INTRODUCTION: Out-of-hospital-cardiac-arrest (OHCA) is a major public health challenge and post-return-of-spontaneous-circulation (ROSC) goals have shifted from just survival to survival with intact neurology. Although post-ROSC care is crucial for survival with intact neurology, there are insufficient well-established protocols for post-resuscitation care. We aimed to evaluate post-resuscitation care in the emergency department (ED) of adult (aged ≥16 years) OHCA patients with sustained ROSC and its associated neurologically intact survival. METHODS: A retrospective review of electronic medical records was conducted for OHCA patients with sustained ROSC at the ED. Data including demographics, pre-hospital resuscitation, ED resuscitation, post-resuscitation care and eventual outcomes were analysed. RESULTS: Among 921 OHCA patients, 85 (9.2%) had sustained ROSC at the ED. Nineteen patients (19/85, 22.4%) survived, with 13 (13/85, 15.3%) having intact neurology at discharge. Electrocardiogram and chest X-ray were performed in all OHCA patients, whereas computed tomography (CT) was performed inconsistently, with CT brain being most common (74/85, 87.1%), while CT pulmonary angiogram (6/85, 7.1%), abdomen and pelvis (4/85, 4.7%) and aortogram (2/85, 2.4%) were done infrequently. Only four patients (4.7%) had all five neuroprotective goals of normoxia, normocarbia, normotension, normothermia and normoglycaemia achieved in the ED. The proportion of all five neuroprotective goals being met was significantly higher ( P = 0.01) among those with neurologically intact survival (3/13, 23.1%) than those without (1/72, 1.4%). CONCLUSION Post-resuscitation care at the ED showed great variability, indicating gaps between recommended guidelines and clinical practice. Good quality post-resuscitation care, centred around neuroprotection goals, must be initiated promptly to achieve meaningful survival with intact neurology.
Humans ; Out-of-Hospital Cardiac Arrest/mortality* ; Retrospective Studies ; Male ; Female ; Middle Aged ; Emergency Service, Hospital ; Cardiopulmonary Resuscitation/methods* ; Return of Spontaneous Circulation ; Aged ; Adult ; Treatment Outcome ; Electrocardiography ; Tomography, X-Ray Computed ; Aged, 80 and over

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 investigate the relationship between the trajectories of serum potassium changes after intensive care unit (ICU) admission and 30-day death risk in patients with sepsis. METHODS: A retrospective cohort study was conducted, including adult patients with sepsis admitted to the comprehensive ICU, medical intensive care unit (MICU) and emergency intensive care unit (EICU) of Guizhou Medical University Affiliated Hospital from January 2020 to January 2024. The patients who had a minimum of 5 days' hospitalisation in the ICU and who had at least 7 consecutive days of the serum potassium measurements were classified into five trajectories groups according to group-based trajectory modelling (GBTM) using SAS software. This was based on tendency changes in serum potassium levels in patients after admission to the ICU, which was categorized as follows: slowly increased from a low level group, slowly increased from a medium level of normal range group, slowly decreased from a medium level of normal range group, slowly decreased from a high level group, and slowly increased from a high level of normal range group. The patient's gender, age, medical history, and white blood cell count (WBC), platelet count (PLT), procalcitonin (PCT), activated partial thromboplastin time (APTT), prothrombin time (PT), blood sodium, and serum creatinine (SCr) at the time of admission to the ICU were collected. At the same time, the patient's worst sequential organ failure assessment (SOFA) score within 24 hours of admission to the ICU, length of ICU stay, and 30-day outcome were record. The differences in clinical data among different groups of patients were compared. The 30-day cumulative survival rates of the various serum potassium trajectories were plotted using Kaplan-Meier survival curves, the groups were then compared using the Log-Rank test. A multivariate Cox proportional risk regression analysis was developed to evaluate the independent effect of serum potassium trajectory on 30-day death risk. RESULTS: Finally, 342 ICU sepsis patients were enrolled, of which 42 patients in the slowly increased from a low level group (12.28%), 127 patients in the slowly increased from a medium level of normal range group (37.14%), 118 patients in the slowly decreased from a medium level of normal range group (34.50%), 28 patients in the slowly decreased from a high level group (8.19%), and 27 patients in the slowly increased from a high level of normal range group (7.89%). Except for age and APTT differences, there were no statistically significant differences in other clinical characteristics among the patients in the different serum potassium trajectories groups. Kaplan-Meier survival curves showed that there was statistically significant difference in the 30-day cumulative survival rate among the patients in the different serum potassium trajectories groups (Log-Rank test: χ² = 14.696, P = 0.005), with the lowest in the slowly increased from a high level of normal range group (39.3%). Multivariate Cox proportional risk regression analysis showed that the patients with the serum potassium trajectory of slowly increased from a high level of normal range had the highest 30-day death risk [hazard ratio (HR) = 2.341, 95% confidence interval (95%CI) was 1.049-5.226, P = 0.038]. This association persisted after adjustment for variables such as gender, age, medical history, SOFA score, WBC, PLT, PCT, APTT, PT, blood sodium, and SCr (HR = 3.058, 95%CI was 1.249-7.488, P = 0.014). CONCLUSION Compared with the patients whose serum potassium fluctuated within the normal range, the sepsis patients in the ICU with a serum potassium trajectory that slowly increased from a high level of normal range had a significantly higher 30-day death risk.
