OBJECTIVE: To develop a method for identifying high-risk patients among septic populations requiring mechanical ventilation, and to conduct phenotypic analysis based on this method. METHODS: Data from four sources were utilized: the Medical Information Mart for Intensive Care (MIMIC-IV 2.0, MIMIC-III 1.4), the Philips eICU-Collaborative Research Database 2.0 (eICU-CRD 2.0), and the Anhui Medical University Second Affiliated Hospital dataset. The adult patients in intensive care unit (ICU) who met Sepsis-3 and received invasive mechanical ventilation (IMV) on the first day of first admission were enrolled. The MIMIC-IV dataset with the highest data integrity was divided into a training set and a test set at a 6:1 ratio, while the remaining datasets were served as validation sets. The demographic information, comorbidities, laboratory indicators, commonly used ICU scores, and treatment measures of patients were extracted. Clinical data collected within first day of ICU admission were used to calculate the sequential organ failure assessment (SOFA) score. K-means clustering was applied to cluster SOFA score components, and the sum of squared errors (SSE) and Davies-Bouldin index (DBI) were used to determine the optimal number of disease subtypes. For clustering results, normalized methods were employed to compare baseline characteristics by visualization, and Kaplan-Meier curves were used to analyze clinical outcomes across phenotypes. RESULTS: This study enrolled patients from MIMIC-IV dataset (n = 11 166), MIMIC-III dataset (n = 4 821), eICU-CRD dataset (n = 6 624), and a local dataset (n = 110), with the four datasets showing similar median ages and male proportions exceeding 50%; using 85% of the MIMIC-IV dataset as the training set, 15% as the test set, and the rest dataset as the validation set. K-means clustering based on the six-item SOFA score was performed to determine the optimal number of clusters as 3, and patients were finally classified into three phenotypes. In the training set, compared with the patients with phenotype II and phenotype III, those with phenotype I had the more severe circulatory and respiratory dysfunction, a higher proportion of vasoactive drug usage, more obvious metabolic acidosis and hypoxia, and a higher incidence of congestive heart failure. The patients with phenotype II was dominated by respiratory dysfunction with higher visceral injury. The patients with phenotype III had relatively stable organ function. The above characteristics were consistent in both the test and validation sets. Analysis of infection-related indicators showed that the patients with phenotype I had the highest SOFA score within 7 days after ICU admission, initial decreases and later increases in platelet count (PLT), and higher counts of neutrophils, lymphocytes, and monocytes as compared with those with phenotype II and phenotype III, their blood cultures had a higher positivity rates for Gram-positive bacteria, Gram-negative bacteria and fungi as compared with those with phenotype II and phenotype III. The Kaplan-Meier curve indicated that in the training, test, and validation sets, the 28-day cumulative mortality of patients with phenotype I was significantly higher than that of patients with phenotypes II and phenotype III. CONCLUSIONS Three distinct phenotypes in septic patients receiving IMV based on unsupervised machine learning is derived, among which phenotype I, characterized by cardiorespiratory failure, can be used for the early identification of high-risk patients in this population. Moreover, this population is more prone to bloodstream infections, posing a high risk and having a poor prognosis.
