In-hospital mortality and SpO2 incritical care patients with cerebral injury: data from the MIMIC‑IV Database

Association between SpO2 and the risk of death in elderly T2DM patients with cerebral infarction: a retrospective cohort study

Objective

This study aimed to evaluate the SpO2 (transcutaneous oxygen saturation) -mortality link in elderly T2DM (diabetes mellitus type 2) patients with cerebral infarction and identify their optimal SpO2 range.

Methods

In this investigation, we employed a comprehensive approach. Initially, we screened the MIMIC-IV database, identifying elderly T2DM patients with cerebral infarction, utilizing specific ICD-9 and ICD-10 codes. We then harnessed the power of restricted cubic splines to craft a visual representation of the correlation between SpO2 and 1-year mortality. To enhance our analysis, we harnessed Cox multivariate regression, allowing us to compute adjusted hazard ratios (HR) accompanied by 95% confidence intervals (CIs). Additionally, we crafted Cumulative Mortality Curve analyses, augmenting our study by engaging in rigorous subgroup analyses, stratifying our observations based on pertinent covariates.

Results

In this study, 448 elderly T2DM patients with cerebral infarction were included. Within 1-year post-discharge, 161 patients (35.94%) succumbed. Employing Restricted Cubic Spline analysis, a statistically significant U-shaped non-linear relationship between admission ICU SpO2 levels and 1-year mortality was observed (P-value < 0.05). Further analysis indicated that both low and high SpO2 levels increased the mortality risk. Cox multivariate regression analysis, adjusting for potential confounding factors, confirmed the association of low (≤94.5%) and high SpO2 levels (96.5-98.5%) with elevated 1-year mortality risk, particularly notably high SpO2 levels (>98.5%) [HR = 2.06, 95% CI: 1.29-3.29, P-value = 0.002]. The cumulative mortality curves revealed the following SpO2 subgroups from high to low cumulative mortality at the 365th day: normal levels (94.5% < SpO2 ≤ 96.5%), low levels (SpO2 ≤ 94.5%), high levels (96.5% < SpO2 ≤ 98.5%), and notably high levels (>98.5%). Subgroup analysis demonstrated no significant interaction between SpO2 and grouping variables, including Sex, Age, Congestive heart failure, Temperature, and ICU length of stay (LOS-ICU; P-values for interaction were >0.05).

Conclusions

Striking an optimal balance is paramount, as fixating solely on lower SpO2 limits or neglecting high SpO2 levels may contribute to increased mortality rates. To mitigate mortality risk in elderly T2DM patients with cerebral infarction, we recommend maintaining SpO2 levels within the range of 94.5-96.5%.

Development and validation of a nomogram for predicting in-hospital mortality of patients with cervical spine fractures without spinal cord injury

BACKGROUND

The incidence of cervical spine fractures is increasing every day, causing a huge burden on society. This study aimed to develop and verify a nomogram to predict the in-hospital mortality of patients with cervical spine fractures without spinal cord injury. This could help clinicians understand the clinical outcome of such patients at an early stage and make appropriate decisions to improve their prognosis.

METHODS

This study included 394 patients with cervical spine fractures from the Medical Information Mart for Intensive Care III database, and 40 clinical indicators of each patient on the first day of admission to the intensive care unit were collected. The independent risk factors were screened using the Least Absolute Shrinkage and Selection Operator regression analysis method, a multi-factor logistic regression model was established, nomograms were developed, and internal validation was performed. A receiver operating characteristic (ROC) curve was drawn, and the area under the ROC curve (AUC), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were calculated to evaluate the discrimination of the model. Moreover, the consistency between the actual probability and predicted probability was reflected using the calibration curve and Hosmer-Lemeshow (HL) test. A decision curve analysis (DCA) was performed, and the nomogram was compared with the scoring system commonly used in clinical practice to evaluate the clinical net benefit.

