Metabolomics covers a wide range of applications in life sciences, biomedicine, and phytology. Data acquisition (to achieve high coverage and efficiency) and analysis (to pursue good classification) are two key segments involved in metabolomics workflows. Various chemometric approaches utilizing either pattern recognition or machine learning have been employed to separate different groups. However, insufficient feature extraction, inappropriate feature selection, overfitting, or underfitting lead to an insufficient capacity to discriminate plants that are often easily confused. Using two ginseng varieties, namely Panax japonicus (PJ) and Panax japonicus var. major (PJvm), containing the similar ginsenosides, we integrated pseudo-targeted metabolomics and deep neural network (DNN) modeling to achieve accurate species differentiation. A pseudo-targeted metabolomics approach was optimized through data acquisition mode, ion pairs generation, comparison between multiple reaction monitoring (MRM) and scheduled MRM (sMRM), and chromatographic elution gradient. In total, 1980 ion pairs were monitored within 23 min, allowing for the most comprehensive ginseng metabolome analysis. The established DNN model demonstrated excellent classification performance (in terms of accuracy, precision, recall, F1 score, area under the curve, and receiver operating characteristic (ROC)) using the entire metabolome data and feature-selection dataset, exhibiting superior advantages over random forest (RF), support vector machine (SVM), extreme gradient boosting (XGBoost), and multilayer perceptron (MLP). Moreover, DNNs were advantageous for automated feature learning, nonlinear modeling, adaptability, and generalization. This study confirmed practicality of the established strategy for efficient metabolomics data analysis and reliable classification performance even when using small-volume samples. This established approach holds promise for plant metabolomics and is not limited to ginseng.

Objective To investigate the differences in subchondral bone mineral density(BMD)between the femoral and tibial sides in patients of knee osteoarthritis(KOA)with normal lines of force and varus.Methods The data of 450 knee joints with a definite diagnosis of KOA were included in this study including weight-bearing full-length X-ray films and quantitative computed tomography(QCT)scans of both lower limbs.Among them,131 were in the normal force line group and 319 were in the knee varus group.The hip-knee-ankle(HKA)angle and BMD of the femoral medial condyle,femoral lateral condyle,tibial medial plateau and tibial lateral plateau were measured.BMD ratio of tibial medial plateau to tibial lateral plateau and the BMD ratio of femoral medial condyle to femoral lateral condyle were calculated.BMD in medial and lateral compartments of the femur and tibia were compared between the two groups,followed by subgroup analyses based on gender and age.Spearman correlation was used to analyze the correlation between the BMD ratio of tibial medial plateau to tibial lateral plateau,the BMD ratio of femoral medial condyle to femoral lateral condyle and the degree of varus in the knee varus group.Results The BMD of the medial femoral condyles and medial tibial platforms were higher in the knee varus group than those in the normal force line group.The BMD of femoral lateral condyle and lateral tibial platform was lower in the knee varus group than that in the normal force line group.The BMD ratio of the medial to lateral tibial plateaus was greater than one in both groups,and the ratio of the knee varus group was greater.The BMD ratio of femoral medial to lateral condyle in the knee varus group was significantly higher than that in the normal force line group.For women,these findings were more pronounced and were independent of age.Correlation analysis showed that the BMD ratio of medial tibial plateau to lateral tibial plateau was negatively correlated with HKA angle(rs=-0.436,P＜0.01),and the BMD ratio of the medial femoral condyle to lateral femoral condyle was also negatively correlated with HKA angle(rs=-0.394,P＜0.01).Conclusion The BMD of medial femoral and tibial compartment is increased and the BMD of lateral compartment is decreased in the genu varus group compared with the normal force line group.