Humans ; Machine Learning ; Sepsis/therapy* ; Respiration, Artificial ; Intensive Care Units ; Organ Dysfunction Scores ; Male ; Female ; Middle Aged ; Adult

Objective:To evaluate the clinical and prognostic differences in acute myeloid leukemia with myelodysplasia-related changes (AML-MRC) children under different diagnostic criteria (World Health Organization (WHO) 2016 and WHO 2022 criteria).Methods:In this retrospective cohort study, clinical characteristics and prognosis information of 260 acute myeloid leukemia (AML) children admitted to Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences from August 2017 to August 2021 were analyzed retrospectively. According to WHO 2016 and WHO 2022 diagnostic criteria, patients were divided into AML-MRC group and non-AML-MRC group, the prognostic and genetic differences between two groups were compared respectively. Meanwhile, the characteristics of children with 8 MRC-related genes defined in WHO 2022 diagnostic criteria were described. Mann-Whitney U test, chi-square test were used for comparison between groups. Survival curve was plotted by Kaplan-Meier method, and comparison between groups was performed by Log-Rank method. Results:Among the 260 children, there were 148 males and 112 females. The follow-up time was 26 (16, 38) months. A total of 28 children (10.8%) were diagnosed with AML-MRC according to the WHO 2016 diagnostic criteria. Compared with non-AML-MRC children, the frequency of PTPN11, RUNX11, SH2B3, MPL and STAG2 mutations was higher in AML-MRC children (25.0% (7/28) vs. 4.3% (10/232), 14.3% (4/28) vs. 3.9% (9/232), 10.7% (3/28) vs. 2.2% (5/232), 10.7% (3/28) vs. 2.2% (5/232), 10.7% (3/28) vs. 0.9% (2/232), all P<0.05). The 2-year overall survival (OS) and events free survival (EFS) rate of 28 AML-MRC children under WHO 2016 diagnostic criteria were worse than those of 232 non-AML-MRC children ((62.1±10.8)% vs. (94.5±1.6)%, χ2=22.1 ,P<0.001;(48.0±10.6)% vs. (70.9±3.2)%, χ2=6.33, P=0.012). Twenty-seven children (10.4%) were eventually diagnosed with AML-MRC according to WHO 2022 criteria, their 2-year OS rate were worse than 233 non-AML-MRC children ((60.8±11.1)% vs. (94.5±1.6)%, χ2=24.49 ,P<0.001), and there was no statistically significant difference in EFS rate between two groups at 2 years ((55.1±10.8)% vs. (70.1±3.2)%, χ2=2.44 , P=0.119). Conclusions:Compared with the 2022 WHO diagnostic criteria, the survival rates of children with AML-MRC under the 2016 WHO diagnostic criteria were worse than that of children without MRC.The new version of the AML-MRC diagnostic criteria emphasizes the importance of genes.

Objective:To investigate the clinical features and prognosis of testicular relapse in pediatric acute lymphoblastic leukemia (ALL).Methods:Clinical data including the age, time from initial diagnosis to recurrence, relapse site, and therapeutic effect of 37 pediatric ALL with testicular relapse and treated in Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences between November 2011 and December 2022 were analyzed retrospectively. Patients were grouped according to different clinical data. Kaplan-Meier analysis was used to evaluate the overall survival (OS) rate and event free survival (EFS) rate for univariate analysis, and Cox proportional-hazards regression model was used to evaluate the influencing factors of OS rate and EFS rate for multivariate analysis.Results:The age at initial diagnosis of 37 pediatric testicular relapse patients was (5±3) years and the time from initial diagnosis to testicular recurrence was (37±15) months. The follow-up time was 43 (22, 56) months. Twenty-three patients (62%) were isolated testis relapse. The 5-year OS rate and EFS rate of the 37 relapsed children were (60±9) % and (50±9) % respectively. Univariate analysis showed that the 2-year EFS rate in the group of patients with time from initial diagnosis to testicular recurrence >28 months was significantly higher than those ≤28 months ((69±10)% vs. (11±11)%, P<0.05), 2-year EFS rate of the isolated testicular relapse group was significantly higher than combined relapse group ((66±11)% vs. (20±13) %, P<0.05), 2-year EFS rate of chimeric antigen receptor T (CAR-T) cell treatment after relapse group was significantly higher than without CAR-T cell treatment after relapse group ((78±10)% vs. (15±10)%, P<0.05). ETV6-RUNX1 was the most common genetic aberration in testicular relapsed ALL (38%, 14/37). The 4-year OS and EFS rate of patients with ETV6-RUNX1 positive were (80±13) % and (64±15) %, respectively. Multivariate analysis identified relapse occurred≤28 months after first diagnosis ( HR=3.09, 95% CI 1.10-8.72), combined relapse ( HR=4.26, 95% CI 1.34-13.52) and CAR-T cell therapy after relapse ( HR=0.15,95% CI 0.05-0.51) were independent prognostic factors for 2-year EFS rate (all P<0.05). Conclusions:The outcome of testicular relapse in pediatric ALL was poor. They mainly occurred 3 years after initial diagnosis. ETV6-RUNX1 is the most common abnormal gene.Patients with ETV6-RUNX1 positive often have a favorable outcome. Early relapse and combined relapse indicate unfavorable prognosis, while CAR-T cell therapy could significantly improve the survival rate of children with testicular recurrence.