RESULTS

The nomogram indicators included the systolic blood pressure, oxygen saturation, respiratory rate, bicarbonate, and simplified acute physiology score (SAPS) II. The results showed that our model had satisfactory predictive ability, with an AUC of 0.907 (95% confidence interval [CI] = 0.853-0.961) and 0.856 (95% CI = 0.746-0.967) in the training set and validation set, respectively. Compared with the SAPS-II system, the NRI values of the training and validation sets of our model were 0.543 (95% CI = 0.147-0.940) and 0.784 (95% CI = 0.282-1.286), respectively. The IDI values of the training and validation sets were 0.064 (95% CI = 0.004-0.123; P = 0.037) and 0.103 (95% CI = 0.002-0.203; P = 0.046), respectively. The calibration plot and HL test results confirmed that our model prediction results showed good agreement with the actual results, where the HL test values of the training and validation sets were P = 0.8 and P = 0.95, respectively. The DCA curve revealed that our model had better clinical net benefit than the SAPS-II system.

CONCLUSION

We explored the in-hospital mortality of patients with cervical spine fractures without spinal cord injury and constructed a nomogram to predict their prognosis. This could help doctors assess the patient's status and implement interventions to improve prognosis accordingly.

Predicting ICU readmission risks in intracerebral hemorrhage patients: Insights from machine learning models using MIMIC databases

BACKGROUND

Intracerebral hemorrhage (ICH) is a stroke subtype characterized by high mortality and complex post-event complications. Research has extensively covered the acute phase of ICH; however, ICU readmission determinants remain less explored. Utilizing the MIMIC-III and MIMIC-IV databases, this investigation develops machine learning (ML) models to anticipate ICU readmissions in ICH patients.

METHODS

Retrospective data from 2242 ICH patients were evaluated using ICD-9 codes. Recursive feature elimination with cross-validation (RFECV) discerned significant predictors of ICU readmissions. Four ML models-AdaBoost, RandomForest, LightGBM, and XGBoost-underwent development and rigorous validation. SHapley Additive exPlanations (SHAP) elucidated the effect of distinct features on model outcomes.

RESULTS

ICU readmission rates were 9.6% for MIMIC-III and 10.6% for MIMIC-IV. The LightGBM model, with an AUC of 0.736 (95% CI: 0.668-0.801), surpassed other models in validation datasets. SHAP analysis revealed hydrocephalus, sex, neutrophils, Glasgow Coma Scale (GCS), specific oxygen saturation (SpO2) levels, and creatinine as significant predictors of readmission.

CONCLUSION

The LightGBM model demonstrates considerable potential in predicting ICU readmissions for ICH patients, highlighting the importance of certain clinical predictors. This research contributes to optimizing patient care and ICU resource management. Further prospective studies are warranted to corroborate and enhance these predictive insights for clinical utilization.

The association between high-oxygen saturation and prognosis for intracerebral hemorrhage

BACKGROUND

Concerns about the adverse effects of excessive oxygen have grown over the years. This study investigated the relationship between high oxygen saturation and short-term prognosis of patients with spontaneous intracerebral hemorrhage (sICH) after liberal use of oxygen.

METHODS

This retrospective cohort study collected data from the Medical Information Mart for Intensive Care III (MIMIC-III) database (ICU cohort) and a tertiary stroke center (general ward cohort). The data on pulse oximetry-derived oxygen saturation (SpO2) during the first 24 h in ICU and general wards were respectively extracted.

RESULTS

Overall, 1117 and 372 patients were included in the ICU and general ward cohort, respectively. Among the patients from the ICU cohort, a spoon-shaped association was observed between minimum SpO2 and the risk of in-hospital mortality (non-linear P<0.0001). In comparison with minimum SpO2 of 93-97%, the minimum SpO2>97% was associated with a significantly higher risk of in-hospital mortality after adjustment for confounders. Sensitivity analysis conducted using propensity score matching did not change this significance. The same spoon-shaped association between minimum SpO2 and the risk of in-hospital mortality was also detected for the general ward cohort. In comparison with the group with 95-97% SpO2, the group with SpO2>97% showed a stronger association with, but non-significant risk for, in-hospital mortality after adjustment for confounders. The time-weighted average SpO2>97% was associated significantly with in-hospital mortality in both cohorts.