Humans ; Retrospective Studies ; Intensive Care Units ; Sepsis/blood* ; Potassium/blood* ; Male ; Female ; Middle Aged ; Aged ; Risk Factors ; Hospital Mortality ; Prognosis

OBJECTIVE: To evaluate the impact of critical care warning platform (CWP) on clinical outcomes of patients transferred from internal medical ward to intensive care unit (ICU) based on real-world data. METHODS: A retrospective cohort study was conducted. The patients transferred from internal medical ward to ICU of Zhongda Hospital, Southeast University, between January 2022 and October 2024, were enrolled. They were divided into critical care warning group and conventional treatment group based on whether they were connected to the CWP. The patients in the critical care warning group were connected to the CWP, which collected real-time vital signs and treatment data. The platform automatically calculated severity scores, generated individualized risk assessments, and triggered warning alerts, allowing clinicians to adjust treatment plans accordingly. The patients in the conventional treatment group were not connected to the CWP and relied on conventional clinical judgment and nursing measures for treatment management. Baseline characteristics [gender, age, body mass index (BMI), admission type, severity score of illness, underlying diseases, and disease type at ICU admission], primary clinical outcome (in-hospital mortality), and secondary clinical outcomes [ICU mortality, length of ICU stay, total length of hospital stay, and mechanical ventilation and continuous renal replacement therapy (CRRT) status] were collected. Multivariate Logistic regression was used to analyze the impact of CWP on in-hospital death, and subgroup analyses were performed based on different patient characteristics. RESULTS: A total of 1 281 patients were enrolled, with 768 in the critical care warning group and 513 in the conventional treatment group. Compared with the conventional treatment group, the proportion of patients in the critical care warning group with underlying diseases of diabetes and malignancy and transferred to ICU due to sepsis was lowered, however, there were no statistically significant differences in other baseline characteristics between the two groups. Regarding the primary clinical outcome, the in-hospital mortality in the critical care warning group was significantly lower than that in the conventional treatment group [17.6% (135/768) vs. 25.7% (132/513), P < 0.01]. For secondary clinical outcomes, compared with the conventional treatment group, the patients in the critical care warning group had significantly fewer days of mechanical ventilation within 28 days [days: 2 (1, 6) vs. 2 (1, 8), P < 0.05], significantly shorter length of ICU stay [days: 3 (2, 8) vs. 4 (2, 10), P < 0.01], and significantly lower ICU mortality [15.1% (116/768) vs. 21.4% (110/513), P < 0.01]. Multivariate Logistic regression analysis showed that, after adjusting for age and underlying diseases, the use of CWP was significantly associated with a reduction of in-hospital mortality among patients transferred from internal medical ward to ICU [odds ratio (OR) = 0.670, 95% confidence interval (95%CI) was 0.502-0.894, P = 0.006]. Further subgroup analysis revealed that, among patients transferred to ICU due to sepsis, the use of CWP significantly reduced in-hospital mortality (OR = 0.514, 95%CI was 0.367-0.722, P < 0.001). In patients aged ≥ 70 years old (OR = 0.587, 95%CI was 0.415-0.831, P = 0.003) and those with underlying diseases of malignancy (OR = 0.124, 95%CI was 0.046-0.330, P < 0.001), CWP also showed significant protective effects on in-hospital prognosis. CONCLUSION The use of CWP is significantly associated with a reduction in in-hospital mortality among patients transferred from internal medical ward to ICU, demonstrating its potential in assessing the deterioration of hospitalized patients.
