A trauma-related survival predictive model of acute respiratory distress syndrome

Association between red blood cell distribution width and the all-cause mortality of patients with aortic stenosis: A retrospective study

BACKGROUND

It is essential to assess the risk stratification of patients with aortic stenosis (AS).

OBJECTIVE

To clarify the predictive value of red blood cell distribution width (RDW) in AS patients using a large cohort from the MIMIC-IV database.

METHODS

Restricted cubic spline, the Kaplan-Meier method, and logistic and Cox regression analyses were used to explore the association between RDW and all-cause mortality in AS patients. Multivariate adjustments, propensity score matching and weighting, and subgroup analysis were conducted to exclude confounding factors. Receiver operating characteristic (ROC) and decision curve analysis (DCA) curves were drawn to evaluate the predictive performance of RDW.

RESULTS

1,148 patients with AS were included. Their death risks gradually increased with the elevation of RDW. Multivariate-adjusted 90-day (OR: 2.12; HR: 1.90; p = 0.001) and 1-year (OR: 2.07; HR: 1.97; p < 0.001) all-cause mortalities were significantly higher in patients with RDW≥14.7 %, which remained robust after propensity score matching and subgroup analysis. For AS patients with high RDW, those < 75 years old had higher death risks than those ≥ 75 years old. The area under the ROC curve of RDW were 0.741 and 0.75 at 90-day and 1-year follow-ups, respectively, exhibiting comparable performance to acute physiology score III and outperforming other critical illness scores in predicting the prognosis of AS patients. DCA curves also illustrated that RDW had a wide range of net benefits.

CONCLUSIONS

High RDW was independently associated with increased 90-day and 1-year all-cause mortalities of AS patients, with strong predictive capability of prognosis.

Association between BUN/creatinine ratio and the risk of in-hospital mortality in patients with trauma-related acute respiratory distress syndrome: a single-centre retrospective cohort from the MIMIC database

OBJECTIVE

Recent studies have shown that blood urea nitrogen to creatinine (BUN/Cr) ratio might be an effective marker for the prognosis of patients with respiratory diseases. Herein, we aimed to assess the association between BUN/Cr ratio and the risk of in-hospital mortality in patients with trauma-related acute respiratory distress syndrome (ARDS).

DESIGN

A retrospective cohort study.

SETTING AND PARTICIPANTS

1034 patients were extracted from the Medical Information Mart for Intensive Care-III (MIMIC-III) database.

PRIMARY AND SECONDARY OUTCOME MEASURES

The primary outcome of the study was in-hospital mortality, defined by the vital status at the time of hospital discharge (ie, survivors and non-survivors).

RESULTS

Of the total patients, 191 (18.5%) died in hospital. The median follow-up duration was 16.0 (8.3-26.6) days. The results showed that high level of BUN/Cr ratio was significantly associated with an increased risk of in-hospital mortality (15.54-21.43: HR=2.00, 95% CI: (1.18 to 3.38); >21.43: HR=1.76, 95% CI: (1.04 to 2.99)) of patients with trauma-related ARDS. In patients with trauma-related ARDS that aged ≥65 years old, male and female, Onychomycosis Severity Index (OSI)>98, Revised Trauma Score (RTS)>11, Simplified Acute Physiology Score II (SAPS-II)>37 and sequential organ failure assessment (SOFA) scores≤7, BUN/Cr ratio was also related to the increased risk of in-hospital mortality (all p<0.05). The predictive performance of BUN/Cr ratio for in-hospital mortality was superior to BUN or Cr, respectively, with the area under the curve of receiver operator characteristic curve at 0.6, and that association was observed in age, gender, OSI, RTS, SAPS-II and SOFA score subgroups.

CONCLUSION

BUN/Cr ratio may be a potential biomarker for the risk of in-hospital mortality of trauma-related ARDS, which may help the clinicians to identify high-risk individuals and to implement clinical interventions.

Association of serum total bilirubin and potential predictors with mortality in acute respiratory failure: A retrospective cohort study

BACKGROUND

Total serum bilirubin (TBIL) levels are a risk factor in critically ill patients. However, the relationship between the dynamics of TBIL and the prognosis of acute respiratory failure (ARF) patients is unclear.