Objective:To investigate the effectiveness of new cryoprotective agents in preserving and transplanting human adipose tissue.Methods:The adipose tissue samples were obtained from healthy adult females who underwent liposuction at the Department of Plastic Surgery of Peking University Third Hospital from January to March 2022. The adipose tissue samples were centrifuged and then randomly divided into 9 groups. These groups were cryopreserved in liquid nitrogen using different cryoprotective agents [group A, group B, and dimethyl sulfoxide (DMSO) group] and cryopreservation times (1-month, 2-month, and 3-month groups), respectively. The cryoprotective agent formulation in group A was dextrose glycoside 40 (DEX), amino acids, vitamins, and inorganic salts. In group B, the formulation included DMSO and DEX. The ratio of cryoprotective agent in the DMSO group was 10% DMSO, 20% fetal bovine serum (FBS), and 70% DMEM-12. For cryopreservation, 5 ml cryogenic tubes were used with a fat to cryoprotective agent ratio of 3∶2, and each group contains 6 tubes for cryopreservation. After thawing the adipose tissue, HE staining was used to observe the histological morphology. Immunohistochemical staining was employed for the quantitative analysis of lipid droplet-encapsulated protein (Perilipin), and the Perilipin positivity rate was calculated by the ratio of the number of positive cells to the total number of cells. Adipocyte viability was assessed using the CCK-8 method. Thirty-eight healthy, clean nude mice were selected and divided into 3 groups of 12 mice each according to the use of different cryoprotective agents (groups A, B, and DMSO), while the other 2 mice were used as the day 0 control group. The mean fat freezing duration for all groups was 3 months. After nude mice were anesthetized intraperitoneally, 0.9 ml of thawed cryopreserved fat was injected into the dorsum bilaterally. The rate of adipose tissue retention was calculated by MRI scanning and three-dimensional software at 1, 2, and 3 months after transplantation, and compared between the groups. The fat grafts were explanted from the mice after they were sacrificed, and then subjected to histological morphology and quantitative analysis of Perilipin by using HE staining and immunohistochemical staining. GraphPad Prism 8.0 software was used for statistical analysis of the data. The data that conformed to a normal distribution were expressed as Mean ± SD. The overall comparison between multiple groups used analysis of variance for repeated measures. The comparison of data between groups at the same time point used Tukey’s multiple comparison test.Results:The morphology of adipose tissue in different cryoprotective agent groups closely resembled that of normal fresh adipose tissue after being cryopreserved in liquid nitrogen for 1-3 months. The difference in the proportion of Perilipin-stained positive cells in each group was not statistically significant ( P>0.05). The CCK-8 method indicated that the effect of the DMSO group was superior to groups A and B at 1 and 3 months of cryopreservation ( P<0.01), and that the DMSO group and group B were superior to group A at 2 months of cryopreservation ( P<0.01). In the animal experiments, there was no statistically significant difference between the groups in the volume retention rate 1-3 months after cryopreserved fat transplantation ( P>0.05). Additionally, the adipose tissues in each group exhibited varying degrees of localized necrosis accompanied by an inflammatory reaction 1-3 months after transplantation. There was no statistically significant difference in the Perilipin staining positivity between the groups ( P>0.05). Conclusion:The use of new cryoprotective agents for cryopreserving adipose tissue does not show a significant difference compared to the traditional cryoprotective agent. However, it is theoretically safer as it avoids the potential toxic effects of using DMSO or FBS on the human body.