Objective:To investigate the clinical characteristics and long-term prognostic factors of relapsed pediatric acute lymphoblastic leukemia (ALL).Methods:Clinical data including the age, time from initial diagnosis to relapse, relapse site, and molecular biological features of 217 relapsed ALL children primarily treated by the Chinese Children's Leukemia Group (CCLG)-ALL 2008 protocol in Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences between April 2008 and April 2015 were collected and analyzed in this retrospective cohort study. Kaplan-Meier analysis was used to evaluate the overall survival (OS) rate and event free survival (EFS) rate for univariate analysis, and Cox proportional-hazards regression model was used to evaluate the influencing factors of OS rate and EFS rate for multivariate analysis.Results:The age at initial diagnosis of 217 relapsed patients was 5 (3, 7) years. There were 135 males and 82 females. The time from initial diagnosis to relapse of 217 children was 22 (10, 39) months. After relapse, 136 out of 217 children (62.7%) received treatment and the follow-up time was 65 (47, 90) months. The 5-year OS rate and EFS rate of the 136 relapsed children were (37±4) % and (26±4) %, respectively. The predicted 10-year OS rate and EFS rate were (35±5) % and (20±4) %, respectively. Univariate analysis showed that the 5-year OS rate in the group of patients with late relapse (43 cases) was significantly higher than those with very early (54 cases) and early relapse (39 cases) ((72±7)% vs. (16±5)%, (28±8)%, χ2=35.91, P<0.05), 5-year OS rate of the isolated extramedullary relapse group (20 cases) was significantly higher than isolated bone marrow relapse group (102 cases) and combined relapse group (14 cases) ((69±11)% vs. (31±5)%, (29±12)%, χ2=9.14, P<0.05), 5-year OS rate of high-risk group (80 cases) was significantly lower than standard-risk group (10 cases) and intermediate-risk group (46 cases) ((20±5)% vs. (90±10)%, (54±8)%, χ2=32.88, P<0.05). ETV6::RUNX1 was the most common fusion gene (13.2%, 18/136). The predicted 10-year OS rate of relapsed children with positive ETV6::RUNX1 was significantly higher than those without ETV6::RUNX1 (118 cases) ((83±9)% vs. (26±5)%, χ2=14.04, P<0.05). The 5-year OS for those accepted hematopoietic stem cell transplantation (HSCT) after relapse (42 cases) was higher than those without HSCT (94 cases) ((56±8)% vs. (27±5)%, χ2=15.18, P<0.05). Multivariate analysis identified very early/early relapse ( HR=3.91, 95% CI 1.96-7.79; HR=4.15, 95% CI 1.99-8.67), bone marrow relapse including isolated bone marrow relapse and combined relapse ( HR=6.50, 95% CI 2.58-16.34; HR=5.19, 95% CI 1.78-15.16), with ETV6::RUNX1 ( HR=0.23, 95% CI 0.07-0.74) and HSCT after relapse ( HR=0.24, 95% CI 0.14-0.43) as independent prognostic factors for OS (all P<0.05). Conclusions:Relapsed pediatric ALL mainly occurs very early and often affects bone marrow, which confer poor outcome. ETV6::RUNX1 is the most common genetic aberration with a favorable outcome. HSCT could rescue the outcome of relapsed children, though the survival rate is still poor.

