (NRS-2002), and lower levels of albumin and hemoglobin. GLIM-malnutrition was an independent risk factor [odds ratio (OR): 1.753, 95% confidence interval (CI): 1.100-2.795, P=0.018] for post-operative complications. Cox regression analysis showed that GLIM-malnutrition was an independent risk factor for overall survival in elderly patients with CRC. The GLIM criteria are feasible diagnostic criteria for malnutrition of elderly patients with CRC. GLIM-malnutrition is significantly associated with post-operative complications and overall survival in elderly patients with CRC.

Project description:AimsWe aimed to investigate the prognostic impact of malnutrition, defined by the Global Leadership Initiative on Malnutrition (GLIM) criteria, stratified by renal function in hospitalized patients with acute decompensated heart failure (HF).Methods and resultsIn this retrospective study, 314 patients who were hospitalized for acute decompensated HF from August 2019 to October 2020 were enrolled. We evaluated malnutrition using the GLIM criteria during the time of admission. The primary outcome was 90-day all-cause mortality. The median patient age was 82 years, and 90-day mortality was 14.0%. In total, 76 (24.2%) patients were malnourished according to the GLIM criteria. Malnutrition defined by the GLIM criteria [adjusted hazard ratio (HR) 1.41, 95% confidence interval (CI) 1.02-1.91, P = 0.036] and renal insufficiency [adjusted HR 2.59, 95% CI 1.07-6.28, P = 0.035 for estimated glomerular filtration rate (eGFR) < 30 mL/min/1.73 m2 vs. ≥60 mL/min/1.73 m2 ] were identified as independent predictors of 90-day mortality after adjustment for age, systolic blood pressure, and serum sodium level. In the combined setting of both variables, patients with malnutrition and eGFR < 30 mL/min/1.73 m2 had a markedly higher risk of 90-day mortality compared with those without malnutrition and eGFR ≥ 60 mL/min/1.73 m2 (adjusted HR 3.92, 95% CI 1.10-13.9, P = 0.035). Adding both eGFR and malnutrition, defined by the GLIM criteria, to the baseline model with established risk factors improved both net reclassification and integrated discrimination greater than that of the baseline model (0.606, P < 0.001 and 0.050, P = 0.002, respectively), even when compared with the model with malnutrition by the GLIM alone (0.463, P = 0.002 and 0.034, P < 0.001, respectively).ConclusionsNutrition screening using the GLIM criteria stratified by renal function could clearly predict 90-day mortality in hospitalized patients with acute decompensated HF.

Project description:Background and aimsMalnutrition is widely present and influences the prognosis of elderly inpatients, so it is helpful to be able to identify it with a convenient method. However, in the widely accepted criteria for malnutrition, the Global Leadership Initiative on Malnutrition (GLIM), a lot of metrics can be used to define the phenotypic and etiological criteria. To identify muscle mass reduction, anthropometric parameters such as calf circumference (CC) and hand grip strength (HGS) are preferable to other expensive methods in many situations because they are easy and inexpensive to measure, but their applicability needs to be verified in specific clinical scenarios. This study aims to verify the value of CC- and HGS-identified muscle loss in diagnosing malnutrition and predicting in-hospital complications (IHC) and prolonged length of hospital stay (PLOS) in elderly inpatients using machine learning methods.MethodsA sample of 7122 elderly inpatients who were enrolled in a previous multicenter cohort study in China were screened for eligibility for the current study and were then retrospectively diagnosed for malnutrition using 33 GLIM criteria that differ in their combinations of phenotypic and etiological criteria, in which CC or CC+HGS were used to identify muscle mass reduction. The diagnostic consistency with the subjective global assessment (SGA) criteria at admission was evaluated according to Kappa coefficients. The association and the predictive value of the GLIM-defined malnutrition with 30-day IHC and PLOS were evaluated with logistic regression and randomized forest models.ResultsIn total, 2526 inpatients (average age 74.63 ± 7.12 years) were enrolled in the current study. The prevalence of malnutrition identified by the 33 criteria combinations ranged from 3.3% to 27.2%. The main IHCs was infectious complications (2.5%). The Kappa coefficients ranged from 0.130 to 0.866. Logistic regression revealed that malnutrition was identified by 31 GLIM criteria combinations that were significantly associated with 30-day IHC, and 22 were significantly associated with PLOS. Random forest prediction revealed that GLIM 15 (unconscious weight loss + muscle mass reduction, combined with disease burden/inflammation) performs best in predicting IHC; GLIM 30 (unconscious weight loss + muscle mass reduction + BMI reduction, combined with disease burden/inflammation) performs best in predicting PLOS. Importantly, CC alone performs better than CC+HGS in the criteria combinations for predicting adverse clinical outcomes.ConclusionMuscle mass reduction defined by a reduced CC performs well in the GLIM criteria combinations for diagnosing malnutrition and predicting IHC and PLOS in elderly Asian inpatients. The applicability of other anthropometric parameters in these applications needs to be further explored.

Project description:ObjectiveMalnutrition is a nutritional disorder and common syndrome that has a high incidence and is easily ignored in hospitalized older patients. It can lead to multiple poor prognoses, such as frailty. Early identification and correct evaluation of possible malnutrition and frailty are essential to improve clinical outcomes in older patients. Therefore, our objective was to explore the applicability and effectiveness of the Global Leadership Initiative on Malnutrition (GLIM) criteria for identifying malnutrition in older patients.MethodsIn total, 223 participants aged ≥60 years were involved. Nutrition was evaluated using the Mini Nutritional Assessment-Full Form (MNA-FF) and GLIM criteria, which adopt a two-step procedure. The first step was to use three different methods for the screening of nutritional risk: the Nutrition Risk Screening 2002, the Mini Nutritional Assessment Short Form (MNA-SF), and the Malnutrition Universal Screening Tool. The second step was to link a combination of at least one phenotypical criterion and one etiological criterion to diagnose malnutrition. The Clinical Frailty Scale was used to assess frailty. Sensitivity, specificity, Youden index, kappa values, and positive and negative predictive values were used to evaluate the validity of the GLIM criteria. Logistic regression models were used to assess whether there was a correlation between malnutrition, as defined by the GLIM criteria, and frailty.ResultsWe found that 32.3-49.8% of our patient sample were at risk of malnutrition based on the GLIM diagnosis and using the three different screening tools; 19.3-27.8% of the patients were malnourished. GLIM criteria with MNA-SF as a diagnostic validation and MNA-FF as a reference showed high consistency (K = 0.629; p < 0.001), sensitivity (90.5%), and specificity (86.4%). Logistic regression analysis showed that malnutrition, using MNA-SF with the GLIM criteria, was relevant for a higher likelihood of frailty (OR = 1.887; 95% CI 1.184-2.589).ConclusionsThe incidence of GLIM-defined malnutrition was 19.3-27.8% using different screening tools. The consistency between the GLIM criteria using the MNA-SF and the MNA methods was high. Malnutrition, as diagnosed by the GLIM criteria with MNA-SF, was significantly correlated with frailty. GLIM criteria with MNA-SF may be a more reliable malnutrition assessment process in older inpatients.