Objective:To investigate the effectiveness of new cryoprotective agents in preserving and transplanting human adipose tissue.Methods:The adipose tissue samples were obtained from healthy adult females who underwent liposuction at the Department of Plastic Surgery of Peking University Third Hospital from January to March 2022. The adipose tissue samples were centrifuged and then randomly divided into 9 groups. These groups were cryopreserved in liquid nitrogen using different cryoprotective agents [group A, group B, and dimethyl sulfoxide (DMSO) group] and cryopreservation times (1-month, 2-month, and 3-month groups), respectively. The cryoprotective agent formulation in group A was dextrose glycoside 40 (DEX), amino acids, vitamins, and inorganic salts. In group B, the formulation included DMSO and DEX. The ratio of cryoprotective agent in the DMSO group was 10% DMSO, 20% fetal bovine serum (FBS), and 70% DMEM-12. For cryopreservation, 5 ml cryogenic tubes were used with a fat to cryoprotective agent ratio of 3∶2, and each group contains 6 tubes for cryopreservation. After thawing the adipose tissue, HE staining was used to observe the histological morphology. Immunohistochemical staining was employed for the quantitative analysis of lipid droplet-encapsulated protein (Perilipin), and the Perilipin positivity rate was calculated by the ratio of the number of positive cells to the total number of cells. Adipocyte viability was assessed using the CCK-8 method. Thirty-eight healthy, clean nude mice were selected and divided into 3 groups of 12 mice each according to the use of different cryoprotective agents (groups A, B, and DMSO), while the other 2 mice were used as the day 0 control group. The mean fat freezing duration for all groups was 3 months. After nude mice were anesthetized intraperitoneally, 0.9 ml of thawed cryopreserved fat was injected into the dorsum bilaterally. The rate of adipose tissue retention was calculated by MRI scanning and three-dimensional software at 1, 2, and 3 months after transplantation, and compared between the groups. The fat grafts were explanted from the mice after they were sacrificed, and then subjected to histological morphology and quantitative analysis of Perilipin by using HE staining and immunohistochemical staining. GraphPad Prism 8.0 software was used for statistical analysis of the data. The data that conformed to a normal distribution were expressed as Mean ± SD. The overall comparison between multiple groups used analysis of variance for repeated measures. The comparison of data between groups at the same time point used Tukey’s multiple comparison test.Results:The morphology of adipose tissue in different cryoprotective agent groups closely resembled that of normal fresh adipose tissue after being cryopreserved in liquid nitrogen for 1-3 months. The difference in the proportion of Perilipin-stained positive cells in each group was not statistically significant ( P>0.05). The CCK-8 method indicated that the effect of the DMSO group was superior to groups A and B at 1 and 3 months of cryopreservation ( P<0.01), and that the DMSO group and group B were superior to group A at 2 months of cryopreservation ( P<0.01). In the animal experiments, there was no statistically significant difference between the groups in the volume retention rate 1-3 months after cryopreserved fat transplantation ( P>0.05). Additionally, the adipose tissues in each group exhibited varying degrees of localized necrosis accompanied by an inflammatory reaction 1-3 months after transplantation. There was no statistically significant difference in the Perilipin staining positivity between the groups ( P>0.05). Conclusion:The use of new cryoprotective agents for cryopreserving adipose tissue does not show a significant difference compared to the traditional cryoprotective agent. However, it is theoretically safer as it avoids the potential toxic effects of using DMSO or FBS on the human body.

OBJECTIVE: To observe the expression level of cytokines in patients with sepsis and its effect on prognosis. METHODS: The clinical data of sepsis patients admitted to the intensive care unit (ICU) of the First Affiliated Hospital of Zhengzhou University from January 2020 to December 2022 were analyzed retrospectively, including gender, age, and acute physiology and chronic health evaluation II (APACHE II), blood routine, procalcitonin (PCT), C-reactive protein (CRP), and cytokines levels [interleukins (IL-2, IL-4, IL-6, IL-10, IL-17), tumor necrosis factor-α (TNF-α), and interferon-γ (IFN-γ)] within 24 hours of admission to ICU. The 28-day prognosis of the patients was followed up. The patients were divided into survival group and death group according to the prognosis. The clinical data between the two groups of sepsis patients with different prognosis were compared. Binary Logistic regression analysis was used to analyze the independent risk factors affecting the prognosis