Humans ; Intensive Care Units ; Retrospective Studies ; Hospital Mortality ; Prognosis ; Critical Care ; Male ; Female ; Patient Transfer ; Middle Aged ; Aged ; Cohort Studies

OBJECTIVE: To assess the relationship between blood pressure reactivity index (BPRI) and in-hospital mortality risk in patients with sepsis-associated acute kidney injury (SA-AKI). METHODS: A retrospective cohort study was conducted to collect data from patients admitted to the intensive care unit (ICU) and clinically diagnosed with SA-AKI between 2008 and 2019 in the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database in the United States. The collected data included demographic characteristics, comorbidities, vital signs, laboratory parameters, sequential organ failure assessment (SOFA) and simplified acute physiology scoreII(SAPSII) within 48 hours of SA-AKI diagnosis, stages of AKI, treatment regimens, mean BPRI during the first and second 24 hours (BPRI_0_24, BPRI_24_48), and outcome measures including primary outcome (in-hospital mortality) and secondary outcomes (ICU length of stay and total hospital length of stay). Variables with statistical significance in univariate analysis were included in LASSO regression analysis for variable selection, and the selected variables were subsequently incorporated into multivariate Logistic regression analysis to identify independent predictors associated with in-hospital mortality in SA-AKI patients. Restricted cubic spline (RCS) analysis was employed to examine whether there was a linear relationship between BPRI within 48 hours and in-hospital mortality in SA-AKI patients. Basic prediction models were constructed based on the independent predictors identified through multivariate Logistic regression analysis, and receiver operator characteristic curve (ROC curve) was plotted to evaluate the predictive performance of each basic prediction model before and after incorporating BPRI. RESULTS: A total of 3 517 SA-AKI patients admitted to the ICU were included, of whom 826 died during hospitalization and 2 691 survived. The BPRI values within 48 hours of SA-AKI diagnosis were significantly lower in the death group compared with the survival group [BPRI_0_24: 4.53 (1.81, 8.11) vs. 17.39 (5.16, 52.43); BPRI_24_48: 4.76 (2.42, 12.44) vs. 32.23 (8.85, 85.52), all P < 0.05]. LASSO regression analysis identified 20 variables with non-zero coefficients that were included in the multivariate Logistic regression analysis. The results showed that respiratory rate, temperature, pulse oxygen saturation (SpO₂), white blood cell count (WBC), hematocrit (HCT), activated partial thromboplastin time (APTT), lactate, oxygenation index, SOFA score, fluid balance (FB), BPRI_0_24, and BPRI_24_48 were all independent predictors for in-hospital mortality in SA-AKI patients (all P < 0.05). RCS analysis revealed that both BPRI showed "L"-shaped non-linear relationships with the risk of in-hospital mortality in SA-AKI patients. When BPRI_0_24 ≤ 14.47 or BPRI_24_48 ≤ 24.21, the risk of in-hospital mortality in SA-AKI increased as BPRI values decreased. Three basic prediction models were constructed based on the identified independent predictors: Model 1 (physiological indicator model) included respiratory rate, temperature, SpO₂, and oxygenation index; Model 2 (laboratory indicator model) included WBC, HCT, APTT, and lactate; Model 3 (scoring indicator model) included SOFA score and FB. ROC curve analysis showed that the predictive performance of the basic models ranked from high to low as follows: Model 3, Model 2, and Model 1, with area under the curve (AUC) values of 0.755, 0.661, and 0.655, respectively. The incorporation of BPRI indicators resulted in significant improvement in the discriminative ability of each model (all P < 0.05), with AUC values increasing to 0.832 for Model 3+BPRI, 0.805 for Model 2+BPRI, and 0.808 for Model 1+BPRI. CONCLUSIONS BPRI is an independent predictor factor for in-hospital mortality in SA-AKI patients. Incorporating BPRI into the prediction model for in-hospital mortality risk in SA-AKI can significantly improve its predictive capability.