Aim To study whether genetically predicted serum testosterone level is causally associated with sys-temic multisite atherosclerosis.Methods Based on two pooled databases of genome-wide association studies on testos-terone and atherosclerosis in European populations from two separate foreign countries,the causal effect between testosterone and atherosclerosis was assessed using two-sample Mendelian randomization analysis with data on testosterone-associated genetic variants as instrumental variables(Ⅳ),and by using the inverse-variance weighted(IVW)method,MR-Egger regression,and weighted median estimation.Results IVW results showed that genetically predicted circu-lating testosterone levels were negatively associated with the risk of peripheral atherosclerosis(OR=0.93,95％CI:0.86～1.00,P=0.01),and that elevated testosterone level may reduce the risk of developing peripheral atherosclerosis,while no evidence of a potential causal association was found with cerebral atherosclerosis,coronary atherosclerosis and other athero-sclerosis type(P＞O.05).Conclusion The final analysis showed a causal relationship between genetically predicted testosterone level and the risk of developing peripheral atherosclerosis,and the role of testosterone therapy in the prevention and treatment of atherosclerosis deserves attention and further study.

Venous thromboembolism (VTE) is a complication in children with acute lymphoblastic leukemia (ALL). The Chinese Children's Cancer Group-ALL-2015 protocol was carried out in China, and epidemiology, clinical characteristics, and risk factors associated with VTE were analyzed. We collected data on VTE in a multi-institutional clinical study of 7640 patients with ALL diagnosed in 20 hospitals from January 2015 to December 2019. First, VTE occurred in 159 (2.08%) patients, including 90 (56.6%) during induction therapy and 108 (67.92%) in the upper extremities. T-ALL had a 1.74-fold increased risk of VTE (95% CI 1.08-2.8, P = 0.022). Septicemia, as an adverse event of ALL treatment, can significantly promote the occurrence of VTE (P < 0.001). Catheter-related thrombosis (CRT) accounted for 75.47% (n = 120); and, symptomatic VTE, 58.49% (n = 93), which was more common in patients aged 12-18 years (P = 0.023), non-CRT patients (P < 0.001), or patients with cerebral thrombosis (P < 0.001). Of the patients with VTE treated with anticoagulation therapy (n = 147), 4.08% (n = 6) had bleeding. The VTE recurrence rate was 5.03% (n = 8). Patients with VTE treated by non-ultrasound-guided venous cannulation (P = 0.02), with residual thrombus (P = 0.006), or with short anticoagulation period (P = 0.026) had high recurrence rates. Thus, preventing repeated venous puncture and appropriately prolonged anticoagulation time can reduce the risk of VTE recurrence.
Humans ; Child ; Venous Thromboembolism/etiology* ; East Asian People ; Precursor Cell Lymphoblastic Leukemia-Lymphoma/epidemiology* ; Risk Factors ; Thrombosis/chemically induced* ; China/epidemiology* ; Anticoagulants/adverse effects* ; Recurrence

Objective:To explore the clinical features of bloodstream infections (BSI) in children with acute myeloid leukemia (AML) during the first induction chemotherapy.Methods:The clinical data, pathogen of BSI, antibiotic susceptibility in vitro, complications and prognosis of 204 newly diagnosed AML children admitted to Blood Diseases Hospital, Chinese Academy of Medical Sciences from August 2009 to December 2015 were analyzed retrospectively. χ 2 test was used for the comparison between groups and Logistic regression was used for BSI risk factor analysis. Results:Among 204 patients, 116 were males and 88 were females. The age was 8 (ranged from 1 to 14) years. Among them, 170 patients received MAE chemotherapies (etoposide, mitoxantrone and cytarabine) and 25 received IAE chemotherapies (etoposide, idarubicin and cytarabine). The other 9 patients used granulocyte colony stimulating factor (G-CSF)-priming regimen (aclacinomycin or homoharringtonine, cytarabine and G-CSF) for induction treatments. A total of 28 patients experienced BSI and the incidence rate was 13.7% (28/204), 26 of them developed BSI once and 2 patients developed twice. Gram-positive bacteria were predominant pathogens accounting for 53.3% (16/30) while gram-negative bacteria accounting for 40.0% (12/30) and fungal accounted for 6.7% (2/30). The most common detected pathogens were Coagulase negative Staphylococcus (CoNS, 26.7% (8/30)), followed by Streptococcus spp. (13.3% (4/30)) and Escherichia coli (13.3% (4/30)). Among Gram-negative bacteria (GNB), 3 cases showed carbapenem resistance and 2 cases were Stenotrophomonas maltophilia. BSI-related mortality was 28.6% (8/28). Infections caused by drug-resistant GNB or fungi resulted in 6 fatal cases. The incidence rate of BSI in group with severe neutropenia was higher than in group without it (16.6% (25/151) vs. 5.7% (3/53), χ2=3.933, P=0.047). Multivariable analysis showed severe neutropenia at the onset of fever was independent risk factor of BSI ( OR=4.258,95% CI 1.097-16.524, P=0.036). Conclusions:During the first induction chemotherapy courses, Gram-positive bacteria cause most of the BSI. Drug-resistant bacteria related infection often result in fatal outcomes. Severe neutropenia is a significant risk factor.