of patients with sepsis, and the receiver operator characteristic curve (ROC curve) was drawn to evaluate the predictive value of each risk factor for the prognosis of patients with sepsis. RESULTS: (1) A total of 227 patients with sepsis were enrolled, including 168 patients in the survival group (survival rate 74.0%) and 59 patients in the death group (mortality 26.0%). There were no significant differences in age (years old: 55.97±2.13 vs. 54.67±1.11) and gender (male: 71.2% vs. 57.1%) between the death group and the survival group (both P > 0.05), indicating that the baseline data of the two groups were comparable. (2) The APACHE II (19.37±0.99 vs. 14.88±0.61, P < 0.001) and PCT (μg/L: 12.39±2.94 vs. 4.14±0.90, P < 0.001) in the death group were significantly higher than those in the survival group, while the platelet count [PLT (×10⁹/L): 144.75±12.50 vs. 215.99±11.26, P = 0.001] and thrombocytocrit [(0.14±0.01)% vs. (0.19±0.01)%, P = 0.001] were significantly lower than those in the survival group. (3) The level of IL-6 in the death group was significantly higher than that in the survival group (ng/L: 577.66±143.16 vs. 99.74±33.84, P < 0.001). There were no statistically significant differences in other cytokines, IL-2, IL-4, IL-10, TNF-α, IFN-γ and IL-17 between the death group and the survival group [IL-2 (ng/L): 2.44±0.38 vs. 2.63±0.27, P = 0.708; IL-4 (ng/L): 3.26±0.67 vs. 3.18±0.34, P = 0.913; IL-10 (ng/L): 33.22±5.13 vs. 39.43±2.85, P = 0.262; TNF-α (ng/L): 59.33±19.21 vs. 48.79±29.87, P = 0.839; IFN-γ (ng/L): 6.69±5.18 vs. 1.81±0.16, P = 0.100; IL-17 (ng/L): 2.05±0.29 vs. 2.58±0.33, P = 0.369]. (4) Binary Logistic regression analysis showed that APACHE II and IL-6 were independent risk factors affecting the prognosis of patients with sepsis [odds ratio (OR) and 95% confidence interval (95%CI) were 1.050 (1.008-1.093) and 1.001 (1.000-1.002), P values were 0.019 and 0.026, respectively]. (5) ROC curve analysis showed that APACHE II and IL-6 had certain predictive value for the prognosis of patients with sepsis, the area under the ROC curve (AUC) was 0.754 (95%CI was 0.681-0.827) and 0.592 (95%CI was 0.511-0.673), P values were < 0.001 and 0.035, respectively. When the optimal cut-off value of APACHE II was 16.50 score, the sensitivity was 72.6% and the specificity was 69.9%. When the optimal cut-off value of IL-6 was 27.87 ng/L, the sensitivity was 67.2% and the specificity was 52.8%. CONCLUSIONS APACHE II score and IL-6 level have certain predictive value for the prognosis of patients with sepsis, the higher APACHE II score and IL-6 level, the greater the probability of death in patients with sepsis.
Humans ; Male ; Interleukin-10 ; Interleukin-17 ; Cytokines ; Tumor Necrosis Factor-alpha ; Interleukin-6 ; Retrospective Studies ; Interleukin-2 ; Interleukin-4 ; ROC Curve ; Sepsis/diagnosis* ; Prognosis ; Procalcitonin ; Interferon-gamma ; Intensive Care Units

Objective:To investigate the alterations in soft tissue morphology and thickness in the mid-face region of patients with cleft lip and palate (UCLP) secondary to maxillofacial deformity following Le Fort I osteotomy.Methods:A total of 22 patients (16 males and 6 females aged from 17 to 28 years with an average of 20 years) diagnosed with cleft lip and palate secondary to maxillofacial deformity were collected from the Wuhan University Hospital of Stomatology from July 2012 to August 2020. All patients underwent Le Fort I osteotomy. CBCT scans were obtained at T0 (3 days before surgery), T1 (7 days after surgery), and T2 (1 year after surgery). The Dolphin11.95 software and 3D Slicer software were utilized to measure and analyze the soft tissue near the mid-face osteotomy line. Differences in soft tissue thickness before and after surgery were compared.Results:Before and after the operation, the soft tissue thickness at P3, P5, P6, and P8 on the affected side was thicker than that on the healthy side, and the difference was statistically significant, with a P-value of <0.05. At P5, P6, P7, P8, and P9 below the osteotomy line at T2-T0, the degree of postoperative thinning on the affected side was more apparent than that on the healthy side, and there was statistical significance at P6 ( P<0.05). The postoperative soft tissue asymmetry in the Ck region was improved compared with the preoperative one. The preoperative average protruding of the affected side was 0.63 compared with the healthy side, and the postoperative value was 0.17. The preoperative and postoperative Mann-Whitney U tests showed significantly statistical difference. Conclusions:After Le Fort I osteotomy, the facial asymmetry of patients with unilateral cleft lip and palate secondary to maxillofacial deformity is improved. However, there is still a difference in the soft tissue thickness between the healthy side and the affected side, and the change in soft tissue thickness on the affected side is more significant than that on the healthy side.