OBJECTIVES

This study aimed to investigate the impact of different levels of TBIL during hospitalization on mortality in ARF patients.

METHODS

This study used a retrospective cohort study. We extracted information on ARF patients from the Medical Information Bank for Intensive Care (MIMIC)-III (version 1.4). We used propensity score matching (PSM) to adjust for the level of potential baseline-level differences between groups. Cox regression was used to analyze mortality risk factors in patients with ARF. Subgroup analysis was used to explore special populations.

RESULTS

2673 patients were included in the study, and 19.7% developed hyperbilirubinemia (TBIL ≥ 2 mg/dL) during their hospitalization. After PSM, multivariate Cox regression showed a 50% and 135% increased risk of death for a maximum value of TBIL ≥ 5 mg/dL and minimum value of TBIL ≥ 2 mg/dL during hospitalization, respectively, compared to the control population. In addition, age ≥ 65 years, previous comorbid malignancies, respiratory rate ≥ 22 beats/min, SpO2 ≥ 95, BUN ≥ 20 mg/dL, lactate ≥ 5 mmol/L, platelet < 100 * 10 ^ 9/L were independent risk factors for 1-year mortality in ARF patients. Subgroup analysis showed that high bilirubin had a greater effect on patients aged less than 65 years (P for interaction < 0.05).

CONCLUSIONS

Hyper TBIL (TBIL max ≥ 5 mg/dL or TBIL min ≥ 2 mg/dL) was an independent risk factor for 1-year mortality in patients with ARF. This study suggests that clinicians should be aware of TBIL levels and intervene early in these patients.

Development and validation of a clinical risk model to predict the hospital mortality in ventilated patients with acute respiratory distress syndrome: a population-based study

Background

Large variability in mortality exists in patients of acute respiratory distress syndrome (ARDS), especially those with invasive ventilation. The aim of this study was to develop a model to predict risk of in-hospital death in ventilated ARDS patients.

Methods

Ventilated patients with ARDS from two public databases (MIMIC-III and eICU-CRD) were randomly divided as training cohort and internal validation cohort. Least absolute shrinkage and selection operator (LASSO) and then Logistic regression was used to construct a predictive model with demographic, clinical, laboratory, comorbidities and ventilation variables ascertained at first 24 h of ICU admission and invasive ventilation. Our model was externally validated using data from another database (MIMIC-IV).

Results

A total of 1075 adult patients from MIMIC-III and eICU were randomly divided into training cohort (70%, n = 752) and internal validation cohort (30%, n = 323). 521 patients were included from MIMIC-IV. From 176 potential predictors, 9 independent predictive factors were included in the final model. Five variables were ascertained within the first 24 h of ICU admission, including age (OR, 1.02; 95% CI: 1.01–1.03), mean of respiratory rate (OR, 1.04; 95% CI: 1.01–1.08), the maximum of INR (OR, 1.14; 95% CI: 1.03–1.31) and alveolo-arterial oxygen difference (OR, 1.002; 95% CI: 1.001–1.003) and the minimum of RDW (OR, 1.17; 95% CI: 1.09–1.27). And four variables were collected within the first 24 h of invasive ventilation: mean of temperature (OR, 0.70; 95% CI: 0.57–0.86), the maximum of lactate (OR, 1.15; 95% CI: 1.09–1.22), the minimum of blood urea nitrogen (OR, 1.02; 95% CI: 1.01–1.03) and white blood cell counts (OR, 1.03; 95% CI: 1.01–1.06). Our model achieved good discrimination (AUC: 0.77, 95% CI: 0.73–0.80) in training cohort but the performance declined in internal (AUC: 0.75, 95% CI: 0.69–0.80) and external validation cohort (0.70, 95% CI: 0.65–0.74) and showed modest calibration.

Conclusions

A risk score based on routinely collected variables at the start of admission to ICU and invasive ventilation can predict mortality of ventilated ARDS patients, with a moderate performance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12890-022-02057-0.