Humans ; Acute Kidney Injury/mortality* ; Sepsis/complications* ; Retrospective Studies ; Hospital Mortality ; Prognosis ; Blood Pressure ; Intensive Care Units ; Male ; Female ; Length of Stay ; Middle Aged ; Aged ; Adult ; Logistic Models

OBJECTIVE: To construct a longitudinal trajectory model of arterial oxygen tension (PaO₂) within 24 hours after cardiac arrest (CA). METHODS: A retrospective cohort study was conducted. CA patients admitted to the ICU from 2014 to 2015 were selected from the eICU Collaborative Research Database (eICU-CRD). Data about patients' demographic characteristics, history of comorbidities, laboratory test indicators within 24 hours of intensive care unit (ICU) admission [including all PaO₂ data and arterial carbon dioxide tension (PaCO₂)], vasopressor use, and clinical outcomes were extracted from the database. The primary outcome variable was all-cause in-hospital mortality. Group-based trajectory model (GBTM) were built based on the changes in PaO₂ within 24 hours of ICU admission, and patients were grouped according to their initial static PaO₂ values upon ICU admission. Multivariable adjusted Poisson regression analysis was used to compare the in-hospital mortality risk among patients in different PaO₂ dynamic trajectory groups. Sensitivity analyses were performed using multivariable logistic regression and multivariable adjusted Poisson regression without imputation of missing values. RESULTS: A total of 3 866 CA patients were included. Three GBTM trajectory groups were identified based on PaO₂ changes within 24 hours of ICU admission: Group-1 (low level first increased then decreased, 148 cases), Group-2 (sustained low level, 3 040 cases), and Group-3 (first high level then decreased, 678 cases). Significant differences were found among the three groups in age, body weight, maximum serum potassium, maximum PaCO₂, minimum hemoglobin (Hb), vasopressor use, total hospitalization time, ICU stay, and hospital mortality. After incorporating variables with significant differences into the multivariable adjusted Poisson regression model, results showed that compared to Group-2 patients, patients in Group-1 and Group-3 had an increased risk of all-cause in-hospital mortality [Group-1 adjusted relative risk (aRR) = 1.20, 95% confidence interval (95%CI) was 1.02-1.41; Group-3 aRR = 1.11, 95%CI was 1.01-1.24]. Based on initial static PaO₂ values at ICU admission, patients were divided into four groups: PaO₂ < 100 mmHg (1 mmHg = 0.133 kPa; 1 217 cases), PaO₂ 100-200 mmHg (569 cases), PaO₂ 201-300 mmHg (547 cases), and PaO₂ > 300 mmHg (1 082 cases). Multivariable adjusted Poisson regression analysis indicated a significant upward trend in aRR for the latter three groups compared to the PaO₂ < 100 mmHg group. Sensitivity analyses revealed that compared to Group-2, patients in Group-1 and Group-3 had a significantly increased risk of all-cause in-hospital mortality (both P < 0.05). CONCLUSIONS Within 24 hours after return of spontaneous circulation in CA patients, PaO₂ exhibits different dynamic trajectories, and patients with hyperoxia have an increased risk of in-hospital mortality.
Humans ; Retrospective Studies ; Hospital Mortality ; Heart Arrest/blood* ; Prognosis ; Oxygen/blood* ; Intensive Care Units ; Cardiopulmonary Resuscitation ; Male ; Female ; Middle Aged

OBJECTIVE: To explore the timing of weaning extracorporeal membrane oxygenation (ECMO) and analyze the risk factors that affect survival outcomes before weaning. METHODS: A retrospective case-control study was conducted. Patients who received ECMO treatment and were weaned according to physicians' orders at the Second Xiangya Hospital of Central South University from January 2020 to June 2024 were enrolled as the study subjects. The general information, underlying diseases, indications and processes of ECMO, vital signs and arterial blood gas analysis 1 hour before weaning test, and biochemical indicators 24 hours before weaning test were collected through the hospital electronic medical record system. The primary outcome measure was the hospital mortality. The variables with P < 0.1 in univariate analysis and correlation analysis were included into binary Logistic regression analysis to identify risk factors. A nomogram model was constructed to predict the risk of weaning death in patients with ECMO, and receiver operator characteristic curve (ROC curve) and calibration curve were drawn to evaluate the model. Decision curve analysis (DCA) was used to evaluate the clinical net benefit rate of the model. RESULTS: A total of 32 ECMO patients were included, among whom 10 received veno-arterial ECMO (VA-ECMO) and 22 received veno-venous ECMO (VV-ECMO). During the hospitalization period, 23 patients survived, while 9 died. The time from mechanical ventilation to ECMO activation in the death group was significantly longer than that in the survival group, and the time from ECMO cessation to discharge was significantly shorter than that in the survival group. The levels of diastolic blood pressure (DBP) and albumin (Alb) before weaning were