Humans ; Critical Illness/rehabilitation* ; Child ; China

PURPOSE: Septic shock is associated with high mortality and poor outcomes among sepsis patients with coagulopathy. Although traditional statistical methods or machine learning (ML) algorithms have been proposed to predict septic shock, these potential approaches have never been systematically compared. The present work aimed to develop and compare models to predict septic shock among patients with sepsis. METHODS: It is a retrospective cohort study based on 484 patients with sepsis who were admitted to our intensive care units between May 2018 and November 2022. Patients from the 908th Hospital of Chinese PLA Logistical Support Force and Nanchang Hongdu Hospital of Traditional Chinese Medicine were respectively allocated to training (n=311) and validation (n=173) sets. All clinical and laboratory data of sepsis patients characterized by comprehensive coagulation indexes were collected. We developed 5 models based on ML algorithms and 1 model based on a traditional statistical method to predict septic shock in the training cohort. The performance of all models was assessed using the area under the receiver operating characteristic curve and calibration plots. Decision curve analysis was used to evaluate the net benefit of the models. The validation set was applied to verify the predictive accuracy of the models. This study also used Shapley additive explanations method to assess variable importance and explain the prediction made by a ML algorithm. RESULTS: Among all patients, 37.2% experienced septic shock. The characteristic curves of the 6 models ranged from 0.833 to 0.962 and 0.630 to 0.744 in the training and validation sets, respectively. The model with the best prediction performance was based on the support vector machine (SVM) algorithm, which was constructed by age, tissue plasminogen activator-inhibitor complex, prothrombin time, international normalized ratio, white blood cells, and platelet counts. The SVM model showed good calibration and discrimination and a greater net benefit in decision curve analysis. CONCLUSION The SVM algorithm may be superior to other ML and traditional statistical algorithms for predicting septic shock. Physicians can better understand the reliability of the predictive model by Shapley additive explanations value analysis.
Humans ; Shock, Septic/blood* ; Machine Learning ; Male ; Female ; Retrospective Studies ; Middle Aged ; Aged ; Sepsis/complications* ; ROC Curve ; Cohort Studies ; Adult ; Intensive Care Units ; Algorithms ; Blood Coagulation ; Critical Illness

OBJECTIVES: To explore the risk factors of feeding intolerance (FI) in critically ill children receiving enteral nutrition (EN) and to construct a prediction nomogram model for FI. METHODS: A retrospective study was conducted to collect data from critically ill children admitted to the Pediatric Intensive Care Unit of Xiangya Hospital, Central South University, between January 2015 and October 2020. The children were randomly divided into a training set (346 cases) and a validation set (147 cases). The training set was further divided into a tolerance group (216 cases) and an intolerance group (130 cases). Multivariate logistic regression analysis was used to screen for risk factors for FI in critically ill children receiving EN. A nomogram was constructed using R language, which was then validated on the validation set. The model's discrimination, calibration, and clinical net benefit were evaluated using receiver operating characteristic curves, calibration curves, and decision curves. RESULTS: Duration of bed rest, shock, gastrointestinal decompression, use of non-steroidal anti-inflammatory drugs, and combined parenteral nutrition were identified as independent risk factors for FI in critically ill children receiving EN (P<0.05). Based on these factors, a nomogram prediction model for FI in critically ill children receiving EN was developed. The area under the receiver operating characteristic curve for the training set and validation set was 0.934 (95%CI: 0.906-0.963) and 0.852 (95%CI: 0.787-0.917), respectively, indicating good discrimination of the model. The Hosmer-Lemeshow goodness-of-fit test showed that the model had a good fit (χ 2=12.559, P=0.128). Calibration curve and decision curve analyses suggested that the model has high predictive efficacy and clinical application value. CONCLUSIONS Duration of bed rest, shock, gastrointestinal decompression, use of non-steroidal anti-inflammatory drugs, and combined parenteral nutrition are independent risk factors for FI in critically ill children receiving EN. The nomogram model developed based on these factors exhibits high predictive efficacy and clinical application value.
Humans ; Critical Illness ; Enteral Nutrition/adverse effects* ; Male ; Risk Factors ; Female ; Child, Preschool ; Infant ; Nomograms ; Retrospective Studies ; Child ; Logistic Models

To guide clinical blood transfusion practices for pediatric patients, the National Health Commission has issued the health standard "Guideline for pediatric transfusion" (WS/T 795-2022). Critically ill children often present with anemia and have a higher demand for transfusions compared to other pediatric patients. This guideline provides guidance and recommendations for blood transfusions in cases of general critical illness, septic shock, acute brain injury, extracorporeal membrane oxygenation, non-life-threatening bleeding, and hemorrhagic shock. This article interprets the background and evidence of the blood transfusion provisions for critically ill and severely bleeding children in the "Guideline for pediatric transfusion", aiming to enhance understanding and implementation of this aspect of the guidelines. Citation:Chinese Journal of Contemporary Pediatrics, 2025, 27(4): 395-403.
Humans ; Critical Illness ; Blood Transfusion/standards* ; Child ; Hemorrhage/therapy* ; Practice Guidelines as Topic

OBJECTIVES: To investigate the association of serum albumin level after human albumin infusion with 28-day mortality in critically ill patients with acute kidney injury (AKI) and its impact on 90-day outcomes of the patients. METHODS: We conducted a retrospective cohort study based on the MIMIC IV database (2008-2019), including 5918 AKI patients treated with albumin in the ICU. Based on serum albumin levels within 72 h after albumin infusion, the patients were divided into low (<30 g/L), medium (30-35 g/L), and high albumin (>35 g/L) groups. Restricted cubic spline regression and multivariate logistic regression were used to analyze the association of albumin levels with patient mortality, and the results were verified in a external validation cohort consisting of 110 sepsis-induced AKI patients treated in Nanfang Hospital between 2017 and 2022 using survival analysis and multivariate adjustment. RESULTS: In the MIMIC training cohort, multivariate logistic regression showed no significant differences in 28-day mortality of the patients with different albumin levels (P>0.05). However, restricted cubic spline analysis indicated a non-linear dose-response relationship between albumin levels and 28-day mortality (threshold effect: risk increased when albumin levels >3.6 g/dL). Secondary endpoint analysis revealed that the patients with high albumin levels had a shorter duration of mechanical ventilation (P<0.001) but a longer ICU stay (P<0.001). In the validation cohort, albumin levels ≥30 g/L were significantly associated with a reduced 28-day mortality rate (P<0.05). CONCLUSIONS The association between increased serum albumin levels following albumin infusion and 28-day mortality of critically ill patients with AKI exhibits a cohort dependency and can be influenced by multiple factors including disease type and severity, infusion strategies, and statistical methods.
Humans ; Acute Kidney Injury/therapy* ; Critical Illness/mortality* ; Retrospective Studies ; Serum Albumin/analysis* ; Male ; Female ; Intensive Care Units ; Middle Aged ; Logistic Models ; Aged

OBJECTIVE: To explore the feasibility of incorporating simple bedside indicators into death predictive model for elderly critically ill patients based on interpretability machine learning algorithms, providing a new scheme for clinical disease assessment. METHODS: Elderly critically ill patients aged ≥ 65 years who were hospitalized in the intensive care unit (ICU) of Tacheng People's Hospital of Ili Kazak Autonomous Prefecture from June 2017 to May 2020 were retrospectively selected. Basic parameters including demographic characteristics, basic vital signs and fluid intake and output within 24 hours after admission, as well acute physiology and chronic health evaluation II (APACHE II), Glasgow coma score (GCS) and sequential organ failure assessment (SOFA) were also collected. According to outcomes in hospital, patients were divided into survival group and death group. Four datasets were constructed respectively, namely baseline dataset (B), including age, body temperature, heart rate, pulse oxygen saturation, respiratory rate, mean arterial pressure, urine output volume, infusion volume, and crystal solution volume; B+APACHE II dataset (BA), B+GCS dataset (BG), and B+SOFA dataset (BS). Then three machine learning algorithms, Logistic regression (LR), extreme gradient boosting (XGboost) and gradient boosting decision tree (GBDT) were used to develop the corresponding mortality predictive models within four datasets. The feature importance histogram of each prediction model was drawn by SHapley additive explanation (SHAP) method. The area under curve (AUC), accuracy and F1 score of each model were compared to determine the optimal prediction model and then illuminate the nomogram. RESULTS: A total of 392 patients were collected, including 341 in the survival group and 51 in the death group. There were statistically significant differences in heart rate, pulse oxygen saturation, mean arterial pressure, infusion volume, crystal solution volume, and etiological distribution between the two groups. The top three causes of death were shock, cerebral hemorrhage, and chronic obstructive pulmonary disease. Among the 12 prognostic models trained by three machine learning algorithms, overall performance of prognostic models based on B dataset was behind, whereas the LR model trained by BA dataset achieved the best performance than others with AUC of 0.767 [95% confidence interval (95%CI) was 0.692-0.836], accuracy of 0.875 (95%CI was 0.837-0.903) and F1 score of 0.190. The top 3 variables in this model were crystal solution volume with first 24 hours, heart rate and mean arterial pressure. The nomogram of the model showed that the total score between 150 and 230 were advisable. CONCLUSION The interpretable machine learning model including simple bedside parameters combined with APACHE II score could effectively identify the risk of death in elderly patients with critically illness.
Humans ; Critical Illness ; Machine Learning ; Aged ; Algorithms ; Intensive Care Units ; Retrospective Studies ; APACHE ; Prognosis ; Organ Dysfunction Scores ; Hospital Mortality ; Male ; Female

OBJECTIVE: To observe the effects of enhanced rehabilitation on the prognosis of critically ill patients in the intensive care unit (ICU). METHODS: A single-center retrospective historical controlled study was conducted, patients admitted to the ICU of Peking University First Hospital from May 1, 2020, to April 30, 2021, and from October 1, 2021, to September 30, 2022 were enrolled. According to the different rehabilitation treatment strategies during different periods, patients were divided into the conventional rehabilitation group (patients receiving conventional rehabilitation treatment from May 1, 2020, to April 30, 2021) and the enhanced rehabilitation group (patients receiving the therapy of multidisciplinary team, ie medical care-rehabilitation-nursing care from October 1, 2021, to September 30, 2022). General data, acute physiology and chronic health evaluation II (APACHE II), and study endpoints were collected. Primary endpoints included rehabilitation-therapy rate, intervention time for rehabilitation, rehabilitation-related adverse events, and prognostic indicators such as (length of stay in hospital, length of stay in the ICU, and duration of mechanical ventilation). Secondary endpoints included incidence of deep vein thrombosis and hospital mortality. Kaplan-Meier curves were used to analyze cumulative discharge rates within 50 days. RESULTS: A total of 539 ICU patients were enrolled, with 245 in the conventional rehabilitation group and 294 in the enhanced rehabilitation group; 322 patients had an APACHE II score ≤ 15, while 217 patients had an APACHE II score > 15. Compared to the conventional rehabilitation group, the enhanced rehabilitation group demonstrated significantly higher rehabilitation-therapy rate [51.70% (152/294) vs. 11.43% (28/245)], earlier intervention time for rehabilitation [days: 2.00 (1.00, 3.00) vs. 4.00 (3.00, 7.00)]; shorter length of stay in hospital [days: 18.00 (12.00, 30.00) vs. 21.00 (13.00, 36.00)] and lower incidence of DVT [17.01% (50/294) vs. 24.08% (59/245)]. The differences were all statistically significant (all P < 0.05). There were no rehabilitation-related adverse events occurred in either group. Kaplan-Meier analysis demonstrated a significantly higher cumulative discharge rate within 50 days in the enhanced rehabilitation group compared to the conventional rehabilitation group [86.7% (255/294) vs. 82.9% (203/245); Log-Rank test: χ² = 4.262, P = 0.039]. Subgroup analysis showed that for patients with APACHE II score ≤ 15, the enhanced rehabilitation subgroup had higher rehabilitation-therapy rate [44.32% (78/176) vs. 6.16% (9/146), P < 0.05]. For patients with APACHE II score > 15, compared to the conventional rehabilitation group, the enhanced subgroup demonstrated higher rehabilitation-therapy rate [62.71% (74/118) vs. 19.19% (19/99), P < 0.05] and shorter length of stay in hospital [days: 20.50 (12.00, 31.25) vs. 26.00 (16.00, 43.00), P < 0.05]. CONCLUSIONS Enhanced rehabilitation therapy with medical care, rehabilitation and nursing care, improved rehabilitation-therapy rate, advanced time of rehabilitation treatment, reduced length of stay in hospital and incidence of deep vein thrombosis in critically ill patients, particularly benefited those with APACHE II score > 15. The enhanced rehabilitation was beneficial to the patient in the intensive care unit with safety and worth more investigation.
Humans ; Retrospective Studies ; Critical Illness/rehabilitation* ; Intensive Care Units ; Prognosis ; Length of Stay ; APACHE ; Historically Controlled Study ; Male ; Female ; Middle Aged ; Aged

Intra-abdominal candidiasis (IAC) is the most common invasive candidiasis, with a high incidence among critically ill patients, which can significantly increase medical costs and affect prognosis. In order to standardize the diagnosis and treatment of IAC in critically ill patients, experts in related fields were organized by the Peking University Critical Care Medicine (PKUCCM), Committee of Critical Care Medicine and Organ Support, China Association for Promotion of Health Science and Technology organized experts in related fields to initiate and form a working group. Expert writers drafted the consensus based on evidence-based medical evidence. A committee composed of critical care physicians, infectious disease physicians, surgeons, dermatologists specializing in antifungal fields, and clinical pharmacists discussed and revised the consensus draft through a standardized process, and finally formulated this consensus. This consensus contains a total of 20 core recommendations, mainly focusing on the epidemiology, high-risk factors, diagnostic techniques and methods (including traditional microbiological culture techniques, clinical risk prediction tools, serological tests, molecular biological tests, and histopathological examinations) of IAC, diagnostic criteria, stratified treatment strategies, antifungal drug selection, control the sources of infection, combined treatment, de-escalation strategies, drug treatment courses, prognosis, and special types of IAC. The aim is to provide expert guidance for the standardized clinical diagnosis and treatment of IAC in critically ill patients, with a view to improving prognosis of patients.
Humans ; Critical Illness ; Intraabdominal Infections/therapy* ; Antifungal Agents/therapeutic use* ; Consensus ; Candidiasis/drug therapy* ; Critical Care ; Candidiasis, Invasive/diagnosis*

OBJECTIVE: To develop and compare risk prediction models for in-hospital post-cardiac arrest brain injury (PCABI) in critically ill patients using nomograms and random forest algorithms, aiming to identify the optimal model for early identification of high-risk PCABI patients and providing evidence for precise treatment. METHODS: A retrospective cohort study was used to collect the first-time in-hospital cardiac arrest (IHCA) patients admitted to the intensive care unit (ICU) from 2008 to 2019 in the Medical Information Mart for Intensive Care-IV (MIMIC-IV) as the study population, and the patients' age, gender, body mass, health insurance utilization, first vital signs and laboratory tests within 24 hours of ICU admission, mechanical ventilation, and critical care scores were extracted. Independent influencing factors of PCABI were identified through univariate and multivariate Logistic regression analyses. The included patients were randomly divided into a training cohort and an internal validation cohort in a 7:3 ratio, and the PCABI risk prediction model was constructed by the nomogram and random forest algorithm, respectively, and the model was evaluated by receiver operator characteristic curve (ROC curve), the calibration curve, and the decision curve analysis (DCA), and after the better model was selected, 179 patients admitted to Tianjin Medical University General Hospital as the external validation cohort for external evaluation were collected by using the same inclusion and exclusion criteria. RESULTS: A total of 1 419 patients with without traumatic brain injury who had their first-time IHCA were enrolled, including 995 in the training cohort (including 176 PCABI and 819 non-PCABI) and 424 in the internal validation cohort (including 74 PCABI and 350 non-PCABI). Univariate and multivariate analysis showed that age, potassium, urea nitrogen, sequential organ failure assessment (SOFA), acute physiology and chronic health evaluation III (APACHE III), and mechanical ventilation were independent influences on the occurrence of PCABI in patients with IHCA (all P < 0.05). Combining the above variables, we constructed a nomogram model and a random forest model for comparison, and the results show that the nomogram model has better predictive efficacy than the random forest model [nomogram model: area under the ROC curve (AUC) of the training cohort = 0.776, with a 95% credible interval (95%CI) of 0.741-0.811; internal validation cohort AUC = 0.776, with a 95%CI of 0.718-0.833; random forest model: AUC = 0.720, with a 95%CI of 0.653-0.787], and they performed similarly in terms of calibration curves, but the nomogram performed better in terms of decision curve analysis (DCA); at the same time, the nomogram model was robust in terms of external validation cohort (external validation cohort AUC = 0.784, 95%CI was 0.692-0.876). CONCLUSIONS A nomogram risk prediction model for the occurrence of PCABI in critically ill patients was successfully constructed, which performs better than the random forest model, helps clinicians to identify the risk of PCABI in critically ill patients at an early stage and provides a theoretical basis for early intervention.
Humans ; Critical Illness ; Retrospective Studies ; Heart Arrest/complications* ; Nomograms ; Brain Injuries/etiology* ; Intensive Care Units ; Algorithms ; Male ; Female ; Middle Aged ; ROC Curve ; Risk Factors ; Risk Assessment ; Logistic Models ; Aged

Extracorporeal membrane oxygenation (ECMO) technology is an important life support method for critically ill patients. A large number of studies have found that ECMO can change the pharmacokinetic (PK) parameters of critically ill patients, thereby affecting the drug effect in vivo. However, there is still a lack of recommendations for the adjustment of commonly used drugs during ECMO support in China, and the selection or dosage adjustment of drugs during ECMO support is not clear. Therefore, a multidisciplinary group of domestic experts in clinical pharmacy and critical care medicine was established by Critical Care Medicine Committee of China International Exchange and Promotive Association for Medical and Health Care, and Hospital Pharmacy Committee of China Pharmaceutical Association, to develop the Expert consensus on drug adjustment during extracorporeal membrane oxygenation support (2025). Eight clinical issues of drug adjustment during ECMO support were discussed in this consensus: (1) Why does the patient's demand for drug dosage change during ECMO support? (2) What factors are related to the degree of drug loss during ECMO support? (3) Considering the features of drugs, which types of drugs may need to be adjusted during ECMO support? (4) How to adjust the dosage when using antibacterial drugs during ECMO support? (5) How to adjust antifungal drugs during ECMO support? (6) Does ECMO support change patients' dosage requirements for antiviral drugs? (7) How to adjust sedative and analgesic drugs during ECMO support? (8) Does ECMO support affect the dosage requirement of vasoactive agents? Eighteen consensus are elaborated based on the latest clinical evidence, aiming to provide recommendations for drug adjustment in critically ill patients receiving ECMO support to ensure the safety and effectiveness of medication.
Humans ; Critical Illness ; Extracorporeal Membrane Oxygenation ; Consensus ; Candidiasis/drug therapy* ; Intraabdominal Infections/therapy*