Font Size: a A A

Construction And Verification Of Risk Prediction Model Of Enteral Nutrition Feeding Intolerance In Severe Acute Pancreatitis

Posted on:2024-03-20Degree:MasterType:Thesis
Country:ChinaCandidate:J F ZangFull Text:PDF
GTID:2544307112987719Subject:Care
Abstract/Summary:PDF Full Text Request
Objective:To analyze the risk factors of feeding intolerance(FI)in Severe Acute Pancreatitis(SAP),build a risk prediction model of feeding intolerance in SAP,and verify and evaluate the model,so as to provide early screening tools for high-risk groups with feeding intolerance in SAP and provide reference for clinical implementation of targeted intervention measures.Method:1.Using computer to search the related literature of SAP enteral nutrition FI,screening the risk factors through the discussion of the research group meeting and expert consultation,designing the clinical data questionnaire,and collecting relevant information by using the hospital electronic medical record system.2.A retrospective method was used to collect 351 patients with severe acute pancreatitis from January2015 to December 2022 in a 3A hospital in Jilin Province as the research object,and246 patients from January 2015 to December 2020 were taken as the modeling group according to the admission time of patients,accounting for 70% of the total cases,so as to construct the risk prediction model.Patients were divided into tolerance group(n=143)and intolerance group(n=103)according to whether they suffered from enteral nutrition intolerance.Through univariate analysis,the risk factors of enteral nutrition intolerance in severe acute pancreatitis were preliminarily screened out,and the statistically significant variables(P < 0.05)were included in multivariate Logisttic regression analysis to construct a risk prediction model.The model was evaluated by ROC and Hosmer-Lemeshow goodness-of-fit test.105 patients from January,2021 to December,2022 were taken as the verification group,accounting for 30% of the total cases.The nomogram model and calibration curve were used to verify the model.Results:1.Among the 351 patients,139 cases developed enteral nutrition feeding intolerance,the incidence rate was 39.6%;Among them,there were 246 cases in the modeling group,and 103 cases developed feeding intolerance,the incidence rate was41.87%;Among 105 cases in the verification group,36 cases developed feeding intolerance,with an incidence rate of 34.29%.2.Binary Logistic regression analysis showed that hypertriglyceridemia(OR=1.962,95% CI: 1.088 ~ 3.538),hypoproteinemia(OR=1.920,95% CI: 1.049 ~3.516),intra-abdominal pressure(OR=2.826,95%).95% CI: 1.720 ~ 5.653)and the time to start enteral nutrition ≥72h(OR=2.350,95% CI: 1.179 ~ 4.683)are independent risk factors for the occurrence of enteral nutrition FI in SAP patients.Adding microecological agents(OR=0.379,95% CI: 0.209)The model formula is:Logit(P)=-1.206+0.674× hypertriglyceridemia +0.652× hypoproteinemia +1.039×intra-abdominal pressure ≥12 mm Hg×+1.137×APACHEⅡ score ≥20 points +0.855×the time to start enteral nutrition ≥ 72h-0.971.3.The discrimination and consistency of the established model are tested.The AUC under ROC curve is 0.793(95% CI: 0.735 ~ 0.851,P < 0.001),the maximum value of Jordan index is 0.498,the optimal critical value is 0.500,the sensitivity is81.8%,and the specificity is 68%.Hosmer-Lemeshow(H-L)test showed that χ~2 value was 3.481,p = 0.901(P > 0.05).4.The discrimination and consistency of the model are tested with the verification group data.In the validation group,the area under ROC curve(AUC)was0.888(95% CI: 0.822 ~ 0.954,P < 0.001),and the χ~2 value of H-L test was 5.985,P =0.649(P > 0.05).The sensitivity,specificity and total accuracy of the model in the validation queue were 75.8%,90% and 81.9%,respectively.Conclusion:In this study,hypertriglyceridemia,hypoproteinemia,APACHEⅡ score,intraabdominal pressure and the time of starting enteral nutrition are the independent risk factors of enteral nutrition intolerance in severe acute pancreatitis,and adding microecological agents is the protective factor of enteral nutrition intolerance in severe acute pancreatitis.A risk prediction model suitable for enteral nutrition therapy for SAP patients was established.It was verified that the model had good regional classification and consistency,and the prediction efficiency was good,which had certain clinical application value.
Keywords/Search Tags:Severe acute pancreatitis, Enteral nutrition, Feeding intolerance, Prediction model
PDF Full Text Request
Related items