CONCLUSION

Higher SpO2 (especially a minimum SpO2>97%) was unrewarding after liberal use of oxygen among patients with sICH and might even be potentially detrimental.

Association between transcutaneous oxygen saturation within 24 h of admission and mortality in critically ill patients with non-traumatic subarachnoid hemorrhage: a retrospective analysis of the MIMIC-IV database

Background

In critically ill patients, transcutaneous oxygen saturation (SpO2) upon admission is typically associated with in-hospital mortality. Nevertheless, the available information for patients with non-traumatic subarachnoid hemorrhage (SAH) is limited. In our study, our objective was to assess the correlation between SpO2 levels and mortality among patients diagnosed with severe SAH.

Methods

In this study, we extracted data from the Medical Information Marketplace in Intensive Care (MIMIC-IV) database, which comprises information on critically ill patients. By employing matching ICD-9 and ICD-10 codes, we identified 3,328 patients diagnosed with SAH. Every individual who was admitted to the intensive care unit (ICU) had their SpO2 data and various covariates, including age, sex, diagnosis, and duration of stay, recorded upon admission. Subsequently, the patients were categorized into three distinct groups according to their SpO2 levels: low (≤95%), moderate (95-98%), and high (≥98%). To investigate the association between percutaneous oxygen saturation and mortality in patients with severe SAH, logistic regression, and cubic spline models were utilized. The main outcomes of interest were 28- and 90-day mortality rates. Additionally, subgroup analyses were conducted to evaluate these correlations and assess the consistency of interactions.

Results

A cohort of 864 patients diagnosed with non-traumatic SAH was included in this study. The correlation between SpO2 and mortality displayed a U-shaped curve when utilizing a finite cubic spline function (non-linearity < 0.001), with the nadir in the probability of in-hospital death at 96%. Mortality at 28 and 90 days showed an inverse correlation with SpO2 < 96% (adjusted odds ratio [OR], 0.8; 95% confidence interval [CI], 0.67-0.95, and 0.76; 95% CI, 0.6-0.96). Conversely, there was a positive correlation between percutaneous oxygen saturation (SpO2) levels of ≥96% and mortality rates at both 28 and 90 days (adjusted OR, 1.17; 95% CI, 1.02-1.35 and 1.2; 95% CI, 1.05-1.39).

Conclusion

In patients with severe subarachnoid hemorrhage, the association between SpO2 and mortality at 28 and 90 days demonstrated a U-shaped pattern. When SpO2 levels were between 95 and 98%, both short- and long-term mortality rates were at their lowest. Patients with significant subarachnoid hemorrhage had a lower chance of survival when their SpO2 values were either high or low.

Oxygen saturation profile in traumatic brain injury animal model after propofol administration

Traumatic brain injury (TBI) is a traumatic that often leads to death due to untreatable cerebral hypoxia, indicated by oxygen saturation of <90%. Cerebral hypoxia is rarely monitored and thereby often overlooked as a cause of mortality and monitoring oxygen saturation is an accurate method to detect the condition. Propofol, an anesthetic agent, is commonly used in the management of TBI; however, its effect on brain tissue and cerebral hypoxia in TBI cases is not well understood. The aim of this study was to evaluate the profile of oxygen saturation in TBI animal model after propofol administration. A laboratory experimental study was conducted, involving 18 male Rattus novergicus rats (aged 4-8 weeks with weight between 150-200 grams) divided into three different treatment groups (non-TBI, TBI without propofol, and TBI with propofol). Oxygen saturation was measured regularly from day 1 to day 8 using pulse oximetry. The oxygen saturation percentages were compared between the TBI rats with and without propofol administration using independent Student t-rest. The results revealed significant reductions of oxygen saturation levels of animals within propofol-treated TBI group compared to that of the untreated-TBI group (p<0.05), with the average oxygen saturation ranging from 80.8%±6.96% vs 86.8%±5.48%. This finding suggests a reducing effect of propofol administration on oxygen saturation levels in rats with TBI and this potentially causes cerebral hypoxia.