Background Papillary thyroid cancer is the result of a variety of pathogenic factors. The prevalence of papillary thyroid cancer varies greatly in different regions, and the disease is more harmful to women. Objective This study aims to explore the relationship between reproductive factors and papillary thyroid cancer, and to provide basic data for prevention and control of the disease. Methods A 1∶1 age (±3 years) matched case-control study was conducted in 331 pairs of newly confirmed papillary thyroid cancer cases and controls from two hospitals in Shanghai from November 2012 to December 2013. Comparisons were made in the history of menstruation, pregnancy, gynecological and breast diseases, and other variables between the two groups. Results The results of univariate analysis indicated that the proportions of education below bachelor degree, married, and mainly manual workers in the case group were significantly higher than those in the control group (P<0.05); the proportion of those with a family history of thyroid diseases in the case group was significantly higher than that in the control group (P<0.05); the proportions of the cases with irregular menstruation (77.34%), a history of using oral contraceptive or hormone drugs (24.77%), a history of benign gynecological diseases (31.72%), and a history of gynecological surgery (9.67%) were significantly higher than the proportions in the control group (P<0.05). After adjusting potential confounding factors such as history of CT examination, age, kinds of family salt, total iodine intake every day, education level, occupation, marital status, body mass index, and family history of thyroid diseases, the results of multiple logistic regression analysis showed that irregular menstruation (OR=1.767, 95%CI: 1.122-2.782; P=0.014), surgical menopause (OR=12.787, 95%CI: 3.202-51.057; P<0.001), pregnancy >1 time (OR =2.490, 95%CI: 1.196-5.184; P=0.015), and the history of using oral contraceptive or hormone drugs (OR=2.389, 95%CI: 1.338-4.268; P=0.003) were the risk factors of papillary thyroid cancer. Conclusion Irregular menstruation, surgical menopause, history of pregnancy, and history of using oral contraceptive or hormone drugs might be the risk factors of papillary thyroid cancer. To reduce the incidence of papillary thyroid cancer, strengthened health education and rational use of contraceptives are recommended.