significantly lower than those in the survival group, and the level of procalcitonin (PCT) was significantly higher than that in the survival group (all P < 0.05). Spearman correlation analysis showed that DBP, PCT, Alb, and thrombin time (TT) were correlated with the weaning outcomes of ECMO patients (r values were -0.450, 0.373, -0.376, -0.346, all P < 0.1). Binary Logistic regression analysis showed that the final indicators entering the regression equation included DBP [odds ratio (OR) = 0.864, 95% confidence interval (95%CI) was 0.756-0.982], PCT (OR = 1.157, 95%CI was 0.679-1.973), and TT (OR = 0.852, 95%CI was 0.693-1.049), and a nomogram model was constructed to predict the weaning outcomes of ECMO patients. ROC curve analysis showed that the area under the curve (AUC) of the nomogram model for predicting the weaning outcome of ECMO patients was 0.831, with a sensitivity of 77.8% and a specificity of 65.2%. Its predictive value was better than that of single indicators DBP, PCT, and TT (AUC of 0.787, 0.739, and 0.722, respectively). The calibration curve showed that the prediction probability of the model was in good consistency with the actual observed results, the Hosmer-Lemeshow goodness of fit test showed that, χ ² = 8.3521, P = 0.400, indicating that the model fits well. DCA showed that across risk threshold of 0-0.8, the net benefit rate was greater than 0, which was significantly better than that of single indicator. CONCLUSIONS The nomogram model constructed with DBP, PCT, and TT has certain predictive value for the weaning outcomes of ECMO patients and can be used as a screening indicator for ECMO weaning timing.
Humans ; Extracorporeal Membrane Oxygenation ; Retrospective Studies ; Risk Factors ; Case-Control Studies ; Hospital Mortality ; Male ; Female ; Nomograms ; Logistic Models ; ROC Curve ; Middle Aged ; Adult ; Ventilator Weaning ; Time Factors

OBJECTIVE: To investigate the correlation between nucleated red blood cell (NRBC) level on the first day of intensive care unit (ICU) admission and 28-day mortality in adult septic patients, and to evaluate the value of NRBC as an independent predictor of death. METHODS: Single-cell transcriptomic analysis was performed using the GSE167363 dataset from the Gene Expression Omnibus (including 2 healthy controls, 3 surviving septic patients, and 2 non-surviving septic patients). A retrospective clinical analysis was conducted using the America Medical Information Mart for Intensive Care-IV (MIMIC-IV) database, including adult patients (≥ 18 years) with first-time admission who met the Sepsis-3.0 criteria, excluding those without NRBC testing on the first ICU day. The demographic information, vital signs, laboratory test indicators, disease severity score and survival data on the first day of admission were collected. The restricted cubic spline (RCS) curve was used to determine the optimal cut-off value of NRBC for predicting 28-day mortality in patients. Patients were divided into low-risk and high-risk groups based on this cut-off value for intergroup comparison, with Kaplan-Meier survival curve analysis conducted. Independent risk factors for 28-day mortality were analyzed using Logistic regression and Cox regression analysis, followed by the construction of regression models. RESULTS: NRBC were detected in the peripheral blood of septic patients by single-cell transcriptomic. A total of 1 291 sepsis patients were included in the clinical analysis, with 576 deaths within 28 days, corresponding to a 28-day mortality of 44.6%. RCS curve analysis showed a nonlinear relationship between the first-day NRBC level and the 28-day mortality. When NRBC ≥ 1%, the 28-day mortality of patients increased significantly. Compared to the low-risk group (NRBC < 1%), the high-risk group (NRBC ≥ 1%) had significantly higher respiratory rate, heart rate, sequential organ failure assessment (SOFA), and simplified acute physiology score II (SAPSII), and significantly lower hematocrit and platelet count. The high-risk group also had a significantly higher 28-day mortality [49.8% (410/824) vs. 35.5% (166/467), P < 0.05], and shorter median survival time (days: 29.8 vs. 208.6, P < 0.05). Kaplan-Meier survival curve showed that compared with the low-risk group, the survival time of high-risk group was significantly shortened (Log-rank test: χ ² = 25.1, P < 0.001). After adjusting for potential confounding factors including body mass, temperature, heart rate, respiratory rate, mean arterial pressure, serum creatinine, pulse oximetry saturation, hemoglobin, hematocrit, Na⁺, K⁺, platelet count, and SOFA score, multivariate regression analysis confirmed that NRBC ≥ 1% was an independent risk factor for 28-day mortality [Logistic regression: odds ratio (OR) = 1.464, 95% confidence interval (95%CI) was 1.126-1.902, P = 0.004; Cox regression: hazard ratio (HR) = 1.268, 95%CI was 1.050-1.531, P = 0.013]. CONCLUSIONS NRBC ≥ 1% on the first day of ICU admission is an independent risk factor for 28-day mortality in septic patients and can serve as a practical indicator for early prognostic assessment.
Humans ; Sepsis/blood* ; Intensive Care Units ; Risk Factors ; Retrospective Studies ; Prognosis ; Male ; Female ; Hospital Mortality ; Middle Aged ; Aged

OBJECTIVE: To explore the correlation between albumin (Alb) combined with diuretic treatment and the mortality risk of septic patients with pre-existing congestive heart failure based on the United States Critical Care Medical Information Database-IV (MIMIC-IV), and to conduct the external validation. METHODS: A retrospective cohort study was conducted. The clinical data of septic patients with pre-existing congestive heart failure admitted to the intensive care unit (ICU) from 2008 to 2019 in the MIMIC-IV 2.0 were extracted, including demographic characteristics, comorbidities, laboratory indicators on the first day of ICU admission, severity of illness, treatment measures, etc. For external validation, clinical data were collected from septic patients with pre-existing congestive heart failure admitted to the ICU of the Second People's Hospital of Shenzhen from October 2022 to December 2023. The patients were divided into Alb alone group and Alb combined with diuretic group. The ICU mortality was defined as the primary outcome event, and the 30-day and 60-day mortality were defined as the secondary outcomes. Multivariate Cox proportional hazard regression analysis was conducted to investigate the relationship between Alb combined with diuretic treatment and the mortality risk of ICU and 30 days in septic patients with pre-existing congestive heart failure, and subgroup analysis was performed. Kaplan-Meier survival curve was plotted to compared the 60-day cumulative survival rate between the Alb alone group and Alb combined with diuretic group. RESULTS: (1) Analysis results of data from MIMIC-IV: a total 1 754 patients were enrolled, of which 378 in the Alb alone group, and 1 376 in the Alb combined with diuretic group. Compared with the Alb alone group, the patients in the Alb combined with diuretic group had significantly lower ICU, 30-day, and 60-day mortality [ICU mortality: 19.11% (263/1 376) vs. 30.42% (115/378), 30-day mortality: 18.90% (260/1 376) vs. 32.54% (123/378), 60-day mortality: 24.49% (337/1 376) vs. 39.15% (148/378), all P < 0.05]. Based on the multivariate Cox proportional hazard regression adjusted models considering demographic characteristics, comorbidities, laboratory indicators, severity of illness, and treatment measures, it was shown that the use of Alb combined with diuretic was significantly associated with a reduced risk death of ICU and 30 days [ICU mortality risk: hazard ratio (HR) = 0.597, 95% confidence interval (95%CI) was 0.460-0.774, P < 0.001; 30-day mortality risk: HR = 0.557, 95%CI was 0.433-0.716, P < 0.001]. Subgroup analysis revealed that after adjusting for variables, regardless of gender, age, and whether or not patients had comorbidities such as hypertension, diabetes, severe liver disease, acute renal insufficiency, and sequential organ failure assessment (SOFA) score, the ICU mortality risk was significantly reduced in patients treated with Alb combined with diuretic (all HR < 1, P < 0.05), with no interaction observed (all P > 0.05). Kaplan-Meier survival curve showed the 60-day cumulative survival rate of patients in the Alb combined with diuretic group was significantly higher than that in the Alb alone group (Log-rank test: χ ² = 49.62, P < 0.05). (2) External validation: a total of 385 patients were enrolled, of which 144 in the Alb alone group, and 241 in the Alb combined with diuretic group. Compared with the Alb alone group, the patients of the Alb combined with diuretic group had significantly lower ICU, 30-day, and 60-day mortality [ICU mortality: 19.92% (48/241) vs. 31.25% (45/144), 30-day mortality: 19.09% (46/241) vs. 28.47% (41/144), 60-day mortality: 24.07% (58/241) vs. 34.03% (49/144), all P < 0.05]. The results of multivariate Cox proportional hazard regression analysis, subgroup analysis, and Kaplan-Meier survival curve analysis were consistent with the data analysis of the MIMIC-IV database. CONCLUSIONS Combination therapy of Alb and diuretic was associated with reduced mortality risk in septic patients with pre-existing congestive heart failure.
Humans ; Heart Failure/mortality* ; Retrospective Studies ; Sepsis/drug therapy* ; Intensive Care Units ; Diuretics/therapeutic use* ; Male ; Female ; Aged ; Middle Aged ; Proportional Hazards Models ; Hospital Mortality

OBJECTIVE: To explore the association between blood glucose trajectories within 7 days of intensive care unit (ICU) admission and mortality in patients with sepsis-associated acute respiratory distress syndrome (ARDS). METHODS: Based on the MIMIC-IV database, sepsis-associated ARDS patients with daily blood glucose monitoring data within 7 days of ICU admission were selected. Blood glucose trajectories were analyzed using group-based trajectory modeling (GBTM), and the optimal number of groups was determined based on the minimum Akaike information criterion (AIC), Bayesian information criterion (BIC), average posterior probability (AvePP), odds of correct classification (OCC), and proportion of group membership (Prop). Baseline characteristics including demographics, comorbidities, severity scores, vital signs, laboratory indicators within the first 24 hours of ICU admission, and treatments were collected. Kaplan-Meier survival curves were used to compare 28-day and 1-year survival across trajectory groups. Multivariate Logistic regression was performed to evaluate the associations between glucose trajectory groups and in-hospital mortality, ICU mortality. The incidence of hypoglycemia within 7 days in the ICU was analyzed among different groups. RESULTS: A total of 3 869 patients with sepsis-associated ARDS were included, with a median age of 63.52 (52.13, 73.54) years; 59.6% (2 304/3 869) were male. Based on glucose levels within 7 days, patients were categorized into three groups: persistent hyperglycemia group (glucose maintained at 10.6-13.1 mmol/L, n = 894), moderate glucose group (7.8-8.9 mmol/L, n = 1 452), and low-normal glucose group (6.1-7.0 mmol/L, n = 1 523). There were statistically significant differences in 28-day mortality and 1-year mortality among low-normal glucose group, moderate glucose group, and persistent hyperglycemia group [28-day mortality: 11.42% (174/1 523), 19.83% (288/1 452), 25.50% (228/894), χ ² = 82.545, P < 0.001; 1-year mortality: 23.31% (355/1 523), 33.75% (490/1 452), 39.49% (353/894), χ ² = 77.376, P < 0.001]. Kaplan-Meier analysis showed that higher glucose trajectories were associated with significantly lower 28-day and 1-year cumulative survival rates (Log-rank test: χ ² were 83.221 and 85.022, both P < 0.001). There were statistically significant differences in in-hospital mortality and ICU mortality among the low-normal glucose group, moderate glucose group, and persistent hyperglycemia group [in-hospital mortality: 9.65% (147/1 523), 19.70% (286/1 452), 24.50% (219/894), χ ² = 102.020, P < 0.001; ICU mortality: 7.22% (110/1 523), 16.05% (233/1 452), 20.13% (180/894), χ ² = 93.050, P < 0.001]. Logistic regression confirmed that, using the persistent hyperglycemia group as the reference, the low-normal glucose group had significantly lower risks of in-hospital mortality and ICU mortality after multiple factor adjustment. Although the moderate glucose group showed a trend toward lower mortality, the differences were not statistically significant. Using the moderate glucose group as a reference, the low-normal glucose group had 43.1% lower in-hospital mortality [odds ratio (OR) = 0.569, 95% confidence interval (95%CI) was 0.445-0.726, P < 0.001] and 42.0% lower ICU mortality (OR = 0.580, 95%CI was 0.439-0.762, P < 0.001). There was no statistically significant difference in the incidence of hypoglycemia within 7 days of ICU admission among low-normal glucose group, moderate glucose group, and persistent hyperglycemia group [2.82% (43/1 523), 2.69% (39/1 452), 3.02% (27/894), χ ² = 0.226, P = 0.893]. CONCLUSIONS Blood glucose trajectories during ICU stay are closely associated with prognosis in patients with sepsis-associated ARDS. Persistent hyperglycemia (10.6-13.1 mmol/L) is linked to significantly higher short- and long-term mortality.
Humans ; Respiratory Distress Syndrome/etiology* ; Sepsis/blood* ; Intensive Care Units ; Male ; Female ; Middle Aged ; Blood Glucose/metabolism* ; Hospital Mortality ; Aged