Objective To screen the risk factors for hypoglycemia in adult intensive care unit(ICU)patients,construct a risk prediction model,and validate its predictive effect.Methods A retrospective study was conducted on adult critically ill patients admitted to the general ICU of Hospital of Chengdu University of Traditional Chinese Medicine from December 2023 to September 2024.Patients admitted from December 2023 to June 2024 served as the modeling group,and those from July to September 2024 as the validation group.A total of 928 patients were included,with 650 in the modeling group and 278 in the validation group.After literature review and expert consultation,27 potential risk factors for hypoglycemia in ICU patients were initially screened,and data were collected including general information[gender,age,acute physiology and chronic health evaluation Ⅱ(APACHEⅡ)score,sequential organ failure assessment(SOFA)score,nutrition risk in critically ill(NUTRIC)score,mechanical ventilation status,hemodialysis status,enteral nutrition status],disease data(sepsis,liver disease history,kidney disease history,diabetes history,hypoglycemia history),blood glucose-related indicators[mean blood glucose,blood glucose coefficient of variation,insulin dosage,intravenous insulin titration use,inotropic drug use,insulin secretagogues(Sulfonylureas and Glinides),and combined use of hypoglycemic drugs(two or more)],and laboratory indicators[serum creatinine(SCr),blood urea nitrogen(BUN),serum albumin(Alb),alanine aminotransferase(ALT),aspartate aminotransferase(AST),total bilirubin(TBil),glomerular filtration rate(GFR)].The patients were divided into a hypoglycemia group and a non-hypoglycemia group based on the occurrence of hypoglycemia.Univariate analysis and binary Logistic regression analysis were used to identify influencing factors of hypoglycemia in adult ICU patients,and a nomogram prediction model was constructed.The area under the receiver operator characteristic curve(AUC)and calibration curves were employed to evaluate the discrimination and calibration of the model.Results The modeling cohort included 552 non-hypoglycemic patients and 98 hypoglycemic patients,with an ICU hypoglycemia incidence rate of 15.1％.Compared with the hypoglycemia group,the non-hypoglycemia group showed significantly lower proportions of patients with renal disease history,diabetes history,hypoglycemia history,undergoing hemodialysis,using intravenous insulin titration,and combined use of hypoglycemic drugs,as well as lower blood glucose coefficient of variation,lower APACHEⅡ scores,and significantly elevated GFR(all P<0.05).Binary Logistic regression analysis was performed using the 9 variables with statistically significant differences in univariate analysis as independent variables and hypoglycemia occurrence as the dependent variable.The results indicated that a history of diabetes,a history of hypoglycemia,APACHEⅡ score,GFR,blood glucose coefficient of variation,and combined use of hypoglycemic drugs were independent risk factors for hypoglycemia in ICU patients[odds ratios(OR)were 1.761,2.095,1.048,0.990,1.029,and 1.975,respectively,and 95％confidence intervals(95％CI)were 1.052-2.949,1.220-3.600,1.022-1.074,0.982-0.997,1.013-1.046,and 1.145-3.408,respectively.The corresponding Pvalues were 0.031,0.007,0.000,0.009,<0.001,0.014].A nomogram prediction model for hypoglycemia in ICU patients was constructed using six independent predictors selected through binary logistic regression analysis.The ROC curve AUC for the modeling group was 0.884(95％CI 0.826-0.941,P=0.250),with a maximum Youden index of 0.713,sensitivity of 92.1％,and specificity of 79.2％.The validation cohort included 38 patients with hypoglycemia and 240 patients without hypoglycemia.Compared with the hypoglycemia group,the non-hypoglycemia group showed significantly lower proportions of patients with a history of diabetes,a history of hypoglycemia,and combined use of hypoglycemic drugs,as well as lower APACHEⅡ scores and lower blood glucose coefficient of variation,with significantly increased GFR(all P<0.05).The ROC curve AUC for the validation cohort was 0.803(95％CI was 0.757-0.849,P=0.138),indicating high discriminatory ability.The predicted probability at the diagnostic cutoff point was P=0.138.The model's diagnostic threshold for predicted probability was P=0.138,while the optimal cut-off value based on the Youden index was 0.513,yielding a sensitivity of 76.5％and specificity of 74.8％,indicating predictive value for hypoglycemia in adult ICU patients.The mean absolute error(MAE)results for the modeling group and validation group were<0.05.The calibration curves of both the modeling and validation groups showed close alignment with the ideal curve,indicating excellent calibration performance of the model.Conclusion The constructed hypoglycemia risk prediction model for adult ICU patients has good predictive performance,which can quickly identify high-risk populations of hypoglycemia in ICU and provide reference for clinical preventive nursing.