Objective:To investigate the predictive value of the maximum aggregation rate (MAR) of platelet for septic shock and septic shock with disseminated intravascular coagulation (DIC).Methods:A retrospective case-control study enrolled patients with sepsis admitted to department of critical care medicine of the First Affiliated Hospital of Zhengzhou University from January 2021 to November 2022. The basic data, dynamic platelet aggregation rate, blood routine, inflammation indicators, sequential organ failure assessment (SOFA) and other clinical indicators within 24 hours after admission were collected. Septic patients were divided into the shock group and the non-shock group according to the presence of septic shock; then refer to the International Society on Thrombosis and Hemostasis (ISTH) standard, patients with septic shock were divided into the shock DIC group and the shock non-DIC group according to the presence of dominant DIC. Compared the differences in platelet aggregation function between these groups, and the receiver operator characteristic curve (ROC curve) was drawn to evaluate the predictive value of the MAR for septic shock and septic shock with DIC. Spearman correlation analysis was used to analyze the correlation of MAR with inflammation indicators and the severity of illness in patients with sepsis.Results:A total of 153 sepsis patients were included and 61 with septic shock (including 17 with dominant DIC and 44 without dominant DIC). Compared with the non-shock group, the level of procalcitonin (PCT), C-reactive protein (CRP), and SOFA score were significantly higher in the shock group [PCT (mg/L): 6.90 (2.50, 23.50) vs. 0.87 (0.26, 5.75), CRP (mg/L): 156.48 (67.11, 230.84) vs. 90.39 (46.43, 182.76), SOFA score: 11.00 (8.00, 14.00) vs. 5.00 (3.00, 8.00), all P < 0.05]. The platelet count (PLT) and the MAR induced by adenosine diphosphate (ADP), adrenaline (A), collagen (COL), and arachidonic acid (AA; ADP-MAR, A-MAR, COL-MAR, AA-MAR) in the shock group were significantly decreased [PLT (×10 9/L): 101.00 (49.00, 163.50) vs. 175.50 (108.25, 254.50), ADP-MAR: 28.50% (22.00%, 38.05%) vs. 45.90% (33.98%, 60.28%), A-MAR: 38.90% (30.00%, 55.40%) vs. 65.15% (54.38%, 72.53%), COL-MAR: 27.90% (20.85%, 36.55%) vs. 42.95% (33.73%, 54.08%), AA-MAR: 24.70% (16.40%, 34.20%) vs. 46.55% (28.33%, 59.20%), all P < 0.05]. Subgroup analysis revealed that, compared with the shock non-DIC group, the SOFA scores were significantly higher in patients in the shock DIC group (13.29±5.23 vs. 10.39±3.58, P < 0.05), the PLT and COL-MAR in the shock DIC group were significantly reduced [PLT (×10 9/L): 36.00 (22.00, 67.50) vs. 115.50 (84.25, 203.75), COL-MAR: 21.50% (17.85%, 32.60%) vs. 30.95% (22.98%, 38.53%), all P < 0.05]. ROC curve analysis showed that A-MAR had a higher predictive value for septic shock, and the area under the ROC curve (AUC) was 0.814 [95% confidence interval (95% CI) was 0.742-0.886, P = 0.000]. When the optimal cut-off value was 51.35%, the sensitivity was 68.9%, the specificity was 82.6%, the positive predictive value was 0.724 and the negative predictive value was 0.800. COL-MAR had some predictive value for septic shock with DIC, and the AUC was 0.668 (95% CI was 0.513-0.823, P = 0.044). When the optimal cut-off value was 21.90%, the sensitivity was 52.9%, the specificity was 79.5%, the positive predictive value was 0.500, and the negative predictive value was 0.813. Spearman correlation analysis showed that the MAR induced by each inducer was negatively correlated with inflammatory indicators and SOFA scores in sepsis patients, with A-MAR showing the strongest correlation with SOFA score ( r = -0.327, P = 0.000). Conclusions:MAR, an indicator of platelet aggregation function, shows predictive value for septic shock and septic shock with DIC, and it could be used to for evaluating the severity of patients with sepsis. In addition, tt alsocan be used as a monitoring index to predict the changes of sepsis patients and to guide the treatment.