Objective:To explore the progress of microcirculation research in shock and sepsis from 2000 to 2019 based on knowledge visualization analysis.Methods:The literatures related to microcirculation of shock and sepsis published in Web of Science and Wanfang databases were collected from 2000 to 2019. Then, data collected was sorted out and used to make bar charts and curves reflecting the growth trend of the literatures using Excel software. The information about country, institution and author were extracted by CiteSpace 5.0 R1 for generating co-occurring network, to find the main research power and each cooperative relation. This software was also used to analyze the related-keywords and cited reference, so that the map of co-citation reference was drawn subsequently to explore the frontiers and hot spots in this field.Results:There were a total of 2 000 Chinese "papers" or "dissertations", 1 823 English "articles" or "reviews", collected preliminarily. The Chinese literatures were only statistically described for the amounts, institutions and authors, however, all retrieved English literatures were enrolled in the visualization analysis eventually. From 2009 to 2019, the number of English literatures rose steadily; however, the number of Chinese literatures had been sliding since 2011. On terms of quality, the impact factors of most literatures were under 6, lacking high quality ones, and the number remained stable every year. According to the analysis of English literatures, the top three countries that published related papers around the world were Germany ( n = 430), the United States ( n = 401) and Netherlands ( n = 223). In the aspect of research institutions, the University of Amsterdam (Netherlands) ranked first in the Web of Science ( n = 113), while the institution in China with the most publications was the Southern Medical University ( n = 71). According to the analysis of co-occurring author network, there were three major globe scientific groups in which Can Ince, Daniel De Backer and Jean-Louis Vincent contributed most to this research field. The top three authors who published most Chinese papers were Niu Chunyu ( n = 20), Zhao Zigang ( n = 18) and Duan Meili ( n = 13). By generating the co-occurring keyword mapping, the research hot spots mainly focused on "blood flow", "nitric oxide", "hemorrhagic shock" and "perfusion". However, these burst keywords, including "cardiogenic shock", "acute kidney injury", "fluid resuscitation", "sublingual microcirculation", "mortality", "oxidative stress", and "critically ill patient", represented the frontiers in microcirculation research of shock and sepsis field. Finally, the results of co-citation reference analysis showed that "sublingual microcirculation" and "mottling score" were most active, indicating that the research of microcirculation monitoring technology could be considered as a hot spot and also the frontier in this field, which was consistent with the results of co-occurring keywords network. Conclusions:Knowledge visualization analysis can visually exhibit the hot spot and trend of microcirculation research in shock and sepsis. The power was mainly focused on a few developed countries in Europe, and the trend was more inclined to the related microcirculation monitoring technology.

Objective: To investigate the efficacy and the prognostic factors of Chinese Academy of Medical Sciences 2005 (CAMS-2005) regimen in the treatment of pediatric acute myeloid leukemia (AML). Methods: Eighty-eight cases of newly-diagnosed AML patients, who were treated with the CAMS-2005 regimen from April 2005 to July 2009, were enrolled in this case observational study. Clinical characteristics, long-term prognosis and prognostic factors were analyzed retrospectively. Overall survival (OS) and event free survival (EFS) rates were estimated by the Kaplan-Meier method. Rates of survival between the groups were compared by the Log-rank test. Prognostic factors were evaluated by COX regression analysis. Results: A total of 82 cases were enrolled in this study, including 34 core binding factor(CBF)-AML patients and 48 non-CBF-AML patients. There were 45 males and 37 females. The median age at diagnosis was 8.0 (0.7-16.0) years. During the induction therapy, 3 patients (4%) developed treatment-related early-death, while 63 patients (77%) achieved complete remission (CR) and 53 patients (65%) achieved CR after 1 course. Twenty-one patients (33%) had relapsed disease. The CR rates of CBF-AML patients and non-CBF-AML patients were 91% (31/34) and 67% (32/48) (χ²=5.410, P=0.020) , while the relapse rates were 29% (9/31) and 38% (12/32) (χ²=0.508, P=0.476) . The 8-year OS and EFS rates of all 82 patients were 59%(48/82) and 51%(42/82). The 8-year OS rates of CBF-AML patients and non-CBF-AML patients were 74% (25/34) and 48%(23/48) (χ²=5.812, P=0.016), while the 8-year EFS rates of CBF-AML patients and non-CBF-AML patients were 71%(24/34) and 38%(18/48) (χ²=8.682, P=0.003). There were statistically significant differences between groups. The 8-year OS rates of patients who achieved CR after 1 course and other patients were 68% (36/53) and 46% (12/26) (χ²=9.606, P=0.002), while the 8-year EFS rates were 66% (35/53) and 27% (7/26) (χ²=19.471, P=0.000), the differences were all statistically significant. COX multivariate analysis showed that CBF-AML or non-CBF-AML and whether achieved CR after 1 course were independent prognostic factors of OS rates (relative risk: 2.538, 2.561) and EFS rates (relative risk: 3.050, 3.686) (P <0.05). Conclusions The efficacy of the CAMS-2005 regimen in the treatment of AML patients was well. CBF-AML or non-CBF-AML and whether achieved CR after 1 course were independent prognostic factors for pediatric AML patients.