Objective To screen the risk factors for hypoglycemia in adult intensive care unit(ICU)patients,construct a risk prediction model,and validate its predictive effect.Methods A retrospective study was conducted on adult critically ill patients admitted to the general ICU of Hospital of Chengdu University of Traditional Chinese Medicine from December 2023 to September 2024.Patients admitted from December 2023 to June 2024 served as the modeling group,and those from July to September 2024 as the validation group.A total of 928 patients were included,with 650 in the modeling group and 278 in the validation group.After literature review and expert consultation,27 potential risk factors for hypoglycemia in ICU patients were initially screened,and data were collected including general information[gender,age,acute physiology and chronic health evaluation Ⅱ(APACHEⅡ)score,sequential organ failure assessment(SOFA)score,nutrition risk in critically ill(NUTRIC)score,mechanical ventilation status,hemodialysis status,enteral nutrition status],disease data(sepsis,liver disease history,kidney disease history,diabetes history,hypoglycemia history),blood glucose-related indicators[mean blood glucose,blood glucose coefficient of variation,insulin dosage,intravenous insulin titration use,inotropic drug use,insulin secretagogues(Sulfonylureas and Glinides),and combined use of hypoglycemic drugs(two or more)],and laboratory indicators[serum creatinine(SCr),blood urea nitrogen(BUN),serum albumin(Alb),alanine aminotransferase(ALT),aspartate aminotransferase(AST),total bilirubin(TBil),glomerular filtration rate(GFR)].The patients were divided into a hypoglycemia group and a non-hypoglycemia group based on the occurrence of hypoglycemia.Univariate analysis and binary Logistic regression analysis were used to identify influencing factors of hypoglycemia in adult ICU patients,and a nomogram prediction model was constructed.The area under the receiver operator characteristic curve(AUC)and calibration curves were employed to evaluate the discrimination and calibration of the model.Results The modeling cohort included 552 non-hypoglycemic patients and 98 hypoglycemic patients,with an ICU hypoglycemia incidence rate of 15.1％.Compared with the hypoglycemia group,the non-hypoglycemia group showed significantly lower proportions of patients with renal disease history,diabetes history,hypoglycemia history,undergoing hemodialysis,using intravenous insulin titration,and combined use of hypoglycemic drugs,as well as lower blood glucose coefficient of variation,lower APACHEⅡ scores,and significantly elevated GFR(all P<0.05).Binary Logistic regression analysis was performed using the 9 variables with statistically significant differences in univariate analysis as independent variables and hypoglycemia occurrence as the dependent variable.The results indicated that a history of diabetes,a history of hypoglycemia,APACHEⅡ score,GFR,blood glucose coefficient of variation,and combined use of hypoglycemic drugs were independent risk factors for hypoglycemia in ICU patients[odds ratios(OR)were 1.761,2.095,1.048,0.990,1.029,and 1.975,respectively,and 95％confidence intervals(95％CI)were 1.052-2.949,1.220-3.600,1.022-1.074,0.982-0.997,1.013-1.046,and 1.145-3.408,respectively.The corresponding Pvalues were 0.031,0.007,0.000,0.009,<0.001,0.014].A nomogram prediction model for hypoglycemia in ICU patients was constructed using six independent predictors selected through binary logistic regression analysis.The ROC curve AUC for the modeling group was 0.884(95％CI 0.826-0.941,P=0.250),with a maximum Youden index of 0.713,sensitivity of 92.1％,and specificity of 79.2％.The validation cohort included 38 patients with hypoglycemia and 240 patients without hypoglycemia.Compared with the hypoglycemia group,the non-hypoglycemia group showed significantly lower proportions of patients with a history of diabetes,a history of hypoglycemia,and combined use of hypoglycemic drugs,as well as lower APACHEⅡ scores and lower blood glucose coefficient of variation,with significantly increased GFR(all P<0.05).The ROC curve AUC for the validation cohort was 0.803(95％CI was 0.757-0.849,P=0.138),indicating high discriminatory ability.The predicted probability at the diagnostic cutoff point was P=0.138.The model's diagnostic threshold for predicted probability was P=0.138,while the optimal cut-off value based on the Youden index was 0.513,yielding a sensitivity of 76.5％and specificity of 74.8％,indicating predictive value for hypoglycemia in adult ICU patients.The mean absolute error(MAE)results for the modeling group and validation group were<0.05.The calibration curves of both the modeling and validation groups showed close alignment with the ideal curve,indicating excellent calibration performance of the model.Conclusion The constructed hypoglycemia risk prediction model for adult ICU patients has good predictive performance,which can quickly identify high-risk populations of hypoglycemia in ICU and provide reference for clinical preventive nursing.

Objective The aim of this study was to investigate the differences of the cell ultrastucture of normal mouse hatched blastocysts and their dormant ones cultured in vitro after freezing-thawing, and to explore whether the dor-mant embryos have a better anti-freezing shock property than the normal hatched mouse embryos .Methods By transmis-sion electron microscopy , the ultrastructure of these two types of mouse embryos was observed and analyzed .Results By comparative analysis of their ultrastructure , the results showed that the dormant embryos before freezing are being austerity and with lower energy metabolism at a ‘ground state ’ .After freezing-thawing and culture , their cellular structure seemed to be similar to that of the normal embryos cultured in vitro before freezing.However, after freezing-thawing and culture, the number of mitochondria decreased , the nuclei were loose , and their heterochromatin also increased .Conclusions From the ultrastructural observation , compared with the normal mouse hatched embryos , the cellular state of dormant mouse em-bryos after freezing-thawing is more favorable for material storage and energy metabolism , thus, indicating that they have a better anti-freezing property than normal hatched embryos .

ObjectiveTo explore the clinical significance of heart-type fatty acid-binding protein (H-FABP) in acute myocardial infarction(AMI) patients. MethodsThe level of H-FABP was assayed within 30 min, 1 h, 2 h, 4 h,6 h and 12 h by enzyme linked immunosorbent assay (ELISA) in 46 AMI patients, and cardiac troponin Ⅰ(cTnⅠ) and creatinine kinase(CK-MB) also was assayed by routine method.The diagnostic accuracy was compared among different methods. ResultsThe diagnostic accuracy of H-FABP[95.7％ (44/46)] was significantly higher than cTnⅠ[65.2％(30/46)] and CK-MB[41.3％ (19/46)](P ＜0.05). The levels of H-FABP, cTnⅠ and CK-MB significantly increased after AMI onset 4,6,12 hrespectively. ConclusionThe diagnosticaccuracy of H-FABP is higher and can be used as a parameter for the early diagnosis of AMI.