Risk Factors of Feeding Intolerance During Enteral Nutrition in Patients with Craniocerebral Injury and Predictive Value of Nomogram

YAO Wei; LI Runfeng; ZHOU Shirui

doi:10.3969/j.issn.1004-437X.2026.15.020

Abstract：Objective To explore the risk factors of feeding intolerance(FI)during enteral nutrition in patients with craniocerebral injury,and to construct a nomogram model to analyze its predictive value.Methods A total of 78 patients with craniocerebral injury who were admitted to the Second People's Hospital of Pingdingshan City from May 2022 to May 2024 were selected as research objects.According to whether FI occurred during enteral nutrition,the patients were divided into FI group(36 cases)and non-FI group(42 cases).Clinical data of the two groups were collected.The difference of clinical data between FI group and non-FI group was analyzed by single factor analysis.Logistic regression analysis was performed to analyze the influencing factors of FI during enteral nutrition in patients with craniocerebral injury.A nomogram prediction model of FI occurrence during enteral nutrition in patients with craniocerebral injury was constructed.The receiver operating characteristic(ROC)curve analysis was used to evaluate the predictive value of risk factors for the occurrence of FI in patients with traumatic brain injury during enteral nutrition.Results Logistic regression model analysis showed that Glasgow coma score(GCS),antibacterial drug use,mechanical ventilation treatment,blood glucose,albumin,ISS score and sedative use were independent risk factors for FI during enteral nutrition in patients with head injury(P＜0.05).Based on GCS score,antibacterial drug use,mechanical ventilation treatment,blood glucose,albumin,ISS score and sedative use,the risk prediction of FI during enteral nutrition in patients with brain injury was established.The score was 300,and the prediction probability was 0.915.The results of ROC analysis showed that the area under the curve was 0.673(0.578-0.768)for GCS,0.775(0.692-0.821)for antibiotics,0.858(0.811-0.905)for mechanical ventilation,0.850(0.809-0.897)for blood glucose and 0.832(0.802-0.933)for albumin,ISS score 0.743(0.663-0.823),sedative use 0.808(0.751-0.879).Verified by Bootstrap self-sampling,the calibration curve had a high degree of conformity with the 45° ideal line,and the model calibration was good.The Hosmer-Lemeshow test x2=7.23(P=0.408),indicating excellent model fitting.When the decision curve analysis showed a threshold probability of 0.15 to 0.85,the net benefit of the model was higher than that of"all treatment/no treatment",which had clinical decision-making value.Conclusion GCS score,antimicrobial drug use,mechanical ventilation treatment,blood glucose,albumin,ISS score and sedate use are independent risk factors for FI occurrence during enteral nutrition in patients with craniocerebral injury.The prediction probability of risk prediction riomogram with a score of 300 based on risk factors is 0.915.ROC analysis shows that each risk factor had certain predictive value for FI.