OBJECTIVE: To establish a deep-learning architecture based on faster region-based convolutional neural networks (Faster R-CNN) algorithm for detection and sorting of red ginseng (Ginseng Radix et Rhizoma Rubra) with internal defects automatically on an online X-ray machine vision system. METHODS: A Faster R-CNN based classifier was trained with around 20 000 samples with mean average precision value (mAP) of 0.95. A traditional image processing method based on feedforward neural network (FNN) obtained a bad performance with the accuracy, recall and specificity of 69.0%, 68.0%, and 70.0%, respectively. Therefore, the Faster R-CNN model was saved to evaluate the model performance on the defective red ginseng online sorting system. RESULTS: An independent set of 2 000 red ginsengs were used to validate the performance of the Faster R-CNN based online sorting system in three parallel tests, achieving accuracy of 95.8%, 95.2% and 96.2%, respectively. CONCLUSION The overall results indicated that the proposed Faster R-CNN based classification model has great potential for non-destructive detection of red ginseng with internal defects.

OBJECTIVE: To explore the relevant clinical test indicators that affect the prognosis of patients with acute fatty liver of pregnancy (AFLP), and to provide a basis for early diagnosis and correct selection of treatment methods. METHODS: A retrospective analysis was conducted. Clinical data of AFLP patients in the intensive care unit (ICU) of the First Affiliated Hospital of Zhengzhou University from January 2010 to May 2021 were collected. According to the 28-day prognosis, the patients were divided into death group and survival group. The clinical data, laboratory examination indicators, and prognosis of the two groups were compared, and further binary Logistic regression analysis was used to analyze the risk factors affecting the prognosis of patients. At the same time, the values of related indicators at each time point (24, 48, 72 hours) after the start of treatment were recorded. The receiver operator characteristic curve (ROC curve) of prothrombin time (PT) and international normalized ratio (INR) for evaluating the prognosis of patients at each time point was drawn, and the area under the ROC curve (AUC) was calculated to evaluate the predictive value of relevant indicators at each time point for the prognosis of AFLP patients. RESULTS: A total of 64 AFLP patients were selected. The patients developed the AFLP during pregnancy (34.5±6.8) weeks, with 14 deaths (mortality of 21.9%) and 50 survivors (survival rate of 78.1%). There was no statistically significant difference in general clinical data between the two groups of patients, including age, time from onset to visit, time from visit to cessation of pregnancy, acute physiology and chronic health evaluations II (APACHE II), hospitalization time in ICU, and total hospitalization cost. However, the proportion of male fetuses and stillbirths in the death group was higher than that in the survival group. The laboratory examination indicators including the white blood cell count (WBC), alanine transaminase (ALT), serum creatinine (SCr), PT extension, INR elevation, and hyperammonia in the death group were significantly higher than those in the survival group (all P < 0.05). Through Logistic regression analysis of the above indicators showed that PT > 14 s and INR > 1.5 were risk factors affecting the prognosis of AFLP patients [PT > 14 s: odds ratio (OR) = 1.215, 95% confidence interval (95%CI) was 1.076-1.371, INR > 1.5: OR = 0.719, 95%CI was 0.624-0.829, both P < 0.01]. ROC curve analysis showed that both PT and INR at ICU admission and 24, 48, and 72 hours of treatment can evaluate the prognosis of AFLP patients [AUC and 95%CI of PT were 0.772 (0.599-0.945), 0.763 (0.608-0.918), 0.879 (0.795-0.963), and 0.957 (0.904-1.000), respectively; AUC and 95%CI of INR were 0.808 (0.650-0.966), 0.730 (0.564-0.896), 0.854 (0.761-0.947), and 0.952 (0.896-1.000), respectively; all P < 0.05], the AUC of PT and INR after 72 hours of treatment was the highest, with higher sensitivity (93.5%, 91.8%) and specificity (90.9%, 90.9%). CONCLUSIONS AFLP often occurs in the middle and late stages of pregnancy, and the initial symptoms are mainly gastrointestinal symptoms. Once discovered, pregnancy should be terminated immediately. PT and INR are good indicators for evaluating AFLP patient efficacy and prognosis, and PT and INR are the best prognostic indicators after 72 hours of treatment.