A propensity score-matching analysis of angiotensin-converting enzyme inhibitor and angiotensin receptor blocker exposure on in-hospital mortality in patients with acute respiratory failure

STUDY OBJECTIVE

To explore the impact of pre-hospital ACEI and ARB exposure on the prognosis of ARF patients.

DESIGN

A single-center retrospective cohort study.

SETTING

Medical Information Mart for Intensive Care-III (MIMIC-III) database.

PATIENTS

The patients meeting ICD-9 code of acute respiratory failure were enrolled.

INTERVENTION

The primary exposure was the pre-hospital exposure of ACEI and ARB.

MEASUREMENT AND MAIN RESULTS

The primary outcome was in-hospital mortality. Multiple logistic regression analysis was conducted to determine the independent effect of ACEI/ARB exposure on mortality. Propensity score matching (PSM) method was adopted to reduce bias of the confounders. Subgroup analysis and sensitivity analysis were used to test the stability of the conclusion. 5335 adult ARF patients were enrolled. Mortality was significantly decreased in patients with ACEI/ARB exposure before and after PSM, and the adjusted odds ratio (OR) of ACEI/ARB exposure was 0.56 (95% CI 0.43-0.72). In the subgroup analysis, ACEI/ARB lost its protective effect in young subgroup, but no significant interaction was found between ACEI/ARB exposure and age (p = 0.082). The point estimation and lower 95% limit of E-value was 2.97 and 2.12. In sensitivity analysis, ACEI/ARB exposure showed similar effect in ARDS cohort, but no significantly difference was found in the MIMIC-IV database, which may be explained by small sample size of the ACEI/ARB group.

CONCLUSIONS

Among patients with acute respiratory failure, pre-hospital ACEI/ARB exposure was associated with better outcomes and acted as an independent factor. The relationship between ACEI/ARB and prognosis of ARF is worth investigating further.

Establishment of ICU Mortality Risk Prediction Models with Machine Learning Algorithm Using MIMIC-IV Database

Objective: The mortality rate of critically ill patients in ICUs is relatively high. In order to evaluate patients’ mortality risk, different scoring systems are used to help clinicians assess prognosis in ICUs, such as the Acute Physiology and Chronic Health Evaluation III (APACHE III) and the Logistic Organ Dysfunction Score (LODS). In this research, we aimed to establish and compare multiple machine learning models with physiology subscores of APACHE III—namely, the Acute Physiology Score III (APS III)—and LODS scoring systems in order to obtain better performance for ICU mortality prediction. Methods: A total number of 67,748 patients from the Medical Information Database for Intensive Care (MIMIC-IV) were enrolled, including 7055 deceased patients, and the same number of surviving patients were selected by the random downsampling technique, for a total of 14,110 patients included in the study. The enrolled patients were randomly divided into a training dataset (n = 9877) and a validation dataset (n = 4233). Fivefold cross-validation and grid search procedures were used to find and evaluate the best hyperparameters in different machine learning models. Taking the subscores of LODS and the physiology subscores that are part of the APACHE III scoring systems as input variables, four machine learning methods of XGBoost, logistic regression, support vector machine, and decision tree were used to establish ICU mortality prediction models, with AUCs as metrics. AUCs, specificity, sensitivity, positive predictive value, negative predictive value, and calibration curves were used to find the best model. Results: For the prediction of mortality risk in ICU patients, the AUC of the XGBoost model was 0.918 (95%CI, 0.915–0.922), and the AUCs of logistic regression, SVM, and decision tree were 0.872 (95%CI, 0.867–0.877), 0.872 (95%CI, 0.867–0.877), and 0.852 (95%CI, 0.847–0.857), respectively. The calibration curves of logistic regression and support vector machine performed better than the other two models in the ranges 0–40% and 70%–100%, respectively, while XGBoost performed better in the range of 40–70%. Conclusions: The mortality risk of ICU patients can be better predicted by the characteristics of the Acute Physiology Score III and the Logistic Organ Dysfunction Score with XGBoost in terms of ROC curve, sensitivity, and specificity. The XGBoost model could assist clinicians in judging in-hospital outcome of critically ill patients, especially in patients with a more uncertain survival outcome.