Risk assessment of dysphagia related to elderly cognitive impairment based on serum biomarkers
Objective To explore the potential of serum biomarkers including GDF15 in risk assessment of dysphagia in elderly patients with cognitive impairment.Methods A retrospective analysis was conducted on data from 83 elderly patients with cognitive impairment admitted to Huai'an First People's Hospital between August 2023 and March 2024.They were divid-ed into two groups based on the presence or absence of dysphagia:the cognitive impairment group(group A)and the cognitive impairment complicated with dysphagia group(group B).The differences in general characteristics,blood biochemical indica-tors and serum biomarker level were compared between the two groups.The Kubota Drinking Test was used to evaluate swallo-wing function.Serum GDF15 level were detected by ELISA.ROC curve was used to analyze the predictive ability of the bio-markers(GDF15,IL6,PAB and ALB)for dysphagia.Results Compared with the group A,the group B was significantly older,had more severe tooth defects,and had a significantly higher rate of hyperlipidemia,suggesting that these factors may in-crease the risk of dysphagia in patients with cognitive impairment.Serum biomarker analysis showed that the levels of GDF15,IL6 and CRP in the group B were significantly higher,while the levels of PAB and ALB were lower,indicating that inflamma-tory response and malnutrition might play an important role in the occurrence of dysphagia.ROC curve analysis further con-firmed the predictive effectiveness of GDF15(AUC=0.808)and PAB(AUC=0.828)for dysphagia.The constructed logis-tic regression model based on GDF15,IL6,PAB,and ALB had an AUC value of 0.909,demonstrating high predictive per-formance and suggesting that the combined use of multiple biomarkers could provide a more accurate risk assessment for dys-phagia.Conclusion Serum GDF15 and PAB may be effective biomarkers for predicting dysphagia in elderly patients with cog-nitive impairment.The combined prediction model based on the above four biomarkers has a high predictive accuracy.