Humans ; Male ; Prognosis ; ROC Curve ; Retrospective Studies ; Intensive Care Units ; Sepsis/diagnosis*

OBJECTIVE: To explore the characteristics of changes in peripheral blood lymphocyte subsets in patients with sepsis in intensive care unit (ICU) and analyze their predictive value for prognosis. METHODS: The clinical data of sepsis patients admitted to the surgical intensive care unit (SICU) of the First Affiliated Hospital of Zhengzhou University from January 2020 to December 2021 were analyzed retrospectively. The patients met the diagnostic criteria of Sepsis-3 and were ≥ 18 years old. Peripheral venous blood samples were collected from all patients on the next morning after admission to SICU for routine blood test and peripheral blood lymphocyte subsets. According to the 28-day survival, the patients were divided into two groups, and the differences in immune indexes between the two groups were compared. Logistic regression analysis was used to analyze the risk factors of immune indexes that affect prognosis. RESULTS: (1) A total of 279 patients with sepsis were enrolled in the experiment, of which 198 patients survived at 28 days (28-day survival rate 71.0%), and 81 patients died (28-day mortality 29.0%). There were no significant differences in age (years old: 57.81±1.71 vs. 54.99±1.05) and gender (male: 60.5% vs. 63.6%) between the death group and the survival group (both P > 0.05), and the baseline data was comparable.(2) Acute physiology and chronic health evalution II (APACHE II: 22.06±0.08 vs. 14.08±0.52, P < 0.001), neutrophil percentage [NEU%: (88.90±1.09)% vs. (84.12±0.77)%, P = 0.001], procalcitonin [PCT (μg/L): 11.97±2.73 vs. 5.76±1.08, P = 0.011], platelet distribution width (fL: 16.81±0.10 vs. 16.57±0.06, P = 0.029) were higher than those in the survival group, while lymphocyte percentage [LYM%: (6.98±0.78)% vs. (10.59±0.86)%, P = 0.012], lymphocyte count [LYM (×10⁹/L): 0.70±0.06 vs. 0.98±0.49, P = 0.002], and platelet count [PLT (×10⁹/L): 151.38±13.96 vs. 205.80±9.38, P = 0.002], and thrombocytocrit [(0.15±0.01)% vs. (0.19±0.07)%, P = 0.012] were lower than those in the survival group. (3) There was no statistically significant difference in the percentage of lymphocyte subsets between the death group and the survival group, but the absolute value of LYM (pieces/μL: 650.24±84.67 vs. 876.64±38.02, P = 0.005), CD3⁺ absolute value (pieces/μL: 445.30±57.33 vs. 606.84±29.25, P = 0.006), CD3⁺CD4⁺ absolute value (pieces/μL: 239.97±26.96 vs. 353.49±18.59, P = 0.001), CD19⁺ absolute value (pieces/μL: 111.10±18.66 vs. 150.30±10.15, P = 0.049) in the death group was lower than those in the survival group. Other lymphocyte subsets in the death group, such as CD3⁺CD8⁺ absolute value (pieces/μL: 172.40±24.34 vs. 211.22±11.95, P = 0.112), absolute value of natural killer cell [NK (pieces/μL): 101.26±18.15 vs. 114.72±7.64, P = 0.420], absolute value of natural killer T cell [NKT (pieces/μL): 33.22±5.13 vs. 39.43±2.85, P = 0.262], CD4^-CD8^- absolute value (pieces/μL: 41.07±11.07 vs. 48.84±3.31, P = 0.510), CD4⁺CD8⁺ absolute value (pieces/μL: 3.39±1.45 vs. 3.47±0.36, P = 0.943) were not significantly different from those in the survival group. (4)Logistic regression analysis showed that lymphocyte subsets were not selected as immune markers with statistical significance for the prognosis of sepsis. CONCLUSIONS The changes of immune indexes in sepsis patients are closely related to their prognosis. Early monitoring of the above indexes can accurately evaluate the condition and prognosis of sepsis patients.
Humans ; Male ; Adolescent ; Retrospective Studies ; ROC Curve ; Sepsis/diagnosis* ; Lymphocyte Count ; Lymphocyte Subsets ; Prognosis ; Killer Cells, Natural