| Objective:By analyzing the changes of body composition in patients with preoperative malnourished gastric cancer and the correlation between bioelectrical impedance analysis indicators and other nutritional indicators,the multivariate logistic regression analysis and ROC curve analysis were used to screen out the sensitive,valuable bioelectrical impedance analysis indicators for diagnosing preoperative malnutrition.Methods:From January 2015 to October 2016,clinical data of 110 patients with gastric cancer who underwent surgery in Shanxi Provincial People’s Hospital were collected.According to the Patient-Generated Subjective Global Assessment(PG-SGA),the clinic date of 110 cases with gastric cancer were divided into two groups named malnutrition group(score≥4)and non-malnutrition group(score≤3),the body composition of patients in two groups was detected by Biomechanical Impedance Analysis(BIA).Pearson correlation analysis was respectively used to analyze the correlation between bioelectrical impedance index and PG-SGA score and laboratory test indexes.The related influencing factors of preoperative malnutrition was analyzed by univariate and multivariate logistic regression analysis in order to identify the risk and protective factors of malnutrition.According to PG-SGA≥4 as the cut-off points of diagnosing malnutrition,the sensitive,valuable bioelectrical impedance analysis indicators for diagnosing preoperative malnutrition in patients with gastric cancer was screen out by drawing the ROC curve and calculating the area under the curve.Results:According to the PG-SGA,110 cases of gastric cancer were enrolled in this study,there were 84 patients(76.36%)in malnutrition group(score≥4)and 26 cases(23.64%)in non-malnutrition group(score≤3).The age and ECW/BCW in the non-malnourished group patients were lower than those in the malnutrition group,but BMI,BFP,FFM,PM,SMM,BCM,Alb,PA and LC were higher than those in malnutrition group patients.The Pearson correlation analysis showed that PG-SGA score was positively correlated with age(r=0.324)and ECW/BCW(r=0.433),but negatively correlated with BMI(r=-0.502),BFP(r=-0.284),FFM(r=-0.346),PM(r=-0.307),SMM(r=-0.328),BCM(r=-0.323),Alb(r=-0.328)(P <0.05).Albumin was positively correlated with BFP(r=0.285),FFM(r=0.334),PM(r=0.367),SMM(r=0.428),BCM(r=0.353),but negatively correlated with ECW/BCW(r=-0.184)(P <0.05).The results of univariate analysis showed that age,BMI,BFP,FFM,PM,SMM,BCM,ECW/BCW,Alb,PA and LC were the influencing factors of preoperative malnutrition,but malnutrition was not related to the gender of patients(P > 0.05),the results of multivariate logistic regression analysis found that elder and higher ECW/BCM were the independent risk factors for preoperative malnutrition in patients with gastric cancer,while higher level of FFM,PM,SMM,BCW,Alb and higher percentage of body fat were the protect factors for malnutrition.According to PG-SGA≥4 as the cut-off points of diagnosing malnutrition,the area under the ROC curve of BFP,FFM,PM,SMM,BCM and ECW/BCM for the diagnosis of malnutrition was 0.874(0.797-0.927),0.843(0.762-0.906),0.798(0.710-0.868),0.812(0.727-0.880),0.813(0.728-0.881)and 0.806(0.720-0.875).Conclusion:Bioelectrical impedance analysis indexes have a good correlation not only with the PG-SGA score,but also with the laboratory test indexes,BIA could be used to evaluate the preoperative nutritional status of patients with gastric cancer,and the bioelectrical impedance indexes such as BFP,FFM,PM,SMM,BCM and ECW/BCM has a high diagnostic value for malnutrition,and can be used as quantitative indexes for diagnosis of preoperative malnutrition in patients with gastric cancer. |
PDF Full Text Request