OBJECTIVE: As an age-related neurodegenerative disease, the prevalence of mild cognitive impairment (MCI) increases with age. Within the framework of traditional Chinese medicine, spleen-kidney deficiency syndrome (SKDS) is recognized as the most frequent MCI subtype. Due to the covert and gradual onset of MCI, in community settings it poses a significant challenge for patients and their families to discern between typical aging and pathological changes. There exists an urgent need to devise a preliminary diagnostic tool designed for community-residing older adults with MCI attributed to SKDS (MCI-SKDS). METHODS: This investigation enrolled 312 elderly individuals diagnosed with MCI, who were randomly distributed into training and test datasets at a 3:1 ratio. Five machine learning methods, including logistic regression (LR), decision tree (DT), naive Bayes (NB), support vector machine (SVM), and gradient boosting (GB), were used to build a diagnostic prediction model for MCI-SKDS. Accuracy, sensitivity, specificity, precision, F1 score, and area under the curve were used to evaluate model performance. Furthermore, the clinical applicability of the model was evaluated through decision curve analysis (DCA). RESULTS: The accuracy, precision, specificity and F1 score of the DT model performed best in the training set (test set), with scores of 0.904 (0.845), 0.875 (0.795), 0.973 (0.875) and 0.973 (0.875). The sensitivity of the training set (test set) of the SVM model performed best among the five models with a score of 0.865 (0.821). The area under the curve of all five models was greater than 0.9 for the training dataset and greater than 0.8 for the test dataset. The DCA of all models showed good clinical application value. The study identified ten indicators that were significant predictors of MCI-SKDS. CONCLUSION The risk prediction index derived from machine learning for the MCI-SKDS prediction model is simple and practical; the model demonstrates good predictive value and clinical applicability, and the DT model had the best performance. Please cite this article as: Ai YT, Zhou S, Wang M, Zheng TY, Hu H, Wang YC, Li YC, Wang XT, Zhou PJ. Development of a machine learning-based risk prediction model for mild cognitive impairment with spleen-kidney deficiency syndrome in the elderly. J Integr Med. 2025; 23(4): 390-397.
Humans ; Cognitive Dysfunction/diagnosis* ; Aged ; Male ; Female ; Machine Learning ; Spleen ; Aged, 80 and over ; Kidney ; Medicine, Chinese Traditional

Objective To explore the efficacy of ransapical-transcatheter aortic valve replacement(TA-TAVR)in patients with severe aortic regurgitation(AR)with different ejection fraction.Methods The clinical data of patients with severe AR undergoing TA-TAVR from November 2017 to January 2023 were retrospective reviewed,severe AR patients with left ventricular ejection fraction(LVEF)≥50％were included in the ejection fraction preserved group and patients with severe AR with LVEF＜50％were included in the reduced ejection fraction group.Clinical data of the patients in group 2 were analyzed.Results A total of 60 patients were included,including 24 in ejection fraction preserved group and 36 in reduced ejection fraction group.All group had significantly improved aortic regurgitation volume 1 day after surgery compared with preoperative volume.The left ventricular remodeling(LVR)indexes[left ventricular end-diastolic diameter(LVEDd),left ventricular end-systolic diameter(LVESD),left ventricular posterior wall thickness(LVPW),LVEF]improved significantly at 1 day after surgery,and remained relatively stable thereafter.The improvement in LVR in reduced ejection fraction group patients was only 3 months after surgery.Two groups of patients had a decrease in amino-terminal probrain natriuretic peptide(NT-proBNP)levels three months after surgery compared to before.Patients in ejection fraction preserved group remained stable thereafter,while patients in reduced ejection fraction group had an increase one year after surgery compared to before.The one-year heart failure rehospitalization rate in reduced ejection fraction group was higher than that in ejection fraction preserved group,and the mean scores of EQ-5D-3L and EQ-VAS were lower than those of patients in ejection fraction preserved group.Comparison of the incidence of atrial fibrillation,third degree atrioventricular block,and permanent pacemaker implantation rate,stroke incidence,leaflet thrombosis incidence,paravalvular leakage incidence,and survival rate,the usage of some drugs at 1 year in both group,there was no statistically significant difference(all P＞0.05).Conclusions TA-TAVR can effectively relieve aortic regurgitation in severe AR patients with different ejection fraction and promote left ventricular reverse remodeling.Intervention in patients with severe AR with preserved ejection fraction should be performed as early as possible.

OBJECTIVE: To analyze the RHD genotype of a blood donor with Del phenotype in Yunnan. METHODS: Rh serological phenotype was identified. RHD gene was detected by PCR-SSP typing, and its 10 exons were sequenced. Exon 9 was amplified for sequencing and analysis. RHD zygosity was detected. RESULTS: The Rh phenotype of this specimen was CcD^elee. Genomic DNA exhibited a 1 003 bp deletion spanning from intron 8, across exon 9 into intron 9. The deletion breakpoints occurred between two 7-bp short tandem repeat sequences. There was no variation in the sequences of the remaining exons. The Rh hybridization box test showed that there was one RHD negative allele. CONCLUSION This specimen is Del type caused by deletion of RHD exon 9.
Humans ; Blood Donors ; Rh-Hr Blood-Group System/genetics* ; China ; Phenotype ; Exons ; Genotype ; Alleles

To explore the predictive efficacy of fragmented red blood cells (FRC) and fragmented red blood cells percentage(FRC%) with regarding for the prognosis of septic patients, along with comparing with routine coagulation parameters. A prospective study was conducted. A total of 101 patients with sepsis who met the requirements admitted to the Intensive Care Department of the Third Xiangya Hospital of Central South University from June 1, 2022 to January 10, 2023 were selected as the research objects, they were divided into survival group and death group according to the 30-day prognosis. The clinical data and laboratory indexes such as FRCs, FRC% and Platelet (PLT) were compared between the survival group and the death group. Univariate logistic regression analysis was used first, then multivariate logistic regression analysis was used to analyze the factors affecting the prognosis. Receiver operator characteristic curve (ROC curve) was drawn to evaluate the predictive value of each parameter in the regression model for the prognosis of sepsis. The results showed that the levels of FRCs, FRC%, prothrombin time (PT), activated partial thromboplastin time (APTT) and D-dimer in death group were significantly higher than those in survival group, and PLT in death group were significantly lower than those in survival group (Z or t values were -3.712,-3.793,-2.119,-2.007,-2.209,all P<0.05). FRCs or FRC% and PLT could be independent predictors of 30-day mortality. The area under the ROC curve (AUC) of PLT for predicting 30-day death in sepsis patients was 0.727 (95%CI 0.629-0.811, P<0.01), when the optimal threshold was 137 ×10⁹/L, the sensitivity was 83.87% and the specificity was 57.14%. The AUC of FRCs for predicting 30-day death in sepsis patients was was 0.732 (95%CI 0.635-0.815, P<0.01), when the optimal threshold was 10.1×10⁹/L, the sensitivity was 77.42%, and the specificity was 67.14%. The AUC of FRC% for predicting 30-day death in sepsis patients was 0.737 (95%CI 0.640-0.820, P<0.01), When the optimal threshold was 0.34%, the sensitivity was 77.42%, and the specificity was 65.71%. In conclusion,PLT, FRCs and FRC% have great application value in the prognosis of sepsis. When the PLT, FRCs and FRC% of sepsis patients are more than 137 ×10⁹/L, 10.1×10⁹/L and 0.34% respectively, it is necessary to take necessary and reasonable clinical intervention measures as soon as possible.

To explore the predictive efficacy of fragmented red blood cells (FRC) and fragmented red blood cells percentage(FRC%) with regarding for the prognosis of septic patients, along with comparing with routine coagulation parameters. A prospective study was conducted. A total of 101 patients with sepsis who met the requirements admitted to the Intensive Care Department of the Third Xiangya Hospital of Central South University from June 1, 2022 to January 10, 2023 were selected as the research objects, they were divided into survival group and death group according to the 30-day prognosis. The clinical data and laboratory indexes such as FRCs, FRC% and Platelet (PLT) were compared between the survival group and the death group. Univariate logistic regression analysis was used first, then multivariate logistic regression analysis was used to analyze the factors affecting the prognosis. Receiver operator characteristic curve (ROC curve) was drawn to evaluate the predictive value of each parameter in the regression model for the prognosis of sepsis. The results showed that the levels of FRCs, FRC%, prothrombin time (PT), activated partial thromboplastin time (APTT) and D-dimer in death group were significantly higher than those in survival group, and PLT in death group were significantly lower than those in survival group (Z or t values were -3.712,-3.793,-2.119,-2.007,-2.209,all P<0.05). FRCs or FRC% and PLT could be independent predictors of 30-day mortality. The area under the ROC curve (AUC) of PLT for predicting 30-day death in sepsis patients was 0.727 (95%CI 0.629-0.811, P<0.01), when the optimal threshold was 137 ×10⁹/L, the sensitivity was 83.87% and the specificity was 57.14%. The AUC of FRCs for predicting 30-day death in sepsis patients was was 0.732 (95%CI 0.635-0.815, P<0.01), when the optimal threshold was 10.1×10⁹/L, the sensitivity was 77.42%, and the specificity was 67.14%. The AUC of FRC% for predicting 30-day death in sepsis patients was 0.737 (95%CI 0.640-0.820, P<0.01), When the optimal threshold was 0.34%, the sensitivity was 77.42%, and the specificity was 65.71%. In conclusion,PLT, FRCs and FRC% have great application value in the prognosis of sepsis. When the PLT, FRCs and FRC% of sepsis patients are more than 137 ×10⁹/L, 10.1×10⁹/L and 0.34% respectively, it is necessary to take necessary and reasonable clinical intervention measures as soon as possible.

To investigate the application of bromocresol green Colorimetry (BCG) method in measuring serum albumin (ALB) and to evaluate its influencing factors in different diseases. This study was a cross-sectional study that included 128 people admitted to the department of nephrology, department of general surgery, department of infectious diseases and other departments of the Third Xiangya Hospital of Central South University in July 2021. They were divided into groups according to disease types, including chronic kidney disease group (47 cases), liver disease group (40 cases), other diseases group (41 cases), serum ALB was detected by BCG method and immunoturbidimetry at the same time, and the results were expressed as ALB_BCG and ALB_I respectively, each group was subdivided into three subgroups according to ALB_I results: relatively high-value subgroup, relatively intermediate-value subgroup and relatively low-value subgroup of albumin. ALB_I and ALB_BCG were compared in all groups and subgroups. Passing-Bablok regression and Bland-Altman diagram analysis were used to evaluate the application of ALB_BCG in each group. Immunoturbidimetry was used as a reference method to evaluate the bias of ALB_BCG, and the differences between ALB_I and ALB_BCG were shown as follows:ΔALB= ALB_BCG-ALB_I. Pearson correlation analysis and multiple linear regression analysis were used to assess the correlation between ΔALB and ALB autoconcentration (ALB_I), α₁-globulin, α₂-globulin, β₁-globulin, β₂-globulin, γ-globulin, creatinine (Cr), urea (UN), uric acid (UA), aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin (TBil), direct bilirubin (DBil), and C-reactive protein (CRP) levels.The results showed that ALB_BCG were higher than ALB_I in the relative low subgroups of total patients group, chronic kidney disease group, liver disease group and other disease groups, and the differences were statistically significant (t value was 8.025, 6.878, 2.628, 4.915, respectively, P<0.05). In the relatively high value subgroup, ALB_BCG was lower than ALB_I, and the differences were statistically significant in the relative high value subgroup of total patients group, liver disease group and other disease groups (t value was -4.388, -2.927, -3.979, P<0.05). Passing-Bablok regression and Bland-Altman analysis showed that the BCG method had proportional bias. In the chronic kidney disease group, the concentrations of ALB_I and Cr had the greatest influence on BCG bias, and the regression model equation was ΔALB=5.437-0.146× Alb_I-0.001 ×Cr, R²=0.505. In the liver disease group, the concentrations of ALB_I, α₁-globulin, β₁-globulin had the greatest influence on BCG bias, and the regression model equation was ΔALB=3.652-0.230×ALB_I+0.398×α₁-globulin+1.171×β₁-globulin, R²=0.658. In the other disease group, the concentration of ALB_I and α₂-globulin had the greatest influence on BCG bias, and the regression equation was ΔALB=5.558-0.225×Alb_I-0.281×α₂-globulin, R²=0.646. The BCG method has a proportion error, and its bias may lead to unacceptable differences. BCG method is mainly affected by the concentration of ALB itself, and may also be affected by α₁-globulin, α ₂-globulin, β₁-globulin, Cr.
BCG Vaccine ; Bilirubin ; Bromcresol Green ; Colorimetry ; Cross-Sectional Studies ; Globulins ; Humans ; Renal Insufficiency, Chronic ; Retrospective Studies ; Serum Albumin/analysis*

Professor
Acupuncture ; Acupuncture Therapy ; Angina, Stable ; Combined Modality Therapy ; Humans ; Moxibustion

【Objective】 To investigate the clinicopathological characteristics and prognosis of T2DM patients with renal involvement. 【Methods】 We conducted a retrospective analysis of clinical and pathologic data from T2DM patients who received renal biopsy(n = 120), and these patients were followed up. Renal outcome is defined as receiving renal replacement therapy or progression to ESRD. 【Results】 Among the 120 patients with T2DM, 57(47.5%) were diagnosed as DN, and 63(52.5%) as non diabetic renal disease(NDRD). The most common subtype of NDRD is membranous nephropathy. Compared with the NDRD group, the DN group had longer course of diabetes, worse renal function and higher proportion of diabetic retinopathy. Kaplan-Meier analysis showed that the renal survival rate in DN group was significantly lower than that in NDRD group. The 1-year, 5-year renal survival rate of DN group was only 81%, 41% Vs NDRD group 95, 84%. After multivariate adjustment, the risk of ESRD in DN patients was 3.81 times higher than that of NDRD patients(95%CI 1.61-9.01, P=0.002) . 【Conclusions】 There is a risk of misdiagnosis DN by clinical manifestations, and accurate diagnosis depends on renal biopsy. The prognosis of DN patients is significantly worse than NDRD, so it is of great clinical significance to distinguish the pathological types of diabetic patients with kidney disease.

【Objective】 To explore the main causes of death and the risk factors for death in maintenance hemodialysis (MHD) patients. 【Methods】 A retrospective study was carried out in 210 MHD patients in The Third Affiliated Hospital of Sun Yat-sen University. Of the 210 patients, 68 dead were designated as the experimental group and the other 142 survivals as the control. Single factor and binary logistic regression analyses were used to investigate the effects of clinical indicators on the prognosis of patients. We further performed subgroup analysis in experimental group, examined the patients died of cardiovascular diseases and those died of non-cardiovascular diseases, and then compared their results of color Doppler echocardiography. 【Results】 The age of the patients was(56.6±16.6) years and 125 of them were male. Compared with those in the control group, the patients in the death group had elder average age, higher rates of complications such as hypertension, diabetes, cardio-cerebro-vascular diseases, lower levels of hemoglobin, albumin, creatinine and uric acid and significantly elevated level of hypersensitive C-reactive protein(P < 0.05). The top three causes of death were cardiocerebro-vascular disease(33 cases, 48.5%), infection(16 cases, 23.5%), malnutrition(8 cases, 11.8%). Logistic regression analysis showed that age, cardio-cerebro-vascular diseases, diabetes mellitus, hypersensitive C-reactive protein and serum creatinine were independent risk factors for death in MHD patients. Subgroup analysis revealed the patients died of CVD diseases had higher incidences of left ventricular hypertrophy and cardiac valve calcification than those died of non-CVD diseases.【Conclusion】Cardio-cerebro-vascular diseases, infection and malnutrition are the main causes of death in MHD patients. Patients died of CVD diseases exhibit significantly higher incidences of left ventricular hypertrophy and cardiac valve calcification compared with those died of non-CVD diseases.

Background/Aims: Esophagogastric junction adenocarci-noma (EJA) is a malignant tumor associated with high mor-bidity and has attracted increasing attention due to a rising incidence and low survival rate. Pathological biopsy is the gold standard for diagnosis, but noninvasive and effective tests are lacking, resulting in diagnoses at advanced stages.This study explored the diagnostic value of insulin-like growth factor binding protein 7 (IGFBP7) in EJA. Methods: A total of 120 EJA patients and 88 normal controls were recruited, and their serum levels of IGFBP7 were measured by enzymelinked immunosorbent assay. Receiver operating character-istic (ROC) curve analysis was used to assess the diagnostic value, and Pearson chi-square analysis was used to evaluate the correlation between IGFBP7 and clinical parameters. Ka-plan-Meier survival analysis was carried out to assess the ef-fect of IGFBP7 on overall survival (OS). Results: The levels of IGFBP7 were higher in both early- and late-stage EJA patients than in normal controls (p<0.001). The area under the ROC curve for EJA patients was 0.794 (95% confidence interval [CI], 0.733 to 0.854), with a cutoff value of 2.716 ng/mL, a sensitivity of 63.3% (95% CI, 54.0% to 71.8%) and a specific-ity of 90.9% (95% CI, 82.4% to 95.7%). For the diagnosis of early-stage EJA, the same cutoff value and specificity were obtained, but the sensitivity of IGFBP7 was 54.3% (95% CI, 36.9% to 70.8%). Patients with low IGFBP7 protein expres-sion had lower OS than those with high expression (p=0.034).The multivariate analysis showed that IGFBP7 is an inde-pendent prognostic factor for EJA (p=0.011). Conclusions Serum IGFBP7 acts as a potential diagnostic and prognostic